Esophageal Cancer - EUR · 2016. 5. 20. · Chapter 10 Recurrence pattern in patients with a...

Esophageal Cancer preoperative management and postoperative outcome

Pieter van Hagen

Esophageal Cancer preoperative managem

ent and postoperative outcome

Pieter van Hagen

Uitnodiging

Voor het bijwonen van de openbare verdediging

van het proefschrift

Esophageal Cancer;preoperative management and

postoperative outcome

door

Pieter van Hagen

Op woensdag 18 mei 2016 om 13.30 uur.

Professor Queridozaal, Onderwijscentrum Erasmus MC.

Wytemaweg 80, Rotterdam

Receptie aansluitend aan de promotie

ParanimfenZarina Lalmahomed

Karel Klop

Pieter van HagenRolderdiephof 13521 DA Utrecht

[email protected]

13631_Hagen_OM.indd 1 15-04-16 16:43

Esophageal Cancer; Preoperative management and postoperative outcome

Pieter van Hagen

ISBN: 978-94-6299-340-2

Printing: Ridderprint BV - www.ridderprint.nl

Lay-out: Nikki Vermeulen - Ridderprint BV

© Pieter van Hagen, 2016

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or

transmitted, in any form or by any means, electronically, mechanically, by photo-copying, recording,

or otherwise, without the prior permission of the author.

Esophageal Cancer; Preoperative management and postoperative outcome

Slokdarmkanker;

Preoperatieve behandeling en postoperatieve uitkomst

Proefschrift

ter verkrijging van de graad van doctor aan de

Erasmus Universiteit Rotterdam

op gezag van de

rector magnificus

prof.dr. H.A.P. Pols

en volgens besluit van het College voor Promoties.

De openbare verdediging zal plaatsvinden op

woensdag 18 mei 2016 om 13.30 uur

door

Pieter van Hagengeboren te Leiden

PROMOTIECOMMISSIE

Promotor: Prof.dr. J.J.B. van Lanschot

Overige leden: Prof.dr. A.H.J. Mathijssen

Prof.dr. E.W. Steyerberg

Prof.dr. R. van Hillegersberg

Copromotor: Dr. B.P.L. Wijnhoven

TABLE OF CONTENTS

Chapter 1 General introduction 7

PART 1: Preoperative analysis and treatment 17

Chapter 2 Preoperative chemoradiotherapy for esophageal or junctional cancer. 19

(2012) New England Journal of Medicine, 366(22), pp. 2074-2084.

Chapter 3 Long-term results of a randomized controlled trial comparing neoadjuvant 35

chemoradiotherapy plus surgery with surgery alone for esophageal or

junctional cancer (CROSS trial).

(2015) Lancet Oncology, Sep;16(9):1090-1098.

Chapter 4 Prolonged time to surgery after neoadjuvant chemoradiotherapy increases 53

histopathological response without improving survival in patients with

esophageal or junctional cancer.

(2014) Annals of Surgery, 260(5):807-14.

Chapter 5 Delay in diagnostic workup and treatment of esophageal cancer. 69

(2010) Journal of Gastrointestinal Surgery, 14(3), pp. 476-483.

Chapter 6 Impact of a multidisciplinary tumor board meeting for upper-GI 85

malignancies on clinical decision making: A prospective cohort study.

(2013) International Journal of Clinical Oncology, 18(2), pp. 214-219.

Chapter 7 Human epidermal growth factor receptor 2 overexpression and amplification 97

in endoscopic biopsies and resection specimens in esophageal and junctional

adenocarcinoma.

(2015) Diseases of the Esophagus, 28(4):380-5.

PART 2: Postoperative outcome and prognostication 109

Chapter 8 Prediction of disease-free survival using relative change in FDG-uptake early during 111

neoadjuvant chemoradiotherapy for potentially curable esophageal cancer.

Accepted for publication, Diseases of the Esophagus

Chapter 9 Patterns of recurrence after surgery alone versus preoperative chemoradiotherapy 127

and surgery in the CROSS trials.

(2014) Journal of Clinical Oncology, 32(5):385-391.

Chapter 10 Recurrence pattern in patients with a pathologically complete response after 143

neoadjuvant chemoradiotherapy and surgery for esophageal cancer.

(2013) British Journal of Surgery, 100(2), pp. 267-273.

Chapter 11 Comparison of the 6th and 7th editions of the UICC-AJCC TNM classification 159

for esophageal cancer.

(2012) Annals of Surgical Oncology, 19(7), pp. 2142-2148.

Chapter 12 Vitamin B12

deficiency following esophagectomy for esophageal 173

and esophagogastric cancer.

Submitted

PART 3: Summary and appendices 185

Summary of the thesis 187

Samenvatting in het Nederlands 193

Future perspectives 197

PhD Portfolio 201

Acknowledgements 203

Curriculum Vitae 207

Chapter 1General introduction

General introduction |

9

Cha

pter

1

Esophageal cancer is an aggressive malignancy. With more than 450,000 newly discovered cases

annually, it is the 8th most common cancer worldwide.1,2 Mortality remains high, with an annual

400,000 deaths worldwide and 5-year overall survival rates for patients with potentially curative

disease rarely exceeding 30%.2-4 In general, two types of esophageal cancer can be distinguished,

i.e. esophageal adenocarcinoma and squamous cell carcinoma. Globally, squamous cell carcinoma

is the most common subtype, accounting for more than 80% of cases. However, in many Western

countries, the incidence of esophageal adenocarcinoma exceeds that of squamous cell carcinoma

and is still rising.1 Because advanced tumors can obstruct the esophageal lumen, most patients with

esophageal cancer present with dysphagia and weight loss. Due to the aggressive character of the

tumor, with invasion into adjacent organs and/or hematogenous dissemination, roughly 50% of

patients are incurable at the time of diagnosis.5,6

Esophageal cancer is most often staged using the tumor-node-metastasis (TNM) classification.7 This

classification is developed by the Union Internationale Contre le Cancer (UICC) and the American

Joint Committee on Cancer (AJCC), and comprises the T-stage that classifies the depth of tumor

infiltration, the N-stage that reflects the presence and number of lymph node metastases, and the

M-stage that describes the presence of distant metastases. The TNM classification is widely used for

defining treatment strategies and prognostication.

Surgery remains the cornerstone of potentially curative treatment for resectable tumors. 8,9 Partly due

to centralization, postoperative mortality is now generally below 5% , however, morbidity remains

high.6,10 Despite extensive preoperative staging, a microscopically radical resection (R0) can only be

performed in 75% of patients treated with primary surgery.11,12 An R0 resection is a strong prognostic

factor and should be pursued whenever possible.13,14 Neoadjuvant chemo(radio)therapy may induce

shrinkage of the tumor, and thereby increase the chance on an R0 resection. Over the past decades,

neoadjuvant treatment with chemo(radio)therapy has been extensively investigated.3,11,15-20 There is

growing evidence that neoadjuvant therapy prolongs long-term survival.21-24

The effect of neoadjuvant chemoradiotherapy can be evaluated in the surgical resection specimen

by quantifying the histopathologic response. This pathological response predicts survival, with a

pathologically complete response having the best long term survival. Partial responders also might

benefit from neoadjuvant therapy, in contrast to non-responders, who only risk exposure to the

toxicity that accompanies the treatment.25-29

Fluorodeoxyglucose Positron Emission Tomography (FDG-PET) is a nuclear imaging technique that

quantifies metabolic activity of tumors, which is increased in most cases of malignant disease.30,31

By administering radioactively labeled glucose, increased glucose uptake in malignant tumors and

possible distant metastases can be detected. This technique has a role in the pretreatment staging

of advanced carcinomas and in detecting metastases in the interval between neoadjuvant treatment

and surgery.32 Also, several studies have been published investigating the potential of FDG-PET to

predict pathological response during or after completion of neoadjuvant chemoradiotherapy.33,34

Unfortunately, it was shown not to be possible to accurately predict histopathological response

early during chemoradiotherapy, which would have facilitated to discontinue the therapy in non-

responders and thus to avoid possible unnecessary toxicity.34

| Chapter 1

10

After completion of neoadjuvant therapy, a certain period of time is needed for the patient to recover

from the therapy before undergoing operative therapy. There is no consensus on how long this

interval between neoadjuvant treatment and surgery should be, but in general, a minimum period

of four weeks is abided. In rectal cancer, it was shown that a longer interval to surgery resulted in a

larger proportion of patients with a pathologically complete response.35-41 In esophageal cancer, there

is only limited evidence for accepting a longer interval.42,43

Besides conventional chemotherapeutic agents, more individualized targeted therapy in esophageal

cancer is currently explored.44 In gastric cancer, targeted palliative therapy with anti-epidermal growth

factor receptor (EGFR) agents was investigated in a randomized trial and proved to lengthen overall

survival in metastatic patients.45 In esophageal cancer, many targets for therapy have been identified,

including Human epidermal growth factor receptor 2 (HER2/neu), a receptor in the EGFR family.46-49

These targets might be used for immunotherapy in a (neo)adjuvant setting, however, so far there is no

clear evidence to justify their use in daily practice for patients with esophageal carcinoma.

Esophagectomy with gastric tube reconstruction changes the anatomy of the upper gastrointestinal

tract, often leading to functional problems and reduced quality of life.50 After a total gastrectomy,

patients are at high risk of developing a vitamin B12

deficiency, as a result of a shortage of intrinsic

factor.51 Intrinsic factor is essential for vitamin B12

metabolism and is produced by the parietal cells

in the gastric mucosa.52 In bariatric patients, vitamin B12

deficiency was also found in patients who

underwent gastric sleeve resections, in which only a part of the gastric mucosa is removed, analogue

to a gastric tube reconstruction.53 So far, it is unknown if there is an increase in vitamin B12

deficiency

after esophagectomy with gastric tube reconstruction.

OUTLINE OF THE THESIS

This thesis is composed of studies that focused on preoperative analysis and treatment (part 1) and on

postoperative outcome and prognostication (part 2). In part 1, neoadjuvant multimodality treatment

(including possible biological targets) of esophageal cancer as well as delay in treatment and other

targets for treatment are discussed. In part 2, the focus is on prognostication after treatment, patterns

of recurrence and long-term consequences of esophageal resection such as vitamin B12

deficiency.

PART 1: PREOPERATIVE ANALYSIS AND TREATMENT

Although surgery remains the cornerstone in the potentially curative treatment of esophageal

cancer, the potential additional value of neoadjuvant treatment has been studied extensively during

the recent decades, with varying results. Multiple meta-analyses were published, however, none

were able to provide definitive support for the use of neoadjuvant chemoradiotherapy for treating

esophageal cancer. Therefore, a multicenter randomized controlled trial comparing neoadjuvant

chemoradiotherapy followed by surgery with surgery alone for patients with potentially curable

esophageal or esophagogastric junction cancer was initiated. In Chapter 2 the long-term results


11

Cha

pter

1

of this trial are described. To further investigate the effects of neoadjuvant chemoradiotherapy on

survival, follow-up was performed five years after completion of the trial. The results of this ultra long

term follow-up are described in Chapter 3.

The time-intervals between preoperative staging, neoadjuvant therapy and surgery have been subject

of debate. The length of these intervals might affect post-operative complications, pathological

staging after surgery, and long term survival. Under analogous circumstances, in rectal cancer, a

positive effect of a prolonged interval between neoadjuvant therapy and surgery was found, resulting

in more pathologically complete responders. In Chapter 4, the effects of a prolonged interval

between neoadjuvant chemoradiotherapy and surgery on pathological outcome and survival are

described. In Chapter 5, we evaluated if the length of the intervals between onset of complaints and

diagnosis (the pre-hospital delay) and between pre-operative staging and surgery (the hospital delay)

were of influence on postoperative morbidity and/or long term survival.

The treatment of esophageal cancer is a multidisciplinary effort, and Dutch guidelines require every

new patient to be presented in a multidisciplinary tumor board prior to treatment. In Chapter 6, the

impact of such a tumor board on clinical decision making is evaluated.

Human epidermal growth factor receptor 2 (HER2/neu) is overexpressed in many solid tumors,

creating a potential target for (neo)adjuvant treatment with immunotherapeutic agents. It has been

shown previously that a subset of esophageal adenocarcinomas also express HER2/neu. However,

it is unclear if overexpression can be reliably established in the preoperatively obtained biopsies,

given the intratumoral heterogeneity. In Chapter 7, the results of a multicenter study are described,

investigating the representativeness of preoperative biopsies for the entire primary tumor in terms of

HER2/neu expression.

PART 2: POSTOPERATIVE OUTCOME AND PROGNOSTICATION 18FluoroDeoxyGlucose Positron Emission Tomography (FDG-PET) proved to successfully distinguish

between responders and non-responders in patients treated with neoadjuvant chemotherapy.

Ancillary to the randomized trial that is described in chapter one, the NEOPEC-trial was initiated.54

In this trial, an FDG-PET was performed immediately prior to nCRT and two weeks after the start of

nCRT. By doing this, a change in metabolic activity of the tumor during therapy can be objectified. For

patients treated with nCRT, FDG-PET was not reliable in predicting pathological response when used

early during therapy. However, the correlation between FDG-PET results and long term survival would

provide a more definitive answer to the usefulness of FDG-PET in prognostication. Therefore, patients

who were included in the NEOPEC- trial were subjected to long term follow-up, which was correlated

with FDG-PET results. This study is described in Chapter 8.The mode of action of chemoradiotherapy is mostly ascribed to its radiotherapy component, while

the chemotherapy is thought to have rather radiosensitizing than systemic effects. To fully understand

the action of both components of neoadjuvant chemoradiotherapy in the treatment of esophageal

cancer, it is essential to scrutinize those cases where multimodality treatment failed. In Chapter 9,

| Chapter 1

12

the pattern of recurrence is described that was encountered in patients treated with neoadjuvant

chemoradiotherapy according to the CROSS-regimen. Furthermore, the recurrence pattern was

established in patients who had a pathologically complete response in the resection specimen after

neoadjuvant chemoradiotherapy. These results are shown in Chapter 10.

For prognostication after surgery in esophageal cancer patients, the TNM classification is the most

widely used. In 2007, the UICC published its 7Th edition of the TNM-classification for solid tumors.

This classification differed strongly from the previous edition, especially for esophageal cancer. To

evaluate and validate the predictive value of the latest edition of the classification, its prognostic

power was compared to that of the sixth edition, using a large external cohort. The results of this

study are described in Chapter 11.

It has been suggested that patients who have undergone esophageal resection with gastric tube

reconstruction are more prone to develop a vitamin B12

deficiency postoperatively, due to the

decreased availability of intrinsic factor. However, the incidence and timing of such deficiency

have never been objectified. Therefore, we commenced a prospective multicenter cohort study

investigating the incidence and prevalence of vitamin B12

deficiency amongst patients who underwent

esophagectomy. In Chapter 12, the results of this study are described.


13

Cha

pter

1REFERENCES1. Arnold M, Soerjomataram I, Ferlay J, Forman D. Global incidence of oesophageal cancer by histological

subtype in 2012. Gut. 2014 Oct 15.

2. Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, et al. Cancer incidence and mortality worldwide: Sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2014 Sep 13.

3. Boonstra JJ, Kok TC, Wijnhoven BP, van Heijl M, van Berge Henegouwen MI, Ten Kate FJ, et al. Chemotherapy followed by surgery versus surgery alone in patients with resectable oesophageal squamous cell carcinoma: long-term results of a randomized controlled trial. BMC Cancer. 2011;11:181.

4. Hulscher JB, van Sandick JW, de Boer AG, Wijnhoven BP, Tijssen JG, Fockens P, et al. Extended transthoracic resection compared with limited transhiatal resection for adenocarcinoma of the esophagus. N Engl J Med. 2002 Nov 21;347(21):1662-9.

5. Koshy M, Esiashvilli N, Landry JC, Thomas CR, Jr., Matthews RH. Multiple management modalities in esophageal cancer: epidemiology, presentation and progression, work-up, and surgical approaches. Oncologist. 2004;9(2):137-46.

6. Enzinger PC, Mayer RJ. Esophageal cancer. N Engl J Med. 2003 Dec 4;349(23):2241-52.

7. Sobin L.H. GMK, Wittekind C. TNM classification of malignant tumors. 7th edition. Oxford: Wiley-Blackwell; 2010.

8. Graham AJ, Shrive FM, Ghali WA, Manns BJ, Grondin SC, Finley RJ, et al. Defining the optimal treatment of locally advanced esophageal cancer: a systematic review and decision analysis. Ann Thorac Surg. 2007 Apr;83(4):1257-64.

9. Low DE. Update on staging and surgical treatment options for esophageal cancer. J Gastrointest Surg. 2011 May;15(5):719-29.

10. van Lanschot JJ, Hulscher JB, Buskens CJ, Tilanus HW, ten Kate FJ, Obertop H. Hospital volume and hospital mortality for esophagectomy. Cancer. 2001 Apr 15;91(8):1574-8.

11. Kelsen DP, Ginsberg R, Pajak TF, Sheahan DG, Gunderson L, Mortimer J, et al. Chemotherapy followed by surgery compared with surgery alone for localized esophageal cancer. N Engl J Med. 1998 Dec 31;339(27):1979-84.

12. van Meerten E, van der Gaast A. Systemic treatment for oesophageal cancer. Eur J Cancer. 2005 Mar;41(5):664-72.

13. Scheepers JJ, van der Peet DL, Veenhof AA, Cuesta MA. Influence of circumferential resection margin on prognosis in distal esophageal and gastroesophageal cancer approached through the transhiatal route. Dis Esophagus. 2009;22(1):42-8.

14. Dexter SP, Sue-Ling H, McMahon MJ, Quirke P, Mapstone N, Martin IG. Circumferential resection margin involvement: an independent predictor of survival following surgery for oesophageal cancer. Gut. 2001 May;48(5):667-70.

15. Bosset JF, Gignoux M, Triboulet JP, Tiret E, Mantion G, Elias D, et al. Chemoradiotherapy followed by surgery compared with surgery alone in squamous-cell cancer of the esophagus. N Engl J Med. 1997 Jul 17;337(3):161-7.

16. Walsh TN, Noonan N, Hollywood D, Kelly A, Keeling N, Hennessy TP. A comparison of multimodal therapy and surgery for esophageal adenocarcinoma. N Engl J Med. 1996 Aug 15;335(7):462-7.

17. Lin CC, Hsu CH, Cheng JC, Wang HP, Lee JM, Yeh KH, et al. Concurrent chemoradiotherapy with twice weekly paclitaxel and cisplatin followed by esophagectomy for locally advanced esophageal cancer. Ann Oncol. 2007 Jan;18(1):93-8.

| Chapter 1

14

18. Reynolds JV, Muldoon C, Hollywood D, Ravi N, Rowley S, O’Byrne K, et al. Long-term outcomes following neoadjuvant chemoradiotherapy for esophageal cancer. Ann Surg. 2007 May;245(5):707-16.

19. Allum WH, Stenning SP, Bancewicz J, Clark PI, Langley RE. Long-term results of a randomized trial of surgery with or without preoperative chemotherapy in esophageal cancer. J Clin Oncol. 2009 Oct 20;27(30):5062-7.

20. van Meerten E, Muller K, Tilanus HW, Siersema PD, Eijkenboom WM, van Dekken H, et al. Neoadjuvant concurrent chemoradiation with weekly paclitaxel and carboplatin for patients with oesophageal cancer: a phase II study. Br J Cancer. 2006 May 22;94(10):1389-94.

21. Fiorica F, Di Bona D, Schepis F, Licata A, Shahied L, Venturi A, et al. Preoperative chemoradiotherapy for oesophageal cancer: a systematic review and meta-analysis. Gut. 2004 Jul;53(7):925-30.

22. Sjoquist KM, Burmeister BH, Smithers BM, Zalcberg JR, Simes RJ, Barbour A, et al. Survival after neoadjuvant chemotherapy or chemoradiotherapy for resectable oesophageal carcinoma: an updated meta-analysis. Lancet Oncol. 2011 Jul;12(7):681-92.

23. Gebski V, Burmeister B, Smithers BM, Foo K, Zalcberg J, Simes J, et al. Survival benefits from neoadjuvant chemoradiotherapy or chemotherapy in oesophageal carcinoma: a meta-analysis. Lancet Oncol. 2007 Mar;8(3):226-34.

24. Urschel JD, Vasan H. A meta-analysis of randomized controlled trials that compared neoadjuvant chemoradiation and surgery to surgery alone for resectable esophageal cancer. Am J Surg. 2003 Jun;185(6):538-43.

25. Mandard AM, Dalibard F, Mandard JC, Marnay J, Henry-Amar M, Petiot JF, et al. Pathologic assessment of tumor regression after preoperative chemoradiotherapy of esophageal carcinoma. Clinicopathologic correlations. Cancer. 1994 Jun 1;73(11):2680-6.

26. Donington JS, Miller DL, Allen MS, Deschamps C, Nichols Iii FC, Pairolero PC. Tumor response to induction chemoradiation: Influence on survival after esophagectomy. Eur J Cardio-Thorac Surg. 2003;24(4):631-7.

27. Van Meerten E, Van Der Gaast A, Tilanus HW, Poley JW, Muller K, Van Dekken H. Pathological analysis after neoadjuvant chemoradiotherapy for esophageal carcinoma: The Rotterdam experience. J Surg Oncol. 2009;100(1):32-7.

28. Schneider PM, Baldus SE, Metzger R, Kocher M, Bongartz R, Bollschweiler E, et al. Histomorphologic tumor regression and lymph node metastases determine prognosis following neoadjuvant radiochemotherapy for esophageal cancer: implications for response classification. Ann Surg. 2005 Nov;242(5):684-92.

29. Swisher SG, Hofstetter W, Wu TT, Correa AM, Ajani JA, Komaki RR, et al. Proposed revision of the esophageal cancer staging system to accommodate pathologic response (pP) following preoperative chemoradiation (CRT). Ann Surg. 2005 May;241(5):810-7; discussion 7-20.

30. Czernin J, Phelps ME. Positron emission tomography scanning: current and future applications. Annu Rev Med. 2002;53:89-112.

31. Gambhir SS, Czernin J, Schwimmer J, Silverman DH, Coleman RE, Phelps ME. A tabulated summary of the FDG PET literature. J Nucl Med. 2001 May;42(5 Suppl):1S-93S.

32. Blom RL, Schreurs WM, Belgers HJ, Oostenbrug LE, Vliegen RF, Sosef MN. The value of post-neoadjuvant therapy PET-CT in the detection of interval metastases in esophageal carcinoma. Eur J Surg Oncol. 2011 Sep;37(9):774-8.

33. Vallbohmer D, Holscher AH, Dietlein M, Bollschweiler E, Baldus SE, Monig SP, et al. [18F]-Fluorodeoxyglucose-positron emission tomography for the assessment of histopathologic response and prognosis after completion of neoadjuvant chemoradiation in esophageal cancer. Ann Surg. 2009 Dec;250(6):888-94.

34. van Heijl M, Omloo JM, van Berge Henegouwen MI, Hoekstra OS, Boellaard R, Bossuyt PM, et al. Fluorodeoxyglucose positron emission tomography for evaluating early response during neoadjuvant chemoradiotherapy in patients with potentially curable esophageal cancer. Ann Surg. 2011 Jan;253(1):56-63.


15

Cha

pter

1

35. Dolinsky CM, Mahmoud NN, Mick R, Sun W, Whittington RW, Solin LJ, et al. Effect of time interval between surgery and preoperative chemoradiotherapy with 5-fluorouracil or 5-fluorouracil and oxaliplatin on outcomes in rectal cancer. J Surg Oncol. 2007 Sep 1;96(3):207-12.

36. Francois Y, Nemoz CJ, Baulieux J, Vignal J, Grandjean JP, Partensky C, et al. Influence of the interval between preoperative radiation therapy and surgery on downstaging and on the rate of sphincter-sparing surgery for rectal cancer: the Lyon R90-01 randomized trial. J Clin Oncol. 1999 Aug;17(8):2396.

37. Tulchinsky H, Shmueli E, Figer A, Klausner JM, Rabau M. An interval >7 weeks between neoadjuvant therapy and surgery improves pathologic complete response and disease-free survival in patients with locally advanced rectal cancer. Ann Surg Oncol. 2008 Oct;15(10):2661-7.

38. Wolthuis AM, Penninckx F, Haustermans K, De Hertogh G, Fieuws S, Van Cutsem E, et al. Impact of interval between neoadjuvant chemoradiotherapy and TME for locally advanced rectal cancer on pathologic response and oncologic outcome. Ann Surg Oncol. 2012 Sep;19(9):2833-41.

39. de Campos-Lobato LF, Geisler DP, da Luz Moreira A, Stocchi L, Dietz D, Kalady MF. Neoadjuvant therapy for rectal cancer: the impact of longer interval between chemoradiation and surgery. J Gastrointest Surg. 2011 Mar;15(3):444-50.

40. Sloothaak DA, Geijsen DE, van Leersum NJ, Punt CJ, Buskens CJ, Bemelman WA, et al. Optimal time interval between neoadjuvant chemoradiotherapy and surgery for rectal cancer. Br J Surg. 2013 Jun;100(7):933-9.

41. Petrelli F, Sgroi G, Sarti E, Barni S. Increasing the Interval Between Neoadjuvant Chemoradiotherapy and Surgery in Rectal Cancer: A Meta-Analysis of Published Studies. Ann Surg. 2013 Nov 20.

42. Ruol A, Rizzetto C, Castoro C, Cagol M, Alfieri R, Zanchettin G, et al. Interval between neoadjuvant chemoradiotherapy and surgery for squamous cell carcinoma of the thoracic esophagus: does delayed surgery have an impact on outcome? Ann Surg. 2010 Nov;252(5):788-96.

43. Kim JY, Correa AM, Vaporciyan AA, Roth JA, Mehran RJ, Walsh GL, et al. Does the timing of esophagectomy after chemoradiation affect outcome? Ann Thorac Surg. 2012 Jan;93(1):207-12; discussion 12-3.

44. Ku GY, Ilson DH. Emerging mAbs for the treatment of esophagogastric cancer. Expert Opin Emerg Drugs. 2014 Dec 6:1-12.

45. Bang YJ, Van Cutsem E, Feyereislova A, Chung HC, Shen L, Sawaki A, et al. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010 Aug 28;376(9742):687-97.

46. Jankowski J, Coghill G, Hopwood D, Wormsley KG. Oncogenes and onco-suppressor gene in adenocarcinoma of the oesophagus. Gut. 1992 Aug;33(8):1033-8.

47. al-Kasspooles M, Moore JH, Orringer MB, Beer DG. Amplification and over-expression of the EGFR and erbB-2 genes in human esophageal adenocarcinomas. Int J Cancer. 1993 May 8;54(2):213-9.

48. Hardwick RH, Shepherd NA, Moorghen M, Newcomb PV, Alderson D. c-erbB-2 overexpression in the dysplasia/carcinoma sequence of Barrett’s oesophagus. J Clin Pathol. 1995 Feb;48(2):129-32.

49. Polkowski W, van Sandick JW, Offerhaus GJ, ten Kate FJ, Mulder J, Obertop H, et al. Prognostic value of Lauren classification and c-erbB-2 oncogene overexpression in adenocarcinoma of the esophagus and gastroesophageal junction. Ann Surg Oncol. 1999 Apr-May;6(3):290-7.

50. Sanghera SS, Nurkin SJ, Demmy TL. Quality of life after an esophagectomy. Surg Clin North Am. 2012 Oct;92(5):1315-35.

51. Pitney WR, Beard MF. Vitamin B12

deficiency following total gastrectomy. AMA Arch Intern Med. 1955 Apr;95(4):591-3.

| Chapter 1

16

52. Kapadia CR. Vitamin B12

in health and disease: part I--inherited disorders of function, absorption, and transport. Gastroenterologist. 1995 Dec;3(4):329-44.

53. Majumder S, Soriano J, Louie Cruz A, Dasanu CA. Vitamin B12 deficiency in patients undergoing bariatric surgery: preventive strategies and key recommendations. Surg Obes Relat Dis. 2013 Nov-Dec;9(6):1013-9.

54. van Heijl M, Omloo JM, van Berge Henegouwen MI, Busch OR, Tilanus HW, Bossuyt PM, et al. NEOadjuvant therapy monitoring with PET and CT in Esophageal Cancer (NEOPEC-trial). BMC Med Phys. 2008;8:3.

PART IPreoperative analysis and treatment

P. van Hagen

M.C.C.M. Hulshof

J.J.B. van Lanschot

E.W. Steyerberg

M.I. van Berge Henegouwen

B.P.L. Wijnhoven

D.J. Richel

G.A.P. Nieuwenhuijzen

G.A.P. Hospers

J.J. Bonenkamp

M.A. Cuesta

R.J.B. Blaisse

O.R.C. Busch

F.J.W. ten Kate

G.-J. Creemers

C.J.A. Punt

J.T.M. Plukker

H.M.W. Verheul

E.J. Spillenaar Bilgen

H. van Dekken

M.J.C. van der Sangen

T. Rozema

K. Biermann

J.C. Beukema

A.H.M. Piet

C.M. van Rij

J.G. Reinders

H.W. Tilanus

A. van der Gaast

for the CROSS Group

The New England Journal of Medicine 2012; 366:2074-2084

Chapter 2Preoperative chemoradiotherapy for

esophageal or junctional cancer

| Chapter 2

20

ABSTRACT

BACKGROUNDThe role of neoadjuvant chemoradiotherapy in the treatment of patients with esophageal or

esophagogastric-junction cancer is not well established. We compared chemoradiotherapy followed

by surgery with surgery alone in this patient population.

METHODSWe randomly assigned patients with resectable tumors to receive surgery alone or weekly

administration of carboplatin (doses titrated to achieve an area under the curve of 2 mg per milliliter

per minute) and paclitaxel (50 mg per square meter of body-surface area) for 5 weeks and concurrent

radiotherapy (41.4 Gy in 23 fractions, 5 days per week), followed by surgery.

RESULTSFrom March 2004 through December 2008, we enrolled 368 patients, 366 of whom were included

in the analysis: 275 (75%) had adenocarcinoma, 84 (23%) had squamous-cell carcinoma, and 7

(2%) had large-cell undifferentiated carcinoma. Of the 366 patients, 178 were randomly assigned

to chemoradiotherapy followed by surgery, and 188 to surgery alone. The most common major

hematologic toxic effects in the chemoradiotherapy–surgery group were leukopenia (6%) and

neutropenia (2%); the most common major nonhematologic toxic effects were anorexia (5%) and

fatigue (3%). Complete resection with no tumor within 1 mm of the resection margins (R0) was

achieved in 92% of patients in the chemoradiotherapy–surgery group versus 69% in the surgery

group (P<0.001). A pathological complete response was achieved in 47 of 161 patients (29%) who

underwent resection after chemoradiotherapy. Postoperative complications were similar in the two

treatment groups, and in-hospital mortality was 4% in both. Median overall survival was 49.4 months

in the chemoradiotherapy–surgery group versus 24.0 months in the surgery group. Overall survival

was significantly better in the chemoradiotherapy–surgery group (hazard ratio, 0.657; 95% confidence

interval, 0.495 to 0.871; P = 0.003).

CONCLUSIONSPreoperative chemoradiotherapy improved survival among patients with potentially curable

esophageal or esophagogastric-junction cancer. The regimen was associated with acceptable

adverse-event rates.

Preoperative chemoradiotherapy for esophageal or junctional cancer |

21

Cha

pter

2

INTRODUCTION

With new diagnoses in more than 480,000 patients annually, esophageal cancer is the eighth most

common cancer worldwide.1 It is a highly lethal disease, causing more than 400,000 deaths per

year.2 The incidence of esophageal adenocarcinoma is rapidly rising, whereas that of squamous-cell

carcinoma remains unchanged.3 Despite adequate preoperative staging, 25% of patients treated

with primary surgery have microscopically positive resection margins (R1), and the 5-year survival

rate rarely exceeds 40%.4 The role of neoadjuvant chemoradiotherapy has been debated for several

decades. In most randomized trials, no survival benefit could be shown, and the trials were criticized

for inadequate trial design, samples that were too small, and poor outcomes in the surgery-alone

group.

Meta-analyses suggest a survival benefit from neoadjuvant chemoradiotherapy, albeit frequently at

the cost of increased postoperative morbidity and mortality.5,6 We previously reported a phase 2 trial

of neoadjuvant chemoradiotherapy consisting of weekly administration of carboplatin and paclitaxel

with concurrent radiotherapy.7 This regimen was associated with a low rate of serious toxic effects,

and a complete resection with no tumor within 1 mm of the resection margins (R0) was achieved in all

patients who underwent resection. These results encouraged us to initiate a multicenter, randomized,

controlled, phase 3 study comparing neoadjuvant chemoradiotherapy followed by surgery with

surgery alone in patients with potentially curable esophageal or esophagogastricjunction carcinoma.8

METHODS

Eligibility criteria

Patients with histologically confirmed, potentially curable squamous-cell carcinoma, adenocarcinoma, or large-cell undifferentiated carcinoma of the esophagus or esophagogastric junction (i.e., tumors

involving both the cardia and the esophagus on endoscopy) were eligible for inclusion in the study.

The upper border of the tumor had to be at least 3 cm below the upper esophageal sphincter. Patients

who had proximal gastric tumors with minimal invasion of the esophagus were excluded. The length

and width of the tumor could not exceed 8 cm and 5 cm, respectively. Only patients with tumors

of clinical stage T1N1 or T2-3N0-1and no clinical evidence of metastatic spread (M0), according to

the International Union against Cancer (UICC) tumor–node–metastasis (TNM) classification,9 were

enrolled. Eligible patients were 18 to 75 years of age, had a World Health Organization (WHO)

performance status score of 2 or lower (on a scale of 0 to 5, with 0 indicating fully active, 1 unable

to carry out heavy physical work, and 2 up and about more than half the day but unable to work),

and had lost 10% or less of body weight. Patients also had to have adequate hematologic, renal,

hepatic, and pulmonary function, as well as no history of other cancer or previous radiotherapy or

chemotherapy.

All patients provided written informed consent. The institutional review board at each participating

center approved the study protocol.8 The protocol, including the statistical analysis plan, is available

with the full text of this article at NEJM.org. No commercial support was involved in the study; the

| Chapter 2

22

drugs were purchased. No one who is not an author contributed to the manuscript. The first, fourth,

and last authors vouch for the accuracy and completeness of the reported data and the fidelity of the

study to the protocol.

Staging

All patients underwent pretreatment staging. This included a history taking; physical examination;

pulmonary-function tests, routine hematologic and biochemical tests; upper gastrointestinal

endoscopy with histologic biopsy and endoscopic ultrasonography; computed tomography of

the neck, chest, and upper abdomen; and external ultrasonography of the neck, with fine-needle

aspiration of lymph nodes when cancer was suspected. For the final analysis, the available endoscopic

reports were centrally reviewed.

Treatment

Chemotherapy

On days 1, 8, 15, 22, and 29, carboplatin targeted at an area under the curve of 2 mg per milliliter per

minute and paclitaxel at a dose of 50 mg per square meter of body-surface area were administered

intravenously. All patients were intravenously premedicated with dexamethasone, clemastine, and

ranitidine as well as standard antiemetic agents. The patients were closely monitored for toxic effects

of chemotherapy with the use of the National Cancer Institute’s Common Terminology Criteria for

Adverse Events, version 3.0.10

Radiotherapy

A total radiation dose of 41.4 Gy was given in 23 fractions of 1.8 Gy each, with 5 fractions administered

per week, starting on the first day of the first chemotherapy cycle. All patients were treated by means

of external-beam radiation. A detailed description of the methods of administration of chemotherapy

and radiotherapy can be found in Appendix 1 in the Supplementary Appendix, available at NEJM.org.

Surgery

Patients in the chemoradiotherapy–surgery group underwent surgery as soon as possible after

completion of chemoradiotherapy (preferably, within 4 to 6 weeks), and patients in the surgery

group were treated as soon as possible after randomization. A transthoracic approach with two-field

lymph-node dissection was performed for tumors extending proximally to the tracheal bifurcation.

For tumors involving the esophagogastric junction, a transhiatal resection was preferred. Peritruncal

dissection was carried out with both approaches. For all other tumors, the approach depended on

the characteristics of the patient and on local preferences. Gastric-tube reconstruction with a cervical

anastomosis was the preferred technique for restoring the continuity of the digestive tract.


23

Cha

pter

2

Pathological analysis

Reports on pathological examination had to describe the tumor type and extension, lymph nodes,

and resection margins. In the absence of macroscopic tumor, any abnormal-appearing tissue was

paraffin-embedded in total in order to make an adequate assessment for the presence of residual

tumor and the effects of therapy.

To grade the response to therapy, we classified the degree of histomorphologic regression into

four categories as follows: grade 1, no evidence of vital residual tumor cells (pathological complete

response); grade 2, less than 10% vital residual tumor cells; grade 3, 10 to 50%; and grade 4, more

than 50%.11,12 If a vital tumor was present at 1 mm or less from the proximal, distal, or circumferential

resection margin, it was considered to be microscopically positive (R1).

Follow-up

During the first year after treatment was completed, patients were seen every 3 months. In the

second year, follow-up took place every 6 months, and then at the end of each year until 5 years

after treatment. Late toxic effects, disease recurrence, and death were documented. Recurrences

were scored at the moment of the first recurrence. During follow-up, diagnostic investigations were

performed only when recurrence was suspected.

Statistical analysis

We calculated that 175 patients were needed in each group in order to detect a difference in median

overall survival of 22 months in the chemoradiotherapy–surgery group versus 16 months in the surgery

group (two-sided test; alpha level, 0.05; beta level, 0.80). Stratification factors included histologic tumor

type, treatment center, lymph-node (N) stage as determined by endoscopic ultrasonography, and

WHO performance score. Block randomization was performed centrally by telephone or at the central

trial office, according to computer-generated randomization lists for each stratum, with random block

sizes of 4 or 6.

Data were analyzed according to the intention-to-treat principle. The primary end point was overall

survival. All other described outcomes were secondary end points. No post hoc analyses were

performed. Survival was calculated from the date of randomization until death. All data collected

through December 2010 were included in the analysis, which guaranteed a potential minimal follow-

up of 2 years.

The Kaplan–Meier method was used to estimate survival, with the log-rank test to determine

significance. A Cox proportional-hazards model was used to estimate the treatment effect with

adjustment for prognostic factors for survival. Moreover, Cox models were used to identify possible

interactions in treatment effect between subgroups, both with and without adjustment for prognostic

factors. Subgroups were predefined according to sex, histologic subtype of tumor, clinical N stage,

and WHO performance score. Statistical analysis was performed with the use of SPSS software, version

17.0 (SPSS).

| Chapter 2

24

837 Assessed and operated

for esophageal / EGJ cancer

469 Not randomized

180 Allocated to CRT+S arm 2 Withdrew their consent

7 Did not receive any CRT

168 Underwent surgery

161 Underwent resecon

188 Allocated to S arm

0 Withdrew their consent

186 Underwent surgery 161 Underwent resecon

178 Analyzed

0 Lost to follow-up 0 Lost to follow-up

188 Analyzed

Enrollment

Allocaon

Analysis

Figure 1. Flow chart

FIGURE 1. Flow chart

RESULTS

Characteristics of the patients

From March 2004 through December 2008, we enrolled 368 patients in the study, of whom 180 were

randomly assigned to the chemoradiotherapy–surgery group, and 188 to the surgery group. Two

patients who were randomly assigned to the chemoradiotherapy–surgery group withdrew consent

and were not included in the analysis (Fig. 1).

Prognostic factors were well balanced between the two treatment groups (Table 1). In both groups,

the median age was 60 years; 134 of 178 patients (75%) in the chemoradiotherapy–surgery group

were men, as compared with 152 of 188 patients (81%) in the surgery group. Most patients (275 of

366 [75%]) had an adenocarcinoma. In both groups, the median tumor length was 4 cm. Most tumors

were located in the distal esophagus (in 211 of 366 patients [58%]) or at the esophagogastric junction

(in 88 [24%]). In the chemoradiotherapy–surgery group, 116 of 178 patients (65%) had positive lymph

nodes as determined by endoscopic ultrasonography, as compared with 120 of 188 (64%) in the

surgery group.


25

Cha

pter

2

TABLE 1. Characteristics of patients with resectable esophageal or esophagogastric-junction cancer, according to treatment group.*

Characteristic Chemoradiotherapy and Surgery (N = 178)

Surgery Alone (N = 188)

Age — yr

Median 60 60

Range 36-79 36-73

Male sex — no. (%) 134 (75) 152 (81)

Tumor type — no. (%)

Adenocarcinoma 134 (75) 141 (75)

Squamous-cell carcinoma 41 (23) 43 (23)

Other 3 (2) 4 (2)

Tumor length — cm†

Median 4 4

Interquartile range 3–6 3–6

Tumor location — no. (%)†

Esophagus

Proximal third 4 (2) 4 (2)

Middle third 25 (14) 24 (13)

Distal third 104 (58) 107 (57)

Esophagogastric junction 39 (22) 49 (26)

Missing data 6 (3) 4 (2)

Clinical T stage — no. (%)‡

cT1 1 (1) 1 (1)

cT2 26 (15) 35 (19)

cT3 150 (84) 147 (78)

cT4 0 1 (1)

Could not be determined§ 1 (1) 4 (2)

Clinical N stage — no. (%)¶

N0 59 (33) 58 (31)

N1 116 (65) 120 (64)

Could not be determined§ 3 (2) 10 (5)

WHO performance status score — no. (%)ǁ

0 144 (81) 163 (87)

1 34 (19) 25 (13)

* Percentages may not add up to 100 because of rounding. WHO denotes World Health Organization.† Tumor length and location were determined by means of endoscopy.‡ Clinical tumor (cT) stage was assessed by means of endoscopic ultrasonography or computed tomography (CT) and was classified according to the International Union against Cancer (UICC) tumor–node–metastasis (TNM) classification.9

§ This category included patients in whom the tumor could not be fully investigated by means of a transducer for endoscopic ultrasonography owing to a stenosis caused by the tumor.¶ Clinical lymph-node (N) stage was assessed by means of endoscopic ultrasonography, CT, or 18F-fluorodeoxyglucose positron-emission tomography and was classified according to UICC TNM classification.9

|| WHO performance status scores are on a scale of 0 to 5, with lower numbers indicating better performance status; 0 indicates fully active, and 1 unable to carry out heavy physical work.

| Chapter 2

26

Delivery and toxic effects of chemoradiotherapy

Seven patients (4%) in the chemoradiotherapy–surgery group did not receive any chemoradiotherapy:

5 because of disease progression before commencing therapy and 2 because they declined

the therapy. A total of 162 patients (91%) received the full treatment regimen of five cycles of

chemoradiotherapy, and 164 (92%) received the full dose of radiotherapy. Two patients (1%) received

a higher dose of radiotherapy (45.0 and 54.0 Gy, respectively). The most common reason for not

completing all chemotherapy cycles was a low platelet count.

In 12 of 171 patients (7%) who received treatment in the chemoradiotherapy–surgery group, grade

3 hematologic toxic effects were observed; a grade 4 hematologic toxic effect and neutropenic fever

developed in 1 patient. One patient died while awaiting surgery after chemoradiotherapy, probably

owing to a perforation of the esophagus, accompanied by major hemorrhage in the absence of

thrombocytopenia. All other major nonhematologic toxic effects of grade 3 or higher occurred in

less than 13% of patients in this group. All serious adverse events that occurred during treatment are

summarized in Table 2.

Surgery

In the chemoradiotherapy–surgery group, 168 patients (94%) underwent surgery, as compared with

186 (99%) in the surgery group (P = 0.01). Reasons for not undergoing surgery were the patient’s

decision (2 patients in the chemoradiotherapy–surgery group), disease progression during treatment

(7 in the chemoradiotherapy–surgery group and 1 in the surgery group), diagnosis of a second

cancer before surgery (1 in the surgery group), and death before surgery due to toxic effects of

chemoradiotherapy (1). No patients were considered medically unfit for surgery.

The median time between randomization and surgery was 97 days in the chemoradiotherapy–surgery

group and 24 days in the surgery group. The median time between the end of chemoradiotherapy

and surgery was 6.6 weeks (interquartile range, 5.7 to 7.9). In 7 of 168 patients (4%) in the

chemoradiotherapy–surgery group, a resection was not possible because the primary tumor or

lymph nodes were identified during surgery as unresectable, as compared with 25 of 186 patients

(13%) in the surgery group (P = 0.002).

Postoperative complications are summarized in Table 2. No significant differences in the occurrence

of complications were found between the two treatment groups. Six of 168 patients (4%) in the

chemoradiotherapy–surgery group died in the hospital, as did 8 of 186 (4%) in the surgery group (P =

0.70). Four patients (2%) in the chemoradiotherapy– surgery group died within 30 days after surgery,

as compared with 5 (3%) in the surgery group (P = 0.85).


27

Cha

pter

2

TABLE 2. Adverse events during neoadjuvant chemoradiotherapy and after surgery.*

Event Chemoradio-therapy and Surgery (N = 171)

Surgery Alone

(N = 186)

Postoperative events — no. of patients/total no. (%)†

Pulmonary complications‡ 78/168 (46) 82/186 (44)

Cardiac complications§ 36/168 (21) 31/186 (17)

Chylothorax¶ 17/168 (10) 11/186 (6)

Mediastinitis‖ǁ 5/168 (3) 12/186 (6)

Anastomotic leakage** 36/161 (22) 48/161 (30)

Death

In hospital 6/168 (4) 8/186 (4)

After 30 days 4/168 (2) 5/186 (3)

Events of any grade during chemoradiotherapy — no. of patients (%)

Anorexia 51 (30

Alopecia 25 (15)

Constipation 47 (27)

Diarrhea 30 (18)

Esophageal perforation 1 (1)

Esophagitis 32 (19)

Fatigue 115 (67)

Nausea 91 (53)

Neurotoxic effects 25 (15)

Vomiting 43 (25)

Leukopenia 103 (60)

Neutropenia 16 (9)

Thrombocytopenia 92 (54)

Events of grade ≥3 during chemoradiotherapy — no. of patients (%)

Anorexia 9 (5)

Constipation 1 (1)

Diarrhea 2 (1)

Esophageal perforation 1 (1)

Esophagitis 2 (1)

Fatigue 5 (3)

Nausea 2 (1)

Vomiting 1 (1)

Leukopenia 11 (6)

Neutropenia 4 (2)

Thrombocytopenia 1 (1)

* Adverse events were graded according to the National Cancer Institute’s Common Terminology Criteria for Adverse Events, version 3.0.10

† Of the 171 patients who received treatment with chemoradiotherapy, 168 underwent surgery.‡ Pulmonary complications were pneumonia (isolation of pathogen from sputum culture and a new or progressive infiltrate on chest radiograph), serious atelectasis (lobar collapse on chest radiograph), pneumothorax (collection of air between the visceral and parietal pleural surfaces, requiring drainage), pleural effusion (collection of fluid between the visceral and parietal pleural surfaces, requiring drainage), pulmonary embolus (embolus detected on spiral CT or a ventilation–perfusion mismatch on a lung scintigram), and acute respiratory failure (partial pressure of arterial oxygen <60 mm Hg while breathing ambient air).

| Chapter 2

28

§ Cardiac complications were arrhythmia (any change in rhythm on the electrocardiogram, requiring treatment), myocardial infarction (two or three of the following: previous myocardial infarction, electrocardiographic changes suggesting myocardial infarction, or enzyme changes suggesting myocardial infarction), and left ventricular failure (marked pulmonary edema on a chest radiograph).¶ Chylothorax was recorded when elevated levels of triglycerides in intrathoracic fluid (>1 mmol per liter [89 mg per deciliter]) were found.|| Mediastinitis was scored when reported by the local investigator.** Anastomotic leakages were recorded when they were diagnosed on physical or radiologic examination in the patients who underwent resection. Leakage was classified as subclinical if it was diagnosed on radiologic examination or endoscopy and as clinical if a salivary fistula was present.

Pathological assessment

An R0 resection was achieved in 148 of 161 patients (92%) in the chemoradiotherapy–surgery group,

as compared with 111 of 161 (69%) in the surgery group (P<0.001). A pathological complete response

(ypT0N0; y denotes underwent neoadjuvant chemoradiotherapy, and p denotes by pathological

assessment) was seen in the resection specimens from 47 patients (29%) in the chemoradiotherapy–

surgery group. A pathological complete response was observed in 28 of 121 patients with

adenocarcinoma (23%) versus 18 of 37 with squamous-cell carcinoma (49%) (P = 0.008). A median

of 15 lymph nodes were resected in patients in the chemoradiotherapy–surgery group, as compared

with 18 in patients in the surgery group (P = 0.77). One or more positive lymph nodes in the resection

specimen were found in 50 patients (31%) in the chemoradiotherapy–surgery group, as compared

with 120 patients (75%) in the surgery group (P<0.001). The pathological findings in all resection

specimens are summarized in Appendix 2 in the Supplementary Appendix.

Survival

For surviving patients, the median follow-up was 45.4 months (range, 25.5 to 80.9). Of the 61 patients

in the chemoradiotherapy–surgery group who underwent resection and died after having been

discharged, 52 (85%) died from recurrent cancer and 9 (15%) from other causes (2 from sepsis, 2

from cardiac failure, 2 from respiratory insufficiency, 1 from kidney failure, 1 from a second primary

tumor, and 1 after reconstructive surgery for a persistent postoperative neo-esophagotracheal

fistula). Of the 83 patients in the surgery group who underwent resection and died after having been

discharged, 78 (94%) died from recurrent cancer, 4 (5%) from other causes (2 from cardiac failure, 1

from respiratory failure, and 1 from a thromboembolic event), and 1 from an unknown cause (P =

0.14). The median disease-free survival for patients who underwent resection was not reached in the

chemoradiotherapy–surgery group and was 24.2 months in the surgery group (hazard ratio, 0.498;

95% confidence interval [CI], 0.357 to 0.693; P<0.001).

An intention-to-treat analysis that included all patients showed a median overall survival of 49.4

months in the chemoradiotherapy–surgery group versus 24.0 months in the surgery group (P = 0.003

by the log-rank test; hazard ratio, 0.657; 95% CI, 0.495 to 0.871) (Fig. 2A). The respective overall survival

rates at 1, 2, 3, and 5 years were 82%, 67%, 58%, and 47% in the chemoradiotherapy–surgery group,

as compared with 70%, 50%, 44%, and 34% in the surgery group. Adjustment for baseline prognostic

factors led to a similar effect estimate (hazard ratio, 0.665; 95% CI, 0.500 to 0.884).


29

Cha

pter

2

Follow up in months

a.

Prop

ortio

nal s

urvi

val

No. at riskCRT+surgery 178 145 119 75 49 28Surgery alone 188 131 94 62 33 17Total 366 276 213 137 82 45

p=0.003

CRT+S

Surgery alone

.

.

.

.

.

.

.

.

.

.

.

Follow up in months

b.

Prop

ortio

nal s

urvi

val

No. at risk AC CRT+S 134 107 87 53 34 18AC S 141 99 73 50 25 10SCC CRT+S 41 35 30 21 15 8SCC S 43 29 19 11 8 4Total 359 270 209 135 82 40

AC p=0.049SCC p=0.011

SCC CRT+S

SCC S

AC CRT+S

AC S

.

.

.

.

.

.

.

.

.

.

.

FIGURE 2. a. Kaplan-Meier plot of the estimated overall 5-year survival of patients with esophageal or esophagogastric junction cancer treated with neoadjuvant chemoradiotherapy (CRT) followed by surgery (solid line, n=178) or with surgery alone (interrupted line, n=188), according to an intention-to-treat analysis. b. Kaplan-Meier plot of the estimated overall 5-year survival of patients with adenocarcinoma treated with neoadjuvant chemoradiotherapy followed by surgery (AC CRT+S, solid grey line, n=134) or with surgery alone (AC S, interrupted grey line, n=141), or squamous cell carcinoma treated with CRT followed by surgery (SCC CRT+S, solid black line, n=41) or with surgery alone (SCC S, interrupted black line, n=43), according to an intention-to-treat analysis. Other tumor types were excluded from this analysis.

Figure 3. Forest plot with Hazard Raos (HR) for death (square) and 95% confidence intervals (CI, error bars) for 366 paents with esophageal or esophagogastric juncon cancer, according to baseline characteriscs. Univariate Hazard raos [95% CI] per subgroup category are shown in the le column. Hazard raos [95% CI] for subgroup effect, adjusted for baseline covariates and p-values are shown in the middle and right column respecvely.

Univariate HR [95% CI] Adjusted HR [95%CI] p-value for adjusted HRAll paents 0.66 [0.495-0.871] 0.67 [0.500-0.884] 0.005All paents by subpopulaon

Sex Female 0.91 [0.482-1.729] 0.93 [0.487-1.766] 0.819 Male 0.61 [0.446-0.841] 0.61 [0.447-0.845] 0.003 Histology Other 0.62 [0.056-6.970] Adenocarcinoma 0.73 [0.524-0.998] 0.74 [0.536-1.024] 0.069 SSC 0.45 [0.243-0.844] 0.42 [0.226-0.788] 0.007 Clinical N-stage cN0 0.41 [0.2340.732] 0.42 [0.239-0.747] 0.003 cN1 0.79 [0.567-1.108] 0.80 [0.576-1.130] 0.212 cNx 0.55 [0.066-4.602] WHO performance score WHO 0 0.62 [0.452-0.844] 0.63 [0.456-0.857] 0.004 WHO 1 0.86 [0.433-1.726] 0.90 [0.753-1.631] 0.765

0.0 1.0 2.0 3.0

Favors CRT+S Favors S

FIGURE 3. Forest plot with Hazard Ratios (HR) for death (square) and 95% confidence intervals (CI, error bars) for 366 patients with esophageal or esophagogastric junction cancer, according to baseline characteristics. Univariate Hazard ratios [95% CI] per subgroup category are shown in the left column. Hazard ratios [95% CI] for subgroup effect, adjusted for baseline covariates and p-values are shown in the middle and right column respectively.

| Chapter 2

30

Separate curves for overall survival according to histologic subtype (i.e., adenocarcinoma or squamous-

cell carcinoma) are shown in Figure 2B. The benefit of neoadjuvant chemoradiotherapy on survival

was consistent across subgroups, without any significant interaction identified. Hazard ratios for the

subgroup effects, with and without adjustment for baseline covariates, are shown in Figure 3.

DISCUSSION

This large, randomized trial of neoadjuvant chemoradiotherapy in patients with esophageal or

esophagogastric-junction cancer showed significantly better overall and disease-free survival among

patients who received a chemoradiotherapy regimen based on carboplatin and paclitaxel, followed by

surgery, as compared with those treated with surgery alone. The chemoradiotherapy was associated

with a low frequency of high-grade toxic effects and could be given as an outpatient treatment.

The preoperative treatment did not result in higher postoperative morbidity or early mortality in this

group, as compared with the surgery group. Patients treated with neoadjuvant chemoradiotherapy

followed by surgery had a 34% lower risk of death during follow-up (hazard ratio, 0.657).

The chemoradiotherapy regimen was designed on the basis of our experience in a previous phase 2

study,7 which used the same dosages of radiotherapy and chemotherapy. In that study, it was possible

to administer this regimen on an outpatient basis, and all the patients had resection margins that

were microscopically negative. On the basis of these results, an alteration of the chemoradiotherapy

regimen was not thought to be necessary.

The present study was designed to detect a difference in median survival of 6 months in favor of the

combined regimen of chemoradiotherapy and surgery, as compared with surgery alone (22 months

vs. 16 months). The observed survival in both groups was superior to the anticipated survival and to

that reported in earlier randomized trials.6,13-17 In line with the results of other studies, the survival of

patients treated with surgery alone has improved,18,19 probably owing to ongoing improvements in

surgical techniques, patient selection, and staging methods over the years. The difference in overall

survival in the present study is not due to poor survival in the surgery group but can clearly be

attributed to improved survival in the chemoradiotherapy–surgery group.

In the chemoradiotherapy–surgery group, 94% of patients underwent surgery, and 90% of tumors

could be resected. In the surgery group, 99% of patients underwent surgery, and 86% underwent

resection. These percentages indicate that the preoperative chemoradiotherapy did not significantly

change the individual chance of undergoing a resection.14,15,17 Postoperative complication rates,

although similar between groups, were higher than expected and higher than reported in other

studies.20 We could not find a plausible explanation for this finding, other than the fact that all

postoperative events were meticulously recorded. This relatively high incidence of postoperative

events in both treatment groups did not result in an increased postoperative mortality, which was

low and similar in the two groups.

Complete remission in both the primary tumor and the lymph nodes (ypT0N0) was the best

possible pathological outcome of chemoradiotherapy. The observed percentage of patients with a


31

Cha

pter

2

pathological complete response (29%) is in line with the reported percentages in other phase 2 and

phase 3 studies.7,13,14,18,21 The substantial downstaging as a result of chemoradiotherapy is also reflected

in the significantly higher percentage of R0 resections in the chemoradiotherapy–surgery group.

Despite the higher rate of pathological complete response among patients with squamous-cell

carcinoma, as compared with those with adenocarcinoma, histologic tumor type was not a prognostic

factor for survival. That is, patients with adenocarcinoma and patients with squamous- cell carcinoma

both benefited from neoadjuvant chemoradiotherapy.

Whether esophageal and esophagogastricjunction tumors should be treated with preoperative

chemoradiotherapy or with perioperative chemotherapy, as suggested by the Medical Research

Council Adjuvant Gastric Infusional Chemotherapy (MAGIC) trial22 and the Actions Concertées dans

les Cancer Colorectaux et Digestifs (ACCORD) 07 trial,23 is unclear. Both trials included gastric tumors as

well as esophagogastric-junction tumors, whereas in the current trial only patients with esophageal or

esophagogastric-junction tumors were treated. In the POET trial, only patients with esophagogastric-

junction tumors were included and randomly assigned to preoperative chemotherapy or

chemoradiotherapy.24 In that study, there was a nonsignificant trend in favor of preoperative

chemoradiotherapy. Because a substantial percentage of patients in the chemoradiotherapy–surgery

group in the present study (22%) had an esophagogastric-junction tumor, we favor preoperative

chemoradiotherapy for such patients, as was also suggested by the POET study, especially because of

the limited toxic effects that were observed with this treatment regimen.

In conclusion, preoperative chemoradiotherapy (five courses of carboplatin and paclitaxel, with 41.4

Gy of concurrent radiotherapy) is safe and leads to a significant increase in overall survival among

patients with adenocarcinoma or squamous-cell carcinoma of the esophagus or esophagogastric

junction.

| Chapter 2

32

REFERENCES1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008:

GLOBOCAN 2008. Int J Cancer 2010;127:2893-917.

2. Parkin DM, Bray FI, Devesa SS. Cancer burden in the year 2000: the global picture. Eur J Cancer 2001;37:Suppl 8:S4-S66.

3. Devesa SS, Blot WJ, Fraumeni JF Jr. Changing patterns in the incidence of esophageal and gastric carcinoma in theUnited States. Cancer 1998;83:2049-53.

4. Kelsen DP, Ginsberg R, Pajak TF, et al. Chemotherapy followed by surgery compared with surgery alone for localized esophageal cancer. N Engl J Med 1998; 339:1979-84.

5. Malthaner R, Fenlon D. Preoperative chemotherapy for resectable thoracic esophageal cancer. Cochrane Database Syst Rev 2001;1:CD001556.

6. Urba SG, Orringer MB, Turrisi A, Iannettoni M, Forastiere A, Strawderman M. Randomized trial of preoperative chemoradiation versus surgery alone in patients with locoregional esophageal carcinoma. J Clin Oncol 2001;19:305-13.

7. van Meerten E, Muller K, Tilanus HW, et al. Neoadjuvant concurrent chemoradiation with weekly paclitaxel and carboplatin for patients with oesophageal cancer: a phase II study. Br J Cancer 2006; 94:1389-94.

8. van Heijl M, van Lanschot JJ, Koppert LB, et al. Neoadjuvant chemoradiation followed by surgery versus surgery alone for patients with adenocarcinoma or squamous cell carcinoma of the esophagus (CROSS). BMC Surg 2008;8:21.

9. International Union against Cancer. TNM classification of malignant tumours. 6th ed. New York: Wiley-Liss, 2002.

10. Trotti A, Colevas AD, Setser A, et al. CTCAE v3.0: development of a comprehensive grading system for the adverse effects of cancer treatment. Semin Radiat Oncol 2003;13:176-81..

11. Mandard AM, Dalibard F, Mandard JC, et al. Pathologic assessment of tumor regression after preoperative chemoradiotherapy of esophageal carcinoma: clinicopathologic correlations. Cancer 1994;73:2680-6.

12. Chirieac LR, Swisher SG, Ajani JA, et al. Posttherapy pathologic stage predicts survival in patients with esophageal carcinoma receiving preoperative chemoradiation. Cancer 2005;103:1347-55.

13. Walsh TN, Noonan N, Hollywood D, Kelly A, Keeling N, Hennessy TP. A comparison of multimodal therapy and surgery for esophageal adenocarcinoma. N Engl J Med 1996;335:462-7. [Erratum, N Engl J Med 1999;341:384.]

14. Bosset JF, Gignoux M, Triboulet JP, et al. Chemoradiotherapy followed by surgery compared with surgery alone in squamous-cell cancer of the esophagus. N Engl J Med 1997;337:161-7.

15. Burmeister BH, Smithers BM, Gebski V, et al. Surgery alone versus chemoradiotherapy followed by surgery for resectable cancer of the oesophagus: a randomized controlled phase III trial. Lancet Oncol 2005;6:659-68.

16. Lin CC, Hsu CH, Cheng JC, et al. Concurrent chemoradiotherapy with twice weekly paclitaxel and cisplatin followed by esophagectomy for locally advanced esophageal cancer. Ann Oncol 2007;18:93-8.

17. Tepper J, Krasna MJ, Niedzwiecki D, et al. Phase III trial of trimodality therapy with cisplatin, fluorouracil, radiotherapy, and surgery compared with surgery alone for esophageal cancer: CALGB 9781. J Clin Oncol 2008;26:1086-92.

18. Lee JL, Park SI, Kim SB, et al. A single institutional phase III trial of preoperative chemotherapy with hyperfractionation radiotherapy plus surgery versus surgery alone for resectable esophageal squamous cell carcinoma. Ann Oncol 2004;15:947-54.


33

Cha

pter

2

19. Portale G, Hagen JA, Peters JH, et al. Modern 5-year survival of resectable esophageal adenocarcinoma: single institution experience with 263 patients. J Am Coll Surg 2006;202:588-96.

20. Fiorica F, Di Bona D, Schepis F, et al. Preoperative chemoradiotherapy for oesophageal cancer: a systematic review andmeta-analysis. Gut 2004;53:925-30.

21. Gannett DE, Wolf RF, Takahashi GW, et al. Neoadjuvant chemoradiotherapy for esophageal cancer using weekly paclitaxel and carboplatin plus infusional 5-fluorouracil. Gastrointest Cancer Res 2007;1:132-8.

22. Cunningham D, Allum WH, Stenning SP, et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N Engl J Med 2006; 355:11-20.

23. Ychou M, Boige V, Pignon JP, et al. Perioperative chemotherapy compared with surgery alone for resectable gastroesophageal adenocarcinoma: an FNCLCC and FFCD multicenter phase III trial. J Clin Oncol 2011;29:1715-21.

24. Stahl M, Walz MK, Stuschke M, et al. Phase III comparison of preoperative chemotherapy compared with chemoradiotherapy in patients with locally advanced adenocarcinoma of the esophagogastric junction. J Clin Oncol 2009;27:851-6.

J. Shapiro

J.J.B. van Lanschot

M.C.C.M. Hulshof

P. van Hagen


B.P.L. Wijnhoven

H.W.M. van Laarhoven

G.A.P. Nieuwenhuijzen

G.A.P. Hospers

J.J. Bonenkamp

M.A. Cuesta

R.J.B. Blaisse

O.R.C. Busch

F.J.W. ten Kate

G.-J. Creemers

C.J.A. Punt

J.T.M. Plukker

H.M.W. Verheul

E.J. Spillenaar Bilgen

H. van Dekken

M.J.C. van der Sangen

T. Rozema

K. Biermann

J.C. Beukema

A.H.M. Piet

C.M. van Rij

J.G. Reinders

H.W. Tilanus

E.W. Steyerberg

A. van der Gaast

For the CROSS study group

Lancet Oncology 2015 Sep;16(9):1090-1098

Chapter 3Neoadjuvant chemoradiotherapy plus surgery versus

surgery alone for oesophageal or junctional cancer (CROSS): long-term results of a randomised controlled trial

| Chapter 3

36

ABSTRACT

BACKGROUND

Initial results of the ChemoRadiotherapy for Oesophageal cancer followed by Surgery Study (CROSS)

comparing neoadjuvant chemoradiotherapy plus surgery versus surgery alone in patients with

squamous cell carcinoma and adenocarcinoma of the oesophagus or oesophagogastric junction

showed a significant increase in 5-year overall survival in favour of the neoadjuvant chemoradiotherapy

plus surgery group after a median of 45 months’ follow-up. In this Article, we report the long-term

results after a minimum follow-up of 5 years.

METHODS

Patients with clinically resectable, locally advanced cancer of the oesophagus or oesophagogastric

junction (clinical stage T1N1M0 or T2–3N0–1M0, according to the TNM cancer staging system, sixth

edition) were randomly assigned in a 1:1 ratio with permuted blocks of four or six to receive either

weekly administration of five cycles of neoadjuvant chemoradiotherapy (intravenous carboplatin

[AUC 2 mg/mL per min] and intravenous paclitaxel [50 mg/m2 of body-surface area] for 23 days) with

concurrent radiotherapy (41·4 Gy, given in 23 fractions of 1·8 Gy on 5 days per week) followed by

surgery, or surgery alone. The primary endpoint was overall survival, analysed by intention-to-treat.

No adverse event data were collected beyond those noted in the initial report of the trial. This trial is

registered with the Netherlands Trial Register, number NTR487, and has been completed.

FINDINGS

Between March 30, 2004, and Dec 2, 2008, 368 patients from eight participating centres (five

academic centres and three large non-academic teaching hospitals) in the Netherlands were enrolled

into this study and randomly assigned to the two treatment groups: 180 to surgery plus neoadjuvant

chemoradiotherapy and 188 to surgery alone. Two patients in the neoadjuvant chemoradiotherapy

group withdrew consent, so a total of 366 patients were analysed (178 in the neoadjuvant

chemoradiotherapy plus surgery group and 188 in the surgery alone group). Of 171 patients who

received any neoadjuvant chemoradiotherapy in this group, 162 (95%) were able to complete the

entire neoadjuvant chemoradiotherapy regimen. After a median follow-up for surviving patients

of 84·1 months (range 61·1–116·8, IQR 70·7–96·6), median overall survival was 48·6 months (95% CI

32·1–65·1) in the neoadjuvant chemoradiotherapy plus surgery group and 24·0 months (14·2–33·7)

in the surgery alone group (HR 0·68 [95% CI 0·53–0·88]; log-rank p=0·003). Median overall survival for

patients with squamous cell carcinomas was 81·6 months (95% CI 47·2–116·0) in the neoadjuvant

chemoradiotherapy plus surgery group and 21·1 months (15·4–26·7) in the surgery alone group (HR

0·48 [95% CI 0·28–0·83]; log-rank p=0·008); for patients with adenocarcinomas, it was 43·2 months

(24·9–61·4) in the neoadjuvant chemoradiotherapy plus surgery group and 27·1 months (13·0–41·2) in

the surgery alone group (HR 0·73 [95% CI 0·55–0·98]; log-rank p=0·038).

Neoadjuvant chemoradiotherapy for oesophageal or junctional cancer (CROSS): long-term results |

37

Cha

pter

3

INTERPRETATION

Long-term follow-up confirms the overall survival benefits for neoadjuvant chemoradiotherapy when

added to surgery in patients with resectable oesophageal or oesophagogastric junctional cancer. This

improvement is clinically relevant for both squamous cell carcinoma and adenocarcinoma subtypes.

Therefore, neoadjuvant chemoradiotherapy according to the CROSS trial followed by surgical resection

should be regarded as a standard of care for patients with resectable locally advanced oesophageal or

oesophagogastric junctional cancer.

INTRODUCTION

Oesophageal cancer is an aggressive disease, characterised by a high degree of locoregional and

distant recurrence after primary surgical resection and poor 5-year overall survival that rarely exceeds

40%.1, 2 and 3 Much effort has been put into improving tumour resectability, long-term locoregional

control, and overall survival, through the addition of chemotherapy, radiotherapy, or both, to surgery,

in a neoadjuvant or adjuvant setting.2, 3, 4 and 5 However, many studies have not shown a significant

long-term survival benefit of such approaches.6 and 7

The randomised controlled ChemoRadiotherapy for Oesophageal cancer followed by Surgery Study

(CROSS) trial8 compared neoadjuvant chemoradiotherapy plus surgery versus surgery alone. The

trial enrolled 368 patients between March 30, 2004, and Dec 2, 2008, from eight Dutch participating

centres (five academic centres and three large non-academic teaching hospitals). Initial results were

published in 2012 after a minimum follow-up of 24 months (median follow-up 45 months [range

25·2-80·9, IQR 32·6-60·6]). We recorded an absolute benefit in 5-year overall survival in favour of the

multimodality group. The neoadjuvant chemoradiotherapy regimen was completed by 162 (95%)

of 171 patients who received any neoadjuvant chemoradiotherapy, with a low occurrence of grade

3 or adverse events for this setting (29 [17%] of 171 patients). Furthermore, a microscopically radical

resection (ie, no vital tumour present at <1 mm from the proximal, distal, or circumferential resection

margins) was achieved in 148 (92%) of 161 patients in the multimodality group, compared with 112

(69%) of 162 in the surgery alone group (p<0·001).

In this Article, we investigate the consistency of longer-term results with our previous findings and

analyse secondary endpoints, such as progression-free survival and disease recurrence patterns.

METHODS

Study design and participants

Full details of patients’ eligibility criteria and the procedures of this open-label, multicentre,

randomised controlled trial have been reported previously.8 and 9 In brief, eligible patients were aged

75 years or younger; had adequate haematological, renal, hepatic, and pulmonary function; and a

WHO performance score of 2 or better, without a past or present history of other malignancy. Only

https://vpn-gw.erasmusmc.nl/+CSCO+0h756767633A2F2F6A6A6A2E667076726170727176657270672E70627A++/science/article/pii/S1470204515000406#bib1












| Chapter 3

38

patients with locally advanced (clinical stage T1N1M0 or clinical stage T2–3N0–1M0, according to

the Union for International Cancer Control [UICC] TNM cancer staging, 6th edition10), histologically

proven, and potentially curable squamous cell carcinoma or adenocarcinoma of the oesophagus or

oesophagogastric junction (i.e., tumours involving both the cardia and the oesophagus on endoscopy)

were eligible for inclusion. The main exclusion criteria were past or current history of malignancy other

than the oesophageal malignancy, previous chemotherapy and/or radiotherapy, and weight loss of

more than 10% of the original bodyweight. The institutional review board at each participating centre

approved the study protocol. All patients provided written informed consent.

Randomisation and masking

Patients were randomly assigned 1:1 to each treatment group, and were stratified according to

histological tumour type (adenocarcinoma vs squamous cell carcinoma), treatment centre, clinical

nodal status (cN0 vs cN1), and WHO performance score (WHO-0 vs WHO-1 vs WHO-2). Randomisation

was done centrally at the Clinical Trial Center at Erasmus MC (Rotterdam, the Netherlands), by computer-

generated randomisation lists for each stratum, with random permuted block sizes of four or six.

Procedures

All patients underwent pretreatment staging, including upper gastrointestinal endoscopy with

histological biopsy and endoscopic ultrasonography; CT scan of the neck, chest, and upper abdomen;

and external ultrasonography of the neck, with fine-needle aspiration of suspected lymph nodes on

indication.

For patients assigned to receive neoadjuvant chemoradiotherapy, carboplatin (AUC 2 mg/mL per

min) and paclitaxel (50 mg/m2 of body-surface area) were administered intravenously for five cycles,

starting on days 1, 8, 15, 22, and 29. A total concurrent radiation dose of 41·4 Gy was given in 23 fractions

of 1·8 Gy, on 5 days per week (excluding weekends), starting on the first day of the first chemotherapy

cycle. The total duration of neoadjuvant treatment was 23 days (5 days per week in weeks 1, 2, 3,

and 4, then 3 days in week 5). If on days 8, 15, 22, or 29 the white blood cell count was lower than

1·0 × 109 cells per L or the platelet count was lower than 50 × 109 per L, administration of neoadjuvant

chemoradiotherapy was delayed by 1 week until recovery above these thresholds. Furthermore, in case

of mucositis with oral ulcers or protracted vomiting despite antiemetic premedication, neoadjuvant

chemoradiotherapy was delayed by 1 week. Further chemotherapy was withheld in case of febrile

neutropenia (defined as a neutrophil count <0·5 × 109 cells per L and a body temperature >38·5°C),

persistent creatinine clearance of less than 50% of the pretreatment level, symptomatic cardiac

arrhythmia or atrioventricular block (with the exception of first-degree atrioventricular block), or other

major organ toxicity at grade 3 or worse (with the exception of oesophagitis). During neoadjuvant

chemoradiotherapy, laboratory tests (including complete blood cell counts and serum creatinine

measurement) were done on a weekly basis, whereas radiological assessments were done only on

indication. All patients in the neoadjuvant chemoradiotherapy plus surgery group were included into



39

Cha

pter

3

the intention-to-treat analysis, irrespective of total dose of neoadjuvant chemoradiotherapy received.

Patients in the surgery alone group were operated on as soon as possible, whereas those in the

neoadjuvant chemoradiotherapy plus surgery group were preferably operated on within 4–6

weeks after completion of chemoradiotherapy. For carcinomas at or above the level of the carina, a

transthoracic oesophageal resection with two-field lymph node dissection was done. For carcinomas

located well below the level of the carina, either a transthoracic approach with two-field lymph node

dissection or a transhiatal approach was used, depending on both patient characteristics and local

preferences. For carcinomas involving the oesophagogastric junction, a transhiatal oesophageal

resection was preferred. In both approaches, an upper abdominal lymphadenectomy, including

resection of nodes along the hepatic artery, splenic artery, and left gastric artery, was done.

For TNM classification, tumour grading, and stage grouping, the sixth edition of the UICC TNM

cancer staging was used.10 Proximal, distal, and circumferential resection margins were assessed.

Microscopically radical resection (R0) was defined as a tumour-free resection margin of at least 1 mm.

During the first year after treatment completion, patients were seen every 3 months. In the second

year, follow-up took place every 6 months, and annually thereafter until 5 years after treatment.

Additional interim visits were scheduled if complaints such as renewed dysphagia and unexplained

weight loss or pain arose before the next scheduled visit. Diagnostic investigations were only

undertaken as necessary during follow-up. No data for adverse events were collected beyond the

initial report of this trial.8

Outcomes

The primary endpoint was overall survival, which was calculated from the date of randomisation to

the date of all-cause death or to the last day of follow-up. Secondary endpoints included progression-

free survival and progression-free interval. Progression-free survival was defined as the interval

between randomisation and the earliest occurrence of disease progression resulting in primary (or

peroperative) irresectability of disease, locoregional recurrence (after completion of therapy), distant

dissemination (during or after completion of treatment), or death from any cause. This definition

for progression-free survival was taken from the modified STEEP criteria for neoadjuvant treatment

trials.11 and 12 The last day of follow-up for progression-free survival varied, depending on the most

recent (scheduled) contact with the patient. Progression-free interval was similar to progression-

free survival, with the difference that treatment-related deaths and non-oesophageal cancer-related

deaths were not counted as events. Locoregional progression was defined as either progression of

locoregional disease during treatment (resulting in irresectability) or as locoregional recurrence after

completion of treatment. Locoregional sites included the mediastinum, the supraclavicular region,

and the coeliac trunk region. Distant progression was defined as occurrence of disseminated disease,

either during or after completion of treatment. Distant disease included cervical and (para-aortic)

lymph node dissemination below the level of the pancreas, malignant pleural effusions, peritoneal

carcinomatosis, and further haematogenous (organ) dissemination.





| Chapter 3

40


Data were analysed according to an intention-to-treat principle. To detect a difference of 6 months in

median overall survival (22 months in the neoadjuvant chemoradiotherapy plus surgery group vs 16

months in the surgery alone group, according to a two-sided test with αlevel 0·05 and β level 0·80),

we calculated that we needed to enrol at least 175 patients in each treatment group. The statistical

significance level was set to 0·05.

We used the Kaplan-Meier method to estimate overall and progression-free survival, with the log-rank

test to ascertain significance. We used univariable and multivariable Cox proportional hazards models

to establish the effect of neoadjuvant chemoradiotherapy in subgroups, adjusting for baseline

covariates.8 Univariable Cox regression modelling was used to analyse differences in progression-free

interval between treatment groups, expressed as hazard ratios (HRs). Follow-up time was divided to

study the temporal distribution of disease progression. Three separate analyses were done, including

follow-up until 6 months, 12 months, and 24 months after randomisation. Progression was defined

as locoregional or distant. Patients in whom both types of disease progression occurred had events

scored in both categories. In the scoring of disease progression in one category, disease progression in

the other category and death without progression were censored. For each timepoint, we compared

the number of events between treatment groups, before the cutoff timepoint and afterwards.

Statistical analysis was done by JS and EWS, using SPSS version 21.0. This trial is registered with the

Netherlands Trial Register, number NTR487.

Role of the funding source

The funder of the study had no role in the study design, data collection, data analysis, data

interpretation, or writing of the report. The corresponding author had full access to all the data in the

study and had final responsibility for the decision to submit for publication.

RESULTS

368 patients from eight participating centres (five academic centres and three large non-academic

teaching hospitals) in the Netherlands were enrolled in the study. 180 patients were randomly

assigned to the neoadjuvant chemoradiotherapy plus surgery group (of whom two later withdrew

consent and were not included in the analysis), and 188 were randomly assigned to the surgery alone

group (figure 1).

Baseline characteristics were well balanced between the two treatment groups (table 1). One

patient in the surgery alone group was originally misclassified as not having received a resection.

However, in the present update, we discovered that this patient had undergone a resection abroad.

This misclassification had no effect on current or previous analyses because of the intention-to-

treat principle. Furthermore, a patient in the neoadjuvant chemoradiotherapy plus surgery group


https://vpn-gw.erasmusmc.nl/+CSCO+0h756767633A2F2F6A6A6A2E667076726170727176657270672E70627A++/science/article/pii/S1470204515000406#fig1

https://vpn-gw.erasmusmc.nl/+CSCO+0h756767633A2F2F6A6A6A2E667076726170727176657270672E70627A++/science/article/pii/S1470204515000406#tbl1


41

Cha

pter

3

moved abroad and was therefore lost to follow-up 73 months after randomisation. Of 171 patients

who received any neoadjuvant chemoradiotherapy in this group, 162 (95%) were able to complete

the entire neoadjuvant chemoradiotherapy regimen. 13 (8%) of 171 patients had grade 3 or worse

haematological toxicity and 18 (11%) had grade 3 or worse non-haematological toxicity. The most

common grade 3 or worse toxicities were leucopenia in 11 (6%) of 171 patients, anorexia in nine (5%),

and fatigue in five (3%).8 In the neoadjuvant chemoradiotherapy plus surgery group, 161 (90%) of

178 patients underwent resection, compared with 162 (86%) of 188 in the surgery alone group. The

proportion of patients with transhiatal resections was similar between both treatment groups (72

[45%] of 161 in the neoadjuvant chemoradiotherapy plus surgery group and 72 [44%] of 162 in the

surgery alone group; χ2=0·01, p=0·96).

837 patients assessed and operated for esophageal or esophagogastric junction cancer

469 excluded

368 patients enrolled and randomly assigned

188 assigned to surgery alone 187 underwent surgery* 161 underwent resection

180 assigned to neoadjuvant chemoradiotherapy and surgery 2 withdrew their consent 5 received CRT only 5 underwent surgery only 5 received neither CRT nor surgery 161 underwent resection

178 were included in the analysis

188 were included in the analysis

FIGURE 1. Trial profile. * One patient had disease progression in waiting period to surgery.


| Chapter 3

42

TABLE 1. Baseline characteristics.

Neoadjuvant chemoradio-therapy plus surgery (n=178) Surgery alone (n=188)

Age, years 60 (55–67) 60 (53–66)

Sex

Women 44 (25%) 36 (19%)

Men 134 (75%) 152 (81%)

Tumour histology

Squamous cell carcinoma 41 (23%) 43 (23%)

Adenocarcinoma 134 (75%) 141 (75%)

Could not be established 3 (2%) 4 (2%)

Tumour length, cm 4 (3–6) 4 (3–6)

Tumour location

Proximal third oesophagus 4 (2%) 4 (2%)

Middle third oesophagus 25 (14%) 24 (13%)

Distal third oesophagus 104 (58%) 107 (57%)

Oesophagogastric junction 39 (22%) 49 (26%)

Missing data 6 (3%) 4 (2%)

Clinical tumour (cT) stage

cT1 1 (1%) 1 (1%)

cT2 26 (15%) 35 (19%)

cT3 150 (84%) 147 (78%)

cT4 0 1 (1%)


Clinical nodal (cN) stage

cN0 59 (33%) 58 (31%)

cN1 116 (65%) 120 (64%)


WHO performance score

0 144 (81%) 163 (87%)

1 34 (19%) 25 (13%)

Data are median (IQR) or n (%).

The final day of follow-up was Dec 31, 2013, guaranteeing a minimum potential follow-up of 60

months for all included patients. At the time of this analysis, median follow-up for surviving patients

was 84·1 months (range 61·1-116·8; IQR 70·7-96·6). Of the 366 analysed patients, 126 were still alive at

the final analysis (73 [41%] of 178 patients in the neoadjuvant chemoradiotherapy plus surgery group

and 53 [28%] of 188 in the surgery alone group). These results correspond with 19 and 21 additional


43

Cha

pter

3

deaths, respectively, since the last follow-up of the original publication.8 Median overall survival was

48·6 months (95% CI 32·1-65·1) in the neoadjuvant chemoradiotherapy plus surgery group and 24·0

months (14·2-33·7) in the surgery alone group (HR 0·68 [95% CI 0·53-0·88]; figure 2A). Median overall

survival for patients with squamous cell carcinomas was 81·6 months (95% CI 47·2-116·0) in the

neoadjuvant chemoradiotherapy plus surgery group and 21·1 months (15·4-26·7) in the surgery alone

group (HR 0·48 [95% CI 0·28-0·83]), and median overall survival for patients with adenocarcinomas was

43·2 months (24·9-61·4) in the neoadjuvant chemoradiotherapy plus surgery group and 27·1 months

(13·0-41·2) in the surgery alone group (HR 0·73 [95% CI 0·55-0·98]; figure 2B). Overall survival was 81%

(95% CI 76-86) at 1 year, 67% (60-74) at 2 years, 58% (51-65) at 3 years, and 47% (39-54) at 5 years in the

neoadjuvant chemoradiotherapy plus surgery group, compared with 70% (63-76), 50% (43-57), 44%

(37-51), and 33% (26-40), respectively, in the surgery alone group (HR 0·57 [95% CI 0·37-0·88] at 1 year,

0·59 [0·43-0·82] at 2 years, 0·65 [0·49-0·88] at 3 years, and 0·67 [0·51-0·87] at 5 years). During follow-up,

16 patients died from treatment-related causes (ie, during neoadjuvant chemoradiotherapy or during

postoperative hospital stay), of whom nine were in the neoadjuvant chemoradiotherapy plus surgery

group and seven in the surgery alone group. 23 patients died from non-disease-related causes beyond

the first 90 days postoperatively (13 in the neoadjuvant chemoradiotherapy plus surgery group and

ten in the surgery alone group).

The estimated number of patients who need to be treated to prevent one additional death at 5

years was 7·1 (95% CI 4·6-13·2).13 The overall survival benefit of neoadjuvant chemoradiotherapy plus

surgery was generally confirmed across subgroups (table 2). The concordance of the multivariable

model for overall survival in all patients was 0·584.15 The proportionality of hazards assumption for the

main analysis was not violated (χ2=0·77, p=0·38).16

Of the 366 analysed patients, 116 were alive and disease free (eventually without evidence of

recurrent disease) at final analysis: 69 (39%) of 178 patients in the neoadjuvant chemoradiotherapy

plus surgery group and 47 (25%) of 188 patients in the surgery alone group. Median progression-

free survival was 37·7 months (95% CI 23·7-51·8) in the neoadjuvant chemoradiotherapy plus surgery

group and 16·2 months (10·7-21·7) in the surgery alone group (HR 0·64 [95% CI 0·49-0·82]; figure

3A). Median progression-free survival for patients with squamous cell carcinomas was 74·7 months

(95% CI 55·1-94·4) in the neoadjuvant chemoradiotherapy plus surgery group and 11·6 months (4·4-

18·8) in the surgery alone group (HR 0·48 [95% CI 0·28-0·82]; figure 3B). Median progression-free

survival for patients with adenocarcinomas was 29·9 months (95% CI 15·9-43·9) in the neoadjuvant

chemoradiotherapy plus surgery group and 17·7 months (11·9-23·5) in the surgery alone group (HR

0·69 [95% CI 0·52-0·92]; figure 3B). Progression-free survival in the neoadjuvant chemoradiotherapy

plus surgery group was 71% (95% CI 65-78) at 1 year, 60% (52-67) at 2 years, 51% (43-58) at 3 years,

and 44% (37-52) at 5 years, compared with 54% (47-61), 41% (34-48), 35% (28-42), and 27% (21-33),

respectively, in the surgery alone group (HR 0·55 [95% CI 0·39-0·77] at 1 year, 0·57 [0·42-0·77] at 2 years,

0·62 [0·47-0·82] at 3 years, and 0·61 [0·47-0·78] at 5 years).












| Chapter 3

44

TABL

E 2.

Uni

varia

ble

and

mul

tivar

iabl

e H

Rs fo

r all-

caus

e m

orta

lity,

acc

ordi

ng to

sub

grou

p ch

arac

teris

tics.

Uni

vari

able

ana

lysi

sM

ultiv

aria

ble

anal

ysis

Neo

adju

vant

ch

emor

adio

ther

apy

plus

sur

gery

(n=1

78)

Surg

ery

alon

e (n

=188

)In

tera

ctio

n p

valu

eH

R (9

5% C

I)p

valu

eaH

R (9

5% C

I)p

valu

e

All

patie

nts

105

(59%

)13

5 (7

2%)

0·07

80·

68 (0

·53-

0·88

)0·

003

0·69

(0·5

3-0·

89)

0·00

4

Sex

0·45

1

Wom

en25

(14%

)24

(13%

)0·

83 (0

·47-

1·45

)0·

502

0·85

(0·4

8-1·

50)

0·57

0

Men

80 (4

5%)

111

(59%

)0·

65 (0

·49-

0·86

)0·

003

0·66

(0·4

9-0·

88)

0·00

4

Tum

our h

isto

logy

0·20

7

Squa

mou

s ce

ll ca

rcin

oma

21 (1

2%)

32 (1

7%)

0·48

(0·2

8-0·

83)

0·00

90·

46 (0

·26-

0·79

)0·

005

Ade

noca

rcin

oma

81 (4

6%)

101

(54%

)0·

73 (0

·55-

0·98

)0·

037

0·75

(0·5

6-1·

01)

0·05

9

Clin

ical

nod

al (c

N) s

tage

0·17

0

cN0

27 (1

5%)

42 (2

2%)

0·50

(0·3

1-0·

80)

0·00

40·

49 (0

·30-

0·80

)0·

004

cN1

77 (4

3%)

85 (4

5%)

0·81

(0·5

9-1·

10)

0·17

60·

83 (0

·61-

1·13

)0·

237

WH

O p

erfo

rman

ce s

core

0·72

9

084

(47%

)11

7 (6

2%)

0·66

(0·5

0-0·

88)

0·00

40·

67 (0

·51-

0·90

)0·

006

121

(12%

)18

(10%

)0·

75 (0

·40-

1·41

)0·

367

0·79

(0·4

1-1·

51)

0·47

3

Dat

a ar

e n

(%),

unle

ss o

ther

wis

e in

dica

ted.

Mul

tivar

iabl

e an

alys

is in

clud

ed th

e fo

llow

ing

base

line

char

acte

ristic

s: se

x, tu

mou

r his

tolo

gy, c

linic

al ly

mph

nod

e (N

) sta

ge, a

nd

WH

O p

erfo

rman

ce s

core

. Clin

ical

N s

tage

was

bas

ed o

n en

dosc

opic

ultr

ason

ogra

phy,

CT,

or 18

F-flu

orod

eoxy

gluc

ose

PET

(in w

hich

cN

0=no

nod

es s

uspe

cted

or p

ositi

ve,

and

cN1=

at le

ast

one

node

sus

pect

ed o

r po

sitiv

e). W

HO

per

form

ance

sta

tus:14

gra

de 0

=ab

le t

o ca

rry

out

all n

orm

al a

ctiv

ity w

ithou

t re

stric

tions

, gra

de 1

=re

stric

ted

in

phys

ical

ly s

tren

uous

act

ivity

but

am

bula

tory

and

abl

e to

do

light

wor

k. H

R=ha

zard

ratio

(nC

RT p

lus

surg

ery

vs s

urge

ry a

lone

). aH

R=ad

just

ed h

azar

d ra

tio.



45

Cha

pter

3

The estimated number of patients who need to be treated to prevent one additional disease

progression at 5 years was 6·1 (95% CI 4·2–10·0).13 The progression-free survival benefit of neoadjuvant

chemoradiotherapy plus surgery was generally confirmed across subgroups (appendix p 3).

TABLE 3. Patients with locoregional or distant progression in the two treatment groups.

Neoadjuvant chemoradiotherapy plus surgery (n=178)

Surgery alone (n=188)

HR (95% CI) p value

Locoregional progression 39 (22%) 72 (38%) 0·45 (0·30–0·66) <0·0001

Distant progression 70 (39%) 90 (48%) 0·63 (0·46–0·87) 0·0040

Overall progression 87 (49%) 124 (66%) 0·58 (0·44–0·76) <0. 0001

Data are n (%), unless otherwise indicated. Comparison between treatment groups was based on univariable cause-specific Cox regression modelling of progression-free intervals. Deaths from non-disease-related causes were censored. Overall progression was defined as either locoregional progression or distant progression. Patients with both locoregional disease progression and distant disease progression (22 patients in the neoadjuvant chemoradiotherapy plus surgery group and 38 in the surgery alone group) were counted in both locoregional progression and distant progression categories. HR=hazard ratio.

We studied the progression-free intervals, in addition to progression-free survival, to focus in more

detail on recurrence patterns in both treatment groups. From randomisation, 211 patients showed

disease progression (table 3). In the neoadjuvant chemoradiotherapy plus surgery group, 87 patients

had disease progression, of whom 39 had locoregional progression and 70 had distant progression

(22 patients had both locoregional and distant progression). In the surgery alone group, 124 patients

had disease progression, of whom 72 had locoregional progression and 90 had distant progression

(38 patients had both). Disease progression during treatment (causing adjustment from curative

to palliative treatment intent) occurred in 17 patients in the neoadjuvant chemoradiotherapy plus

surgery group and in 26 patients in the surgery alone group.

Compared with patients in the surgery alone group, those in the neoadjuvant chemoradiotherapy

plus surgery group had significantly less locoregional progression and significantly less distant

progression (table 3). The reduction in locoregional progression was already apparent during the first

6 months of follow-up and remained significant after the first 24 months of follow-up (appendix p

5). This finding indicates that the effect of reduction in locoregional progression continued for an

extended period after randomisation. The reduction in distant progression was also already recorded

during the first 6 months of follow-up and remained significant during the first 24 months of follow-

up but not thereafter, which suggests that the reduction in distant progression mainly occurred

within the first 24 months after randomisation.


https://vpn-gw.erasmusmc.nl/+CSCO+0h756767633A2F2F6A6A6A2E667076726170727176657270672E70627A++/science/article/pii/S1470204515000406#sec1





| Chapter 3

46

FIG

URE

2. O

vera

ll su

rviv

al.

(A) B

y tr

eatm

ent g

roup

. (B)

By

trea

tmen

t gro

up a

nd h

isto

logi

cal t

umou

r typ

e; fo

ur p

atie

nts i

n th

e su

rger

y al

one

grou

p an

d th

ree

in th

e ne

oadj

uvan

t che

mor

adio

ther

apy

plus

su

rger

y gr

oup

wer

e ex

clud

ed fr

om th

is a

naly

sis

beca

use

thei

r his

tolo

gica

l tum

our t

ype

coul

d no

t be

asce

rtai

ned.

SCC

=sq

uam

ous

cell

carc

inom

a. A

C=

aden

ocar

cino

ma.


47

Cha

pter

3

FIG

URE

3. P

rogr

essi

on-fr

ee s

urvi

val

(A)B

y tr

eatm

ent g

roup

. (B)

By

trea

tmen

t gro

up a

nd h

isto

logi

cal t

umou

r typ

e; fo

ur p

atie

nts

in th

e su

rger

y al

one

grou

p an

d th

ree

in th

e ne

oadj

uvan

t che

mor

adio

ther

apy

plus

sur

gery

gro

up w

ere

excl

uded

from

this

ana

lysi

s be

caus

e hi

stol

ogic

al tu

mou

r typ

e co

uld

not b

e as

cert

aine

d. S

CC=

squa

mou

s ce

ll ca

rcin

oma.

AC

=ad

enoc

arci

nom

a

| Chapter 3

48

DISCUSSION

These long-term results, after a median follow-up for surviving patients of 84 months, confirm the

initially reported survival benefit for neoadjuvant chemoradiotherapy plus surgery compared with

surgery alone. The improvement in distant disease control occurred within the first 2 years after

initiation of treatment, whereas the improvement in locoregional control continued for a longer

period. These findings further support the clinical value of this multimodality treatment strategy.

The overall survival benefit and the progression-free survival benefit were confirmed for both

histological subtypes of oesophageal or oesophagogastric junctional cancer and for other clinically

relevant subgroups. Although univariable and multivariable hazard ratios for individual subgroups

were reported for informative purposes, no significant interactions in treatment effect were identified

for any of the subgroups, which means that differences in treatment effect between subgroups

could well have arisen by chance, and the overall treatment effect should be regarded as valid for all

considered subgroups. In other words, no clear evidence exists to support the assumption that the

adjusted overall treatment effect of neoadjuvant chemoradiotherapy does not also apply to patients

with adenocarcinoma. We therefore conclude that both patients with squamous cell carcinoma and

those with adenocarcinoma benefit significantly from the CROSS multimodality treatment regimen.

The addition of neoadjuvant chemoradiotherapy to primary surgery significantly improved

locoregional disease control. Notably, the largest reported trials with neoadjuvant chemotherapy

only showed limited improvement in rates of R0 resection, pathologically complete response, and

locoregional recurrence. Furthermore, two small randomised trials17 and 18 comparing neoadjuvant

chemotherapy plus surgery to neoadjuvant chemoradiotherapy plus surgery both reported similar R0

resection rates between treatment groups, but significantly higher pathologically complete response

rates and lower locoregional recurrence rates in the neoadjuvant chemoradiotherapy plus surgery

groups. Therefore, the results from these trials point towards neoadjuvant radiotherapy combined

with sensitising chemotherapy rather than neoadjuvant chemotherapy alone as the likely cause of

improved locoregional control, as achieved in the CROSS trial.

In the CROSS trial, not only locoregional control, but also distant disease control improved significantly

in the neoadjuvant chemoradiotherapy plus surgery group. Theoretically, several potential

explanations exist for this improved distant disease control. First, if fewer locoregional recurrences

occur, then possibly less distant dissemination develops from these locoregional recurrences.

Second, effective treatment of the primary tumour in the presence of disseminated disease has been

reported to prolong survival in some cancer types.19 and 20 Therefore, a mechanism by which improved

locoregional control might improve distant disease control could be merely control of the primary

tumour itself, thereby removing a presently unknown stimulus for disseminated tumour outgrowth.

A third explanation is that improved distant disease control could be caused by a direct systemic

effect of chemotherapy. In the present study, we recorded a significant reduction in distant disease

progression already within the first 6 months after randomisation (appendix p 5). Such an early

reduction in distant disease progression, without evidence of a reduction beyond the first 24 months,







49

Cha

pter

3

supports a direct systemic effect of this neoadjuvant chemotherapy regimen. The reduction in distant

disease progression achieved by this neoadjuvant chemoradiotherapy regimen is similar to reductions

achieved with more protracted (and more toxic) perioperative chemotherapy regimens.4 and 5

Results from this trial might not be readily extrapolated to patients with poorer performance status,

older patients, or those with tumours located in the proximal or middle oesophagus, because of the

relative scarcity of patients in these categories. The value of this treatment regimen will need to be

confirmed for these patients in future follow-up studies.

Despite recent advances in curative treatment of oesophageal or junctional cancers, the benefit

of (neo)adjuvant treatment is generally quite limited and a definitive statement on the optimum

perioperative treatment in terms of survival is still absent. A recent meta-analysis suggested a

(non-significant) advantage of neoadjuvant chemoradiotherapy over neoadjuvant chemotherapy

alone in both a direct comparison (HR 0·77 [95% CI 0·53–1·12]) and in an indirect comparison (0·88

[0·76–1·01]).7 The ongoing Japanese randomised NExT trial (JCOG1109)21 and the Irish randomised

Neo-AEGIS trial (ICORG 10-14)22 will hopefully provide more definitive evidence on the best possible

perioperative treatment for squamous cell carcinoma and adenocarcinoma, respectively. Unless

convincing results to the contrary become available, strong evidence from the CROSS trial continues

to support neoadjuvant chemoradiotherapy as a standard of care for both squamous cell carcinoma

and adenocarcinoma of the oesophagus or oesophagogastric junction.

In conclusion, chemoradiotherapy according to the CROSS regimen improves long-term overall and

progression-free survival in patients with oesophageal and junctional cancer. This improvement is

statistically significant and clinically relevant for both squamous cell carcinoma and adenocarcinoma

subtypes. Neoadjuvant chemoradiotherapy according to CROSS followed by surgical resection

should be viewed as a standard of care for patients with resectable locally advanced oesophageal or

junctional cancer.






| Chapter 3

50

REFERENCES1. JBF Hulscher, JW van Sandick, AGEM de Boer, et al. Extended transthoracic resection compared with limited

transhiatal resection for adenocarcinoma of the esophagus. N Engl J Med, 347 (2002), pp. 1662–1669

2. DP Kelsen, KA Winter, LL Gunderson, et al. Long-term results of RTOG trial 8911 (USA Intergroup 113): a random assignment trial comparison of chemotherapy followed by surgery compared with surgery alone for esophageal cancer. J Clin Oncol, 25 (2007), pp. 3719–3725

3. WH Allum, SP Stenning, J Bancewicz, PI Clark, RE Langley. Long-term results of a randomized trial of surgery with or without preoperative chemotherapy in esophageal cancer. J Clin Oncol, 27 (2009), pp. 5062–5067

4. D Cunningham, WH Allum, SP Stenning, et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer N Engl J Med, 355 (2006), pp. 11–20

5. M Ychou, V Boige, JP Pignon, et al. Perioperative chemotherapy compared with surgery alone for resectable gastroesophageal adenocarcinoma: an FNCLCC and FFCD multicenter phase III trial. J Clin Oncol, 29 (2011), pp. 1715–1721

6. V Gebski, B Burmeister, BM Smithers, et al. Survival benefits from neoadjuvant chemoradiotherapy or chemotherapy in oesophageal carcinoma: a meta-analysis. Lancet Oncol, 8 (2007), pp. 226–234

7. KM Sjoquist, BH Burmeister, BM Smithers, et al. Survival after neoadjuvant chemotherapy or chemoradiotherapy for resectable oesophageal carcinoma: an updated meta-analysis. Lancet Oncol, 12 (2011), pp. 681–692

8. P van Hagen, MC Hulshof, JJ van Lanschot, et al. Preoperative chemoradiotherapy for esophageal or junctional cancer. N Engl J Med, 366 (2012), pp. 2074–2084

9. M van Heijl, JJB van Lanschot, LB Koppert, et al.Neoadjuvant chemoradiation followed by surgery versus surgery alone for patients with adenocarcinoma or squamous cell carcinoma of the esophagus (CROSS). BMC Surg, 8 (2008), p. 21

10. LH Sobin, C Wittekind, et al. TNM classification of malignant tumors (6th edn), Wiley-Liss, New York (2002)

11. CA Hudis, WE Barlow, JP Costantino, et al. Proposal for standardized definitions for efficacy end points in adjuvant breast cancer trials: the STEEP system. J Clin Oncol, 25 (2007), pp. 2127–2132

12. D Fumagalli, PL Bedard, Z Nahleh, et al. A common language in neoadjuvant breast cancer clinical trials: proposals for standard definitions and endpoints. Lancet Oncol, 13 (2012), pp. e240–e248

13. DG Altman, PK Andersen. Calculating the number needed to treat for trials where the outcome is time to an event. BMJ, 319 (1999), pp. 1492–1495

14. MM Oken, RH Creech, DC Tormey, et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol, 5 (1982), pp. 649–655

15. FE Harrell Jr, RM Califf, DB Pryor, KL Lee, RA Rosati. Evaluating the yield of medical tests. JAMA, 247 (1982), pp. 2543–2546

16. PM Grambsch, TM Therneau. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika, 81 (1994), pp. 515–526

17. M Stahl, MK Walz, M Stuschke, et al. Phase III comparison of preoperative chemotherapy compared with chemoradiotherapy in patients with locally advanced adenocarcinoma of the esophagogastric junction. J Clin Oncol, 27 (2009), pp. 851–856

18. BH Burmeister, JM Thomas, EA Burmeister, et al. Is concurrent radiation therapy required in patients receiving preoperative chemotherapy for adenocarcinoma of the oesophagus? A randomised phase II trial. Eur J Cancer, 47 (2011), pp. 354–360


51

Cha

pter

3

19. RC Flanigan, SE Salmon, BA Blumenstein, et al. Nephrectomy followed by interferon alfa-2b compared with interferon alfa-2b alone for metastatic renal-cell cancer. N Engl J Med, 345 (2001), pp. 1655–1659

20. E Rapiti, HM Verkooijen, G Vlastos, et al. Complete excision of primary breast tumor improves survival of patients with metastatic breast cancer at diagnosis. J Clin Oncol, 24 (2006), pp. 2743–2749

21. K Nakamura, K Kato, H Igaki, et al . Three-arm phase III trial comparing cisplatin plus 5-FU (CF) versus docetaxel, cisplatin plus 5-FU (DCF) versus radiotherapy with CF (CF-RT) as preoperative therapy for locally advanced esophageal cancer (JCOG1109, NExT study). Jpn J Clin Oncol, 43 (2013), pp. 752–755

22. N Keegan, F Keane, S Cuffe, M Cunningham, N Ravi, G Lee. ICORG 10-14: Neo-AEGIS: a randomized clinical trial of neoadjuvant and adjuvant chemotherapy (modified MAGIC regimen) versus neoadjuvant chemoradiation (CROSS protocol) in adenocarcinoma of the esophagus and esophagogastric junction. Proc Am Soc Clin Oncol, 32 (5 suppl) (2014) abstr TPS4145.

P. van Hagen

J. Shapiro

H.F. Lingsma

B.P.L. Wijnhoven

K. Biermann

F.J.W. ten Kate

E.W. Steyerberg

A. van der Gaast

J.J.B. van Lanschot

on behalf of the CROSS Study Group

Annals of Surgery 2014 Nov;260(5):807-13

Chapter 4Prolonged time to surgery after neoadjuvant

chemoradiotherapy increases histopathological response without affecting survival in patients

with esophageal or junctional cancer

| Chapter 4

54

ABSTRACT

OBJECTIVE

To determine the relation between time to surgery (TTS) after neoadjuvant chemoradiotherapy

(nCRT) and pathologically complete response (pCR), surgical outcome, and survival in patients with

esophageal cancer.

BACKGROUND

Standard treatment for potentially curable esophageal cancer is nCRT plus surgery after 4 to 6 weeks.

In rectal cancer patients, evidence suggests that prolonged TTS is associated with a higher pCR rate

and possibly with better survival.

METHODS

We identified patients treated with nCRT plus surgery for esophageal cancer between 2001 and 2011.

TTS (last day of radiotherapy to day of surgery) varied mainly for logistical reasons. Minimal follow-

up was 24 months. The effect of TTS on pCR rate, postoperative complications, and survival was

determined with (ordinal) logistic, linear, and Cox regression, respectively.

RESULTS

In total, 325 patients were included. Median TTS was 48 days (p25-p75 = 40-60). After 45 days, TTS was

associated with an increased probability of pCR [odds ratio (OR) = 1.35 per additional week of TTS,

P = 0.0004] and a small increased risk of postoperative complications (OR = 1.20, P = 0.976 and

P = 0.139, respectively).

CONCLUSIONS

Prolonged TTS after nCRT increases the probability of pCR and is associated with a slightly increased

probability of postoperative complications, without affecting disease-free and overall survivals. We

conclude that TTS can be safely prolonged from the usual 4 to 6 weeks up to at least 12 weeks,

which facilitates a more conservative wait-and-see strategy after neoadjuvant chemoradiotherapy to

be tested.

Prolonged time to surgery after neoadjuvant chemoradiotherapy for esophageal or junctional cancer |

55

Cha

pter

4

INTRODUCTION

Recent studies show that, in patients with potentially curable esophageal or esophagogastric junction

cancer, the addition of neoadjuvant chemoradiotherapy (nCRT) to surgery improves locoregional

control and long-term survival.1,2 Therefore, nCRT plus surgery has become the standard of care for

patients with potentially curable esophageal or junctional cancer in many countries.

Reasons for prolonged time to surgery (TTS) after nCRT include patients’ malnutrition or poor physical

status, comorbidities such as infections and cardiopulmonary problems, and logistical problems such

as hospital bed and operating room availability.

Tumor response to neoadjuvant therapy is classified in several tumor regression grading (TRG)

systems.3-6 A pathologically complete response (pCR; ie, no viable tumor cells found in the resection

specimen) has been recognized as a valuable prognostic factor for long-term survival.3,7-10 It remains

unclear, however, whether prolonged TTS has an impact on pCR rate, on short-term surgical outcome

and on long-term survival.

Theoretically, prolonged TTS might increase pCR rate and possibly improve disease-free survival

(DFS) because of a prolonged effect of nCRT. Conversely, prolonged TTS might lead to residual tumor

outgrowth, increased difficulty of surgical resection with a higher postoperative complication rate,

and possibly a worse overall survival (OS).

In rectal cancer patients, several studies have shown that prolonged TTS probably increases the

percentage of pathologically complete responders.11-18 Perioperative morbidity and mortality, DFS

and OS, seem less clearly correlated with TTS.12,13,15-22 Comparable studies in esophageal cancer

patients have shown no effect of TTS on pCR rate and no (or little) effect on DFS.23,24

We aimed to investigate the impact of TTS after nCRT on pCR rate, short-term surgical outcome, and

DFS and OS in a cohort of patients with potentially curable esophageal or junctional cancer, who

underwent neoadjuvant chemoradiotherapy according to CROSS 2 followed by surgical resection.

PATIENTS AND METHODS

Patients

We included patients treated with nCRT according to CROSS 2 followed by surgical resection within

a previously published single-center phase II trial (February 2001-January 2004),25 a multicenter

randomized controlled phase III trial (March 2004-November 2008)2 and as standard therapy in the

period after completion of the 2 trials (March 2009-December 2011) at the Erasmus MC-University

Medical Center Rotterdam (a tertiary referral hospital in Rotterdam, The Netherlands).

Patients who did not receive at least 80% of planned nCRT or who received a different nCRT regimen

were excluded. Also, patients with irresectable tumors (cT4b) and/or disseminated disease (M1) were

excluded.

| Chapter 4

56

Treatment and pathological assessment

The neoadjuvant treatment regimen has been described before in detail.2 Esophagectomy was

planned 4 to 6 weeks after completion of nCRT. The resection specimens (primary tumor and all

resected lymph nodes) were processed according to a standardized protocol.26 The TRG was scored

with the modified Mandard scoring system as reported by Chirieac et al.3,4 A microscopically radical

resection (R0) was defined as a tumor-free resection margin of 1 mm or more.

Follow-up and data collection

Clinical, surgical, and histopathological characteristics were retrieved from prospectively maintained

institutional databases. TTS was determined as the interval between the last day of radiotherapy and

the day of surgery. After surgery, patients were routinely followed; every 3 months during the first

postoperative year, every 6 months during the second postoperative year, and yearly until the fifth

postoperative year. During follow-up, diagnostic investigations were performed only when recurrence

was suspected clinically. Recurrences were scored at the time of first failure. Survival was determined

using hospital records and municipal registers.

Surgical outcome

Surgical outcome was described by duration of operation in minutes, intraoperative blood loss in

milliliters, and length of hospital stay in days. Postoperative complications were categorized according

to the Clavien-Dindo classification,27,28 a postoperative complication ranking system: grade 0-no

complications; grade I-complications requiring no or minimal (nonpharmacological) treatment; grade

II-complications requiring pharmacological treatment; grade III-complications requiring surgical,

endoscopic, or radiological intervention; grade-IV-life-threatening complications requiring intensive

care unit-level care; and grade V-postoperative death.

Potential delay-related confounders

To correct for the possibility that variations in TTS were not only caused by random logistical

difficulties but also by patient-related characteristics (i.e. intentionally longer TTS in more vulnerable

patients), effects of TTS on pCR and on surgical outcome were adjusted for 3 potential delay-related

confounders.

First, this was the Charlson comorbidity index 29 at diagnosis, an indicator of comorbidity. For this

index, each condition present at diagnosis is assigned a standard score of 1, 2, 3, or 6 (determined by

the estimated risk of dying from this condition). The total sum per patient is predictive for mortality.

Second, the Karnofsky performance status 30,31 at the end of nCRT, an indicator of general well-

being. The scale ranges from 100 (normal health) to 0 (death). A score of 90 indicates minor signs and/

or symptoms of disease, 80 indicates mild signs and/or symptoms of disease, etc. Third, weight loss

during nCRT, defined as the difference in body weight (in kilograms) between the start of the first cycle

of chemotherapy and the start of the fifth cycle of chemotherapy. Weight gains were set to zero loss.


57

Cha

pter

4


Baseline data were described as medians with the interquartile range in case of continuous variables

and as frequencies with percentages in case of categorical variables. Missing baseline data were

imputed with a single imputation technique based on correlations with relevant baseline variables

and outcome. TTS was truncated at 5% and 95% to reduce leverage effects of outliers. The effect of

TTS on pCR and on surgical outcome was determined using univariable and multivariable logistic

regression. Nonlinearity of the effect of TTS was assessed with restricted cubic splines (3 knots).32 To

correct for the possibility that TTS was intentionally longer in more vulnerable patients, the effects of

TTS on surgical outcome was adjusted in multivariable regression analysis for the 3 potential delay-

related confounders together with surgical approach (adjusted odds ratio, aOR). Subsequently, the

effects of pCR and TTS on DFS and OS were assessed using Cox regression with adjustment for the

3 potential delay-related confounders alone and in combination with postoperative complications

(adjusted hazard ratio, aHR).

RESULTS

Patient and treatment-related characteristics

In total, 325 patients with potentially curable esophageal or junctional cancer were included who

underwent nCRT according to CROSS 2 followed by resection between February 2001 and December

2011. Fifty-one included patients participated in the phase II trial,25 157 patients participated in

the multimodality arm of the randomized phase III trial,2 and 117 patients were treated with nCRT

followed by resection as standard therapy in the period after completion of the phase III trial. In total,

344 patients completed nCRT within the 3 cohorts. Of these, 4 patients did not undergo surgery

(primarily due to patient preferences) and 15 additional patients did not undergo resection (due to

disease progression).

Clinical, surgical, and histopathological characteristics of included patients are summarized in

Table 1. Median age at diagnosis was 60 years. The majority of patients were male (78%), had an

adenocarcinoma (76%), and were clinically staged as cT3 (77%), cN1 (63%). Before truncation, TTS

ranged from 18 to 291 days, with a median of 48 days (interquartile range: 40-60 days), as also shown

in Figure 1. Median TTS in irresectable patients was 46 days (interquartile range: 40-61 days), which

was similar to TTS in resected patients (P = 0.99). Surgical resection was performed regularly via both

transhiatal approach (48%) and transthoracic approach (50%). Median duration of operation was 368

minutes, median volume of intraoperative blood loss was 900 mL, and median length of hospital

stay was 14 days. Postoperative complications requiring pharmacological, interventional, or surgical

treatment (>= grade-II) were reported in 55% of patients. Postoperative in-hospital mortality was

4%. Histopathological staging showed frequent ypT0 and ypT3 stages (34% and 32%, respectively),

whereas ypN0 was the most frequent nodal stage (68%). A pCR (TRG1) was found in 28% of patients.

The microscopically radical resection rate (R0) was 93%.

| Chapter 4

58

TABLE 1 . Clinical, surgical, and histopathological characteristics of 325 included patients with potentially curable esophageal or junctional cancer, treated with neoadjuvant chemoradiotherapy according to CROSS2 followed by surgical resection.

Characteristic N (total) N %*

Age [years] 325

median (p25-p75) 60 (55 – 67)

Sex 325

Male 253 78%

Female 72 22%

Tumor type 325

Squamous cell carcinoma 73 23%

Adenocarcinoma 247 76%

Other 5 2%

cT-stage§ 323

cT1 9 3%

cT2 60 19%

cT3 250 77%

cT4 4 1%

cN-stage‡ 324

cN0 120 37%

cN1 204 63%

Time to surgery [days] ₪ 325

median (p25-p75) 48 (40 – 60)

Surgical approach 325

Transhiatal approach 156 48%

Transthoracic approach 164 50%

Thoraco-phreno-laparotomy 5 2%

Length of operation [min] 308

median (p25-p75) 368 (262 – 388)

Intraoperative blood loss [ml] 281

median (p25-p75) 900 (550 – 1350)

Length of hospital stay [days] 319

median (p25-p75) 14 (12 – 23)

Clavien–Dindo classification (26, 27) ∆ 324

Grade 0 72 22%

Grade I 73 23%

Grade II 82 25%

Grade III 43 13%


59

Cha

pter

4

TABLE 1 . Continued


Grade IV 40 12%

Grade V 14 4%

ypT-stage• 324

ypT0 110 34%

ypT1 49 15%

ypT2 59 18%

ypT3 103 32%

ypT4 3 1%

ypN-stage• 325

ypN0 222 68%

ypN1 74 23%

ypN2 21 7%

ypN3 8 3%

Tumor regression grade (TRG) ¥ 325

1 90 28%

2 102 31%

3 77 24%

4 56 17%

Radicality◊ 325

R0303 93%

R122 7%

* Percentages may not add up to 100 due to rounding§ cT-stage: clinical T-stage as defined by endo-ultrasonography and/or CT-scanning according to the Union

Internationale Contre le Cancer (UICC) TNM Cancer Staging, 7th edition. ‡ cN-stage: clinical N-stage as defined by endo-ultrasonography and/or CT-scanning and/or FDG-PET-scanning

according to UICC TNM Cancer Staging, 7th edition.₪ Time to surgery was determined as the interval between the last day of radiotherapy and the day of surgery. ∆ Clavien-Dindo classification, a postoperative complication ranking system. Grade 0: no complications,

grade I: complications requiring none or minimal (non-pharmacological) treatment, grade II: complications requiring pharmacological treatment, grade III: complications requiring surgical, endoscopic or radiological intervention, grade IV: life-threatening complications requiring ICU-level care, grade V: postoperative death.

• ypT and ypN-stage: pathological T-stage and pathological N-stage in the resection specimen following neoadjuvant chemoradiotherapy, according to UICC TNM Cancer Staging, 7th edition.

¥ Tumor regression grade was defined as: TRG1: no residual tumor cells found (pathologically complete response= pCR); TRG2: 1-10% residual tumor cells; TRG3: 11-50% residual tumor cells; TRG4: > 50% residual tumor cells.

◊ R0 was defined as a tumor-free resection margin ≥ 1 mm. R

1 was defined as a macroscopically radical resection,

with a microscopically tumor-free resection margin < 1 mm.

| Chapter 4

60

FIGURE 1 . Distribution of TTS (before truncation) in 340 patients who completed neoadjuvant chemoradiotherapy according to CROSS 2 and underwent surgery for potentially curable esophageal or junctional cancer, 15 of these patients were irresectable due to disease progression, whereas 325 patients underwent resection.

Effect of TTS on pCR

Median TTS was 48 days (p25-p75: 40-60 days, p5-p95: 18-83 days). The association of TTS with pCR

was nonlinear (P = 0.025). After approximately 45 days (Supplementary Fig. 1, available at http://links.

lww.com/SLA/A672), additional TTS was associated with an increased probability of pCR (OR = 1.35

per additional week after 45 days, P = 0.0004).

In itself a pathologically complete response was associated with a significantly better DFS and OS

HR = 0.29, P < 0.0001 and 0.44, P < 0.0001, respectively). The effect of pCR on OS was sustained after

correction for the 3 potential delay-related confounders (aHR = 0.44, P < 0.0001).

Effect of TTS on short-term surgical outcome

The effect of TTS on short-term surgical outcome was reasonably linear (nonlinearity, P = 0.920).

Increased TTS was associated with prolonged duration of operation (12 minutes per additional week

of TTS, P < 0.001), prolonged length of hospital stay (1.55 day per additional week of TTS, P = 0.006),

and more severe postoperative complications (OR 1.20 per additional week of TTS, P < 0.001). No

association was found with intraoperative blood loss (-3 mL per additional week of TTS, P = 0.898).

However, these associations were partly explained by the 3 potential delay-related confounders

together with surgical approach. After adjustment, the effects were reduced for duration of operation

(6 minutes per additional week of TTS, P = 0.031), length of hospital stay (1.03 days per additional

week of TTS, P = 0.071), and for postoperative complications (aOR 1.10 per additional week of TTS, P

= 0.132, Table 3.)


61

Cha

pter

4

TABLE 2. Three potential delay-related confounders in 325 included patients with potentially curable esophageal or junctional cancer, treated with neoadjuvant chemoradiotherapy according to CROSS2 followed by surgical resection.


Charlson comorbidity index28 §at initial diagnosis

324

0

243 75%

1

63 19%

2

15 5%

3

2 1%

4

1 0%

Karnofsky performance status 29,30 ‡ at the end of nCRT 269

100 80 30%

90 171 64%

≤80 18 7%

Weight loss [kg] •

during nCRT297

median (p25-p75) 1 (0 – 3)

* Percentages may not add up to 100 due to rounding§ Charlson comorbidity index, an indicator of comorbidity. Each condition present at diagnosis is assigned a

standard score of 1, 2, 3 or 6 (determined by the estimated risk of dying from this condition). The total sum per patient is predictive for mortality.

‡ Karnofsky performance status, an indicator of general well-being. The scale ranges from 100 (normal health) to 0 (death). A score of 90 indicates minor signs and/or symptoms of disease, 80 indicates mild signs and/or symptoms of disease, etc.

• Weight loss during nCRT was defined as the difference in body weight (in kilograms) between the start of the first cycle of chemotherapy and the start of the fifth cycle of chemotherapy. Weight gains were set to zero loss.

| Chapter 4

62

TABLE 3. Effect of time to surgery on short-term surgical outcome and postoperative complications in 325 included patients treated with CROSS 2 followed by surgical resection, adjusted for the three potential delay-related confounders together with surgical approach.

Outcomebeta(per additional week)

Padjusted beta (per additional week)

P

Length of operation (minutes) 12 <0.001 6 0.031

Intraoperative blood loss (ml) -3 0.898 -3 0.898

Length of hospital stay (days) 1.55 0.006 1.03 0.071

Postoperative complications (according to the Clavien-Dindo

classification 26, 27))1.20* <0.001 1.10* 0.132

*Odds ratioAdjustment for the 3 delay-related confounders (Charlson comorbidity index29 at diagnosis, Karnofsky performance status 30,31 during the last week of nCRT, and weight loss during nCRT) together with surgical approach.

The potential delay-related confounders are summarized in Table 2. The majority of patients had a

pretreatment Charlson comorbidity index of 0 (75%) and a Karnofsky performance score at the end of

nCRT of 90 (64%). Median weight loss during nCRT was 1 kg (p25-p75: 0-3).

Effects of TTS on survival

The effect of TTS on DFS and OS was reasonably linear (nonlinearity, P = 0.566). Increased TTS was not

associated with DFS (HR = 1.00, P = 0.976; aHR = 0.98, P = 0.620), nor with OS (HR = 1.06, P = 0.139;

aHR = 1.03, P = 0.465, Table 4).

TABLE 4. Effect of time to surgery on OS in 325 included patients with potentially curable esophageal or junctional cancer, treated with neoadjuvant chemoradiotherapy according to CROSS followed by surgical resection

TTS HR for OS(Per Additional Week) P

Univariate 1.06 0.139

Adjusted for 3 potential delay-related confounders 1.04 0.387

Adjusted for 3 potential delay-relatedconfounders + postoperativecomplications (shows effect ofTTS on OS via increased pCR)

1.030.465

Delay-related confounders: Charlson comorbidity index29 at diagnosis, Karnofsky performance status30,31 during the last week of nCRT, and weight loss during nCRT.


63

Cha

pter

4

DISCUSSION

Results show that prolonged TTS, beyond 45 days after completion of nCRT, significantly increased

the probability of pCR. Longer TTS was also associated with a small increased risk of postoperative

complications. Furthermore, pCR in the resection specimen was associated with significantly

improved DFS. However, prolonged TTS (with increased probability of pCR) was not associated with

improved DFS.

Although prolonged TTS was associated with an increased pCR rate, DFS was not accordingly

improved. The comparable DFS between patients with shorter and longer TTS implies that their

disease is biologically similar, despite differences in pCR rate. The increase in pCR after prolonged TTS

is attributed to patients who presumably would have been noncomplete responders at shorter TTS.

Interestingly, these patients, who become pathologically complete responders relatively late after

completion of nCRT, do not contribute to an improved DFS after prolonged TTS. This can be explained

if “late complete responders” would already behave as biologically complete responders at shorter

TTS, despite a pathologically noncomplete response in their resection specimens. Therefore, the

(minimal) residual disease found at histological examination after shorter TTS in these late complete

responders is presumed to be biologically nonrelevant. We hypothesize that prolonged TTS does

not truly improve the biological pCR rate, but that prolonged TTS simply allows for a more precise

histopathological distinction between viable residual tumor cells and biologically nonrelevant tumor

cell debris by the pathologist.

In this cohort, prolonged TTS was significantly associated with a prolonged duration of operation,

a prolonged length of hospital stay, and increased postoperative complications. However, these

associations were partly explained by the 3 potential delay-related confounders, that is, Charlson

comorbidity index 29 at diagnosis, Karnofsky performance status 30,31 at the end of nCRT, and weight loss

during nCRT (besides surgical approach; Table 3). These potential confounding factors were chosen to

reflect the overall condition of patients at the end of nCRT, which might have influenced the surgeons’

decision to delay surgical resection after nCRT, independent of logistical planning. Therefore, these

results indicate that the association of prolonged TTS with negative short-term surgical outcome

and increased postoperative complications is partly the result of an intentionally prolonged TTS in

patients with a worse overall condition at the end of nCRT and that prolonged TTS has at most a small

association with short-term surgical outcome and postoperative complications.

Two previous studIes investigated the association between TTS and surgical and oncological outcome

in patients with esophageal cancer. Both studies did not find a correlation with pCR rate.23,24 This

discrepancy with the present study might be related to the heterogeneity of nCRT protocols used

in both previous studies. In our cohort, all patients completed the same nCRT protocol, and in all

patients there was a similar initial intention to undergo surgery after 4 to 6 weeks, possibly allowing

for a clearer delineation of the effect of TTS on pCR rate. Also, the present study investigated the

association between TTS and pCR in all patients using logistic regression, resulting in more power to

detect subtle effects as compared to subgroup analyses. Last, this study focused on the association

| Chapter 4

64

between TTS and DFS, which was lacking in the study by Kim and colleagues.24

Currently, controversy exists concerning the management of cancer patients with a clinically

complete response after neoadjuvant therapy. For rectal cancer patients, more conservative strategies

have already been proposed and tested in clinically complete responders.33-36 The goal of these wait-

and-see approaches is organ preservation and overall reduction in treatment-related morbidity and

mortality. However, it has remained difficult to clinically distinguish truly complete responders from

nearly complete responders. Patients belonging to the latter group might be incorrectly identified as

clinically complete responders and could thus be exposed to potentially harmful disease progression

during the wait-and-see period. For esophageal cancer, the challenge to distinguish complete from

nearly complete responders seems even more difficult.37-40

However, if surgery can be safely postponed beyond the usual 4 to 6 weeks, without negatively

affecting oncological outcome as suggested in the present study, more time is allowed to identify

noncomplete responders. Such a strategy would not focus on distinguishing complete from

noncomplete responders at a single instance, but would rather use intense surveillance, with targeting

of the mucosa and submucosa41 during repeated instances, for the detection of residual or recurrent

disease. Thereby, possibly allowing more time to differentiate between truly complete responders and

noncomplete responders.

Limitations of the study

The first limitation is that this study was a retrospective analysis. However, all parameters in this study

were collected prospectively. Furthermore, all 3 cohorts used the same neoadjuvant CROSS regimen

and all 3 cohorts had a similar intention to perform surgery at 4 to 6 weeks after completion of nCRT.

The second limitation of this study is that the number of patients with a TTS beyond 10 weeks was

relatively small. Therefore, a subtle effect of TTS on DFS and OS cannot be excluded in patients with a

TTS beyond 10 weeks.

Conclusions

In patients with potentially curable esophageal or junctional cancer, prolonged TTS (beyond 45

days) after neoadjuvant chemoradiotherapy increases the probability of a pathologically complete

response, without improving DFS and OS. It might also slightly increase the probability of postoperative

complications, which can be partly explained by intentional postponement of surgery in some

high-risk patients with worse overall condition. We hypothesize that prolonged TTS allows for more

accurate determination of histopathological response due to continued disintegration of nonviable

cells in some patients, without affecting actual prognosis after neoadjuvant chemoradiotherapy. We

conclude that TTS can be safely prolonged from the usual 4 to 6 weeks up to at least 12 weeks,

which facilitates a more conservative wait-and-see strategy after neoadjuvant chemoradiotherapy to

be tested.


65

Cha

pter

4

REFERENCES1. Sjoquist KM, Burmeister BH, Smithers BM, et al. Survival after neoadjuvant chemotherapy or chemoradiotherapy

for resectable oesophageal carcinoma: an updated meta-analysis. Lancet Oncol. 2011;12:681-692.

2. van Hagen P, Hulshof MC, van Lanschot JJ, et al. Preoperative chemoradiotherapy for esophageal or junctional cancer. N Engl J Med. 2012;366:2074-2084.

3. Mandard AM, Dalibard F, Mandard JC, et al. Pathologic assessment of tumor regression after preoperative chemoradiotherapy of esophageal carcinoma: clinicopathologic correlations. Cancer. 1994;73:2680-2686.

4. Chirieac LR, Swisher SG, Ajani JA, et al. Posttherapy pathologic stage predicts survival in patients with esophageal carcinoma receiving preoperativeBNchemoradiation. Cancer. 2005;103:1347-1355.

5. Schneider PM, Baldus SE, Metzger R, et al. Histomorphologic tumor regression and lymph node metastases determine prognosis following neoadjuvant radiochemotherapy for esophageal cancer: implications for response classification. Ann Surg.2005;242:684-692.

6. Donohoe CL, O’Farrell NJ, Grant T, et al. Classification of pathologic response to neoadjuvant therapy in esophageal and junctional cancer: assessment of existing measures and proposal of a novel 3-point standard. Ann Surg. 2013;258:784-792.

7. Ancona E, Ruol A, Santi S, et al. Only pathologic complete response to neoadjuvant chemotherapy improves significantly the long-term survival of patients with resectable esophageal squamous cell carcinoma: final report of a randomized, controlled trial of preoperative chemotherapy versus surgery alone. Cancer. 2001;91:2165-7214.

8. Reynolds JV, Muldoon C, Hollywood D, et al. Long-term outcomes following neoadjuvant chemoradiotherapy for esophageal cancer. Ann Surg. 2007;245:707-716.

9. Verlato G, Zanoni A, Tomezzoli A, et al. Response to induction therapy in oesophageal and cardia carcinoma using Mandard tumour regression grade or size of residual foci. Br J Surg. 2010;97:719-725.

10. Vallbohmer D, Holscher AH, DeMeester S, et al. A multicenter study of survival after neoadjuvant radiotherapy/chemotherapy and esophagectomy for ypT0N0M0R0 esophageal cancer. Ann Surg. 2010;252:744-749.

11. Francois Y, Nemoz CJ, Baulieux J, et al. Influence of the interval between preoperative radiation therapy and surgery on downstaging and on the rate of sphincter-sparing surgery for rectal cancer: the Lyon R90-01 randomized trial. J Clin Oncol. 1999;17:2396-2402.

12. Dolinsky CM, Mahmoud NN, Mick R, et al. Effect of time interval between surgery and preoperative chemoradiotherapy with 5-fluorouracil or 5-fluorouracil and oxaliplatin on outcomes in rectal cancer. J Surg Oncol. 2007;96:207-212.

13. Tulchinsky H, Shmueli E, Figer A, et al. An interval >7 weeks between neoadjuvant therapy and surgery improves pathologic complete response and disease-free survival in patients with locally advanced rectal cancer. Ann Surg Oncol. 2008;15:2661-2667.

14. Kalady MF, de Campos-Lobato LF, Stocchi L, et al. Predictive factors of pathologic complete response after neoadjuvant chemoradiation for rectal cancer. Ann Surg. 2009;250:582-589.

15. de Campos-Lobato LF, Geisler DP, da Luz Moreira A, et al. Neoadjuvant therapy for rectal cancer: the impact of longer interval between chemoradiation and surgery. J Gastrointest Surg. 2010;15:444-450.

16. Wolthuis AM, Penninckx F, Haustermans K, et al. Impact of interval between neoadjuvant chemoradiotherapy and TME for locally advanced rectal cancer on pathologic response and oncologic outcome. Ann Surg Oncol. 2012;19:2833-2841.

17. Sloothaak DA, Geijsen DE, van Leersum NJ, et al. Optimal time interval between neoadjuvant chemoradiotherapy and surgery for rectal cancer. Br J Surg. 2013;100:933-939.

| Chapter 4

66

18. Petrelli F, Sgroi G, Sarti E, et al. Increasing the interval between neoadjuvant chemoradiotherapy and surgery in rectal cancer: a meta-analysis of published studies. Ann Surg. November 20, 2013 [Epub ahead of print].

19. Moore HG, Gittleman AE, Minsky BD, et al. Rate of pathologic complete response with increased interval between preoperative combined modality therapy and rectal cancer resection. Dis Colon Rectum. 2004;47:279-286.

20. Habr-Gama A, Perez RO, Proscurshim I, et al. Interval between surgery and neoadjuvant chemoradiation therapy for distal rectal cancer: does delayed surgery have an impact on outcome? Int J Radiat Oncol Biol Phys. 2008;71:1181-1188.

21. Kerr SF, Norton S, Glynne-Jones R. Delaying surgery after neoadjuvant chemoradiotherapy for rectal cancer may reduce postoperative morbidity without compromising prognosis. Br J Surg. 2008;95:1534-1540.

22. Lim SB, Choi HS, Jeong SY, et al. Optimal surgery time after preoperative chemoradiotherapy for locally advanced rectal cancers. Ann Surg. 2008;248:243-251.

23. Ruol A, Rizzetto C, Castoro C, et al. Interval between neoadjuvant chemoradiotherapy and surgery for squamous cell carcinoma of the thoracic esophagus: does delayed surgery have an impact on outcome? Ann SUrg. 2010;252:788-796.

24. Kim JY, Correa AM, Vaporciyan AA, et al. Does the timing of esophagectomy after chemoradiation affect outcome? Ann Thorac Surg. 2012;93:207-213.

25. van Meerten E, Muller K, Tilanus HW, et al. Neoadjuvant concurrent chemoradiation with weekly paclitaxel and carboplatin for patients with oesophageal cancer: a phase II study. Br J Cancer. 2006;94:1389-1394.

26. van Meerten E, van der Gaast A, Tilanus HW, et al. Pathological analysis after neoadjuvant chemoradiotherapy for esophageal carcinoma: the Rotterdam experience. J Surg Oncol. 2009;100:32-37.

27. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240:205-213.

28. Clavien PA, Barkun J, de Oliveira ML, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg. 2009;250:187-196.

29. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373-383.

30. Karnofsky DA, Burchenal JH. The clinical evaluation of chemotherapeutic agents in cancer. In: MacLeod CM, ed. Evaluation of Chemotherapeutic Agents. New York, NY: Columbia University Press; 1949:191-205.

31. Mor V, Laliberte L, Morris JN, et al. The Karnofsky Performance Status Scale. An examination of its reliability and validity in a research setting. Cancer. 1984;53:2002-2007.

32. Harrell FE Jr, Lee KL, Pollock BG. Regression models in clinical studies: determining relationships between predictors and response. J Natl Cancer Inst. 1988;80:1198-1202.

33. Habr-Gama A, Perez RO, Nadalin W, et al. Operative versus nonoperative treatment for stage 0 distal rectal cancer following chemoradiation therapy: long-term results. Ann Surg. 2004;240:711-718.

34. Habr-Gama A, Perez RO. Non-operative management of rectal cancer after neoadjuvant chemoradiation. Br J Surg. 2009;96:125-127.

35. Maas M, Beets-Tan RG, Lambregts DM, et al. Wait-and-see policy for clinical complete responders after chemoradiation for rectal cancer. J Clin Oncol. 2011;29:4633-4640.

36. Bokkerink GM, de Graaf EJ, Punt CJ, et al. The CARTS study: Chemoradiation therapy for rectal cancer in the distal rectum followed by organ-sparing transanal endoscopic microsurgery. BMC Surg. 2011;11:34.


67

Cha

pter

4

37. Lordick F, Ott K, Krause BJ, et al. PET to assess early metabolic response and to guide treatment of adenocarcinoma of the oesophagogastric junction: the MUNICON phase II trial. Lancet Oncol. 2007;8:797-805.

38. Schneider PM, Metzger R, Schaefer H, et al. Response evaluation by endoscopy, rebiopsy, and endoscopic ultrasound does not accurately predict histopathologic regression after neoadjuvant chemoradiation for esophageal cancer. Ann Surg. 2008;248:902-908.

39. Vallbohmer D, Holscher AH, Dietlein M, et al. [18F]-Fluorodeoxyglucose-positron emission tomography for the assessment of histopathologic response and prognosis after completion of neoadjuvant chemoradiation in esophageal cancer. Ann Surg. 2009;250:888-894.

40. van Heijl M, Omloo JM, van Berge Henegouwen MI, et al. Fluorodeoxyglucose positron emission tomography for evaluating early response during neoadjuvant chemoradiotherapy in patients with potentially curable esophageal cancer. Ann Surg. 2011;253:56-63.

41. Shapiro J, ten Kate FJ, van Hagen P, et al. Residual esophageal cancer after neoadjuvant chemoradiotherapy frequently involves the mucosa and submucosa. Ann Surg. 2013;258:678-688.

B.A. Grotenhuis

P. van Hagen

B.P.L. Wijnhoven

M.C.W. Spaander

H.W. Tilanus

J.J.B. van Lanschot

Journal of Gastrointestinal Surgery 2010;14(3):476-83

Chapter 5Delay in diagnostic workup and treatment of esophageal cancer

| Chapter 5

70

ABSTRACT

INTRODUCTION

Esophageal cancer should preferably be detected and treated at an early stage, but this may be

prohibited by late onset of symptoms and delays in referral, diagnostic workup, and treatment.

The aim of this study was to investigate the impact of these delays on outcome in patients with

esophageal cancer.

METHODS

For 491 patients undergoing esophagectomy for cancer between 1991 and 2007, patients’ short- and

long-term outcome were analyzed according to different time intervals between onset of symptoms,

diagnosis, and surgical treatment.

RESULTS

Length of prehospital delay (from onset of symptoms until endoscopic diagnosis) did not affect

patient’s short- or long-term outcome. A shorter hospital delay between establishing the diagnosis

of esophageal cancer on endoscopy and surgery was associated with lower overall morbidity and in-

hospital mortality. Patients of ASA classes I and II experienced a shorter hospital delay than patients

of ASA classes III and IV. Length of hospital delay between endoscopic diagnosis and surgery did not

affect pathological tumor–node–metastasis stage or R0-resection rate. Longer hospital delay did not

result in worse survival: Overall survival after esophagectomy for cancer was not significantly different

between patients with hospital delay <5, 5–8, or >8 weeks (24.7%, 21.7%, and 32.3%, respectively;

p = 0.12).

CONCLUSION

A longer hospital delay (between endoscopic diagnosis and surgery) resulted in worse patient’s short-

term outcome (higher overall morbidity and mortality rates) but not in a worse long-term outcome

(overall survival). This may be explained by a more time-consuming diagnostic workup in patients

with a poorer physical status and not by tumor progression.

Delay in diagnostic workup and treatment of esophageal cancer |

71

Cha

pter

5

INTRODUCTION

The 5-year survival rate for esophageal cancer patients after esophagectomy with curative intent has

improved up to 40%.1-3 As further improvement in survival from a single modality approach, such as

surgery, is unlikely, considerable interest has grown in other strategies that may improve patients’

survival (neoadjuvant chemo- and/or radiotherapy in particular). In many types of cancer, the prognosis

of patients with small, localized tumors is better than with locally advanced or metastatic disease.

Similar to other malignancies such as colorectal and breast cancer, the outcome of esophageal cancer

is related to the pathological tumor–node–metastasis (pTNM) stage of the disease.2,4,5 Therefore,

detection and treatment of esophageal cancer at an early stage could also improve long-term survival.

Early detection of esophageal cancer may be prohibited not only by the late onset of symptoms but

also by delays in referral to an appropriate specialist, establishment of the diagnosis, further diagnostic

workup, and start of treatment. However, the impact of these delays on both short- and long-term

outcome for patients undergoing esophagectomy for cancer is unclear.

In patients with breast cancer, delays of 3–6 months between the onset of symptoms and start

of treatment are associated with lower survival, caused by a more advanced tumor stage.6 In two

systematic reviews, no association was found between diagnostic and therapeutic delay and survival

in colorectal cancer patients7 nor between these delays and disease stage.8 A few studies have

investigated the impact of delays in diagnosis and treatment of esophageal cancer. Drawbacks of

these studies are small numbers of patients included,9 analyses that do not cover the complete track

between onset of symptoms and surgical treatment,10,11 combined patient groups with gastric and

esophageal carcinoma,12,13 and studies lacking survival analyses.9, 11-13

We hypothesized that longer delays between onset of symptoms, endoscopic diagnosis, and surgical

treatment are associated with a worse short-term outcome (morbidity, reoperation rate, and in-

hospital mortality), worse tumor stage, and hence, worse long-term outcome (overall survival)

following potentially curative esophagectomy in patients with esophageal cancer.


The Erasmus Medical Center in Rotterdam is a tertiary referral center for patients with esophageal

cancer in The Netherlands. Most patients are referred to the Erasmus MC outpatient clinic for (surgical)

treatment after the diagnosis of esophageal cancer has been established in a referring hospital (group

A). The minority of patients is directly referred by the general practitioner (GP) to the Erasmus MC for

clinical investigations of symptoms suggestive of cancer (group B). In all patients (groups A and B)

upper gastrointestinal endoscopy with biopsy is (re)done in the Erasmus MC to confirm the diagnosis

of esophageal cancer and to determine the exact location of the tumor. Staging is performed routinely

with endoscopic ultrasonography, CT scanning of thorax and abdomen, and external ultrasound of the

neck. Every patient is discussed in a weekly multidisciplinary oncology meeting in which a definitive

treatment plan is designed. If eligible for surgery, patients are put on the waiting list for surgery. On the

| Chapter 5

72

same day, the patient is referred to the Department of Anesthesiology for preoperative counseling. If

needed, additional cardiac and/or pulmonary function tests are scheduled.

Between January 1991 and December 2007, 791 patients underwent esophagectomy for cancer of

the esophagus or gastroesophageal junction in the Erasmus MC. To obtain a homogeneous cohort

of patients in terms of treatment and to circumvent possible stage migration following chemo-

and/or radiotherapy, patients receiving (neo)adjuvant therapy were excluded from this analysis.

In our hospital, patients received neoadjuvant chemo(radio)therapy in the context of randomized

controlled trials.14,15 Induction chemo- and/or radiotherapy was given in patients with either a cT4

tumor without distant metastases or in patients with gross involvement of celiac trunk lymph nodes

(M1a), who were not considered eligible for primary surgical therapy. There were 214 patients who

were excluded because of chemo- and/or radiotherapy prior to surgery. In 44 patients, the hospital

delay from endoscopic diagnosis to surgery could not be calculated, as the date of their first upper

gastrointestinal endoscopy performed in the referring hospital was unknown. Another 42 patients

were excluded, as they participated in a Barrett’s esophagus surveillance program. Over recent years,

multiple attempts for endoscopic treatment of early lesions delayed referral to the Department of

Surgery in such way that this group was not representative for patients treated for (more advanced)

esophageal cancer. Finally, data of 491 patients were analyzed in the present study. The vast majority

of these patients underwent a transhiatal esophagectomy with locoregional lymphadenectomy only

(N = 477). In 14 patients, a transthoracic resection with extended lymphadenectomy was performed.

The applied surgical techniques have been described previously.3, 16 Tumors were assigned pTNM

stages according to the Union Internationale Contre le Cancer 2002 system.17

Data on patients’ demographics, diagnostic tests, surgery, postoperative morbidity, in-hospital

mortality, and survival have been collected prospectively and stored in a database by a data manager.

From this database, the following time points were defined:

- Date of upper gastrointestinal endoscopy in the referring hospital, on which the diagnosis of

esophageal cancer had been established by histology from biopsies (only applicable for group A)

- Date of first visit at the Erasmus MC outpatient clinic: Department of Surgery, Gastroenterology,

or Medical Oncology

- Date of upper gastrointestinal endoscopy in the Erasmus MC, on which the diagnosis of

esophageal cancer had been established by histology from biopsies

- Date of the multidisciplinary oncology meeting, after which the patient had been put on the

operative waiting list if eligible for surgery

- Date of surgery.

To summarize all different time points that have been marked in the process between onset of

symptoms and surgery, we divided this time span into two major time intervals that have been

analyzed separately: pre-hospital and hospital delay (see Fig. 1). Subsequently, data were analyzed in

three different ways:


73

Cha

pter

5

- Impact of prehospital delay: time from onset of symptoms until diagnosis on first endoscopy

(either in the referring hospital for group A or in Erasmus MC for group B)

- Impact of hospital-delay: time from diagnosis on patient’s first endoscopy undertaken until

surgery

- Impact of specific time intervals between diagnosis on first endoscopy and surgery. In order to

examine the hospital-delay in more detail, the effect of specific time intervals between diagnosis

in the referring hospital, first visit at the outpatient clinic in Erasmus MC, diagnosis on endoscopy in

Erasmus MC, multidisciplinary oncology meeting, and surgery on short- and long-term outcome

were analyzed. For this purpose, only data from patients in group A were used.

Onset of symptoms Diagnosis on endoscopy -Referring hospital (group A)

-Erasmus MC (group B)

Multidisciplinary oncology meeting

Surgery

FIGURE 1. Analysis of prehospital and hospital delays encountered by patients who underwent surgical resection for esophageal cancer in Erasmus MC.

Statistics

Follow-up was recorded until December 2008 or until death if earlier and was complete for all patients.

Statistical analysis for non-parametric data was used. Grouped data were compared using the chi-

square, Mann–Whitney U, or Kruskall–Wallis H test. Patients who died due to complications following

esophagectomy (in-hospital mortality) were not excluded from survival analysis. Overall survival

was calculated from the date of operation until the date of last follow-up or death according to the

Kaplan–Meier method. Disease-free survival was assessed from the date of operation until the date of

disease recurrence in case of locoregional recurrence or distant metastases. Univariate analyses were

performed with the log-rank test to identify prognostic variables associated with overall survival after

esophagectomy. Data analyses were carried out with SPSS version 15.0 (SPSS, Chicago, IL, USA).

RESULTS

Patients’ characteristics are shown in Table 1. Three hundred sixty-five patients (74.3%), in whom the

diagnosis esophageal cancer was established in another hospital, were referred to the Erasmus MC

for further staging and treatment (group A). One hundred twenty-six patients (25.7%) were referred

directly to the Erasmus MC by the general practitioner for investigation of symptoms suggestive

| Chapter 5

74

of esophageal cancer (group B). Patients’ first visit to the Erasmus MC was at the Department of

Surgery (N = 338, 68.8%), Department of Gastroenterology (N = 147, 29.9%), or Department of Medical

Oncology (N = 6, 1.3%).

TABLE 1. Clinicopathological characteristics of 491 patients who underwent surgical resection for esophageal cancer and who were included in the present study.

Characteristic N (range)

Age (in years)a 65 (28–89)

N (%)

Gender

Male 399 (81.3%)

Female 92 (18.7%)

ASA classification

I 77 (15.7%)

II 316 (64.4%)

III 96 (19.6%)

IV 2 (0.4%)

Tumor location

Proximal esophagus 8 (1.6%)

Mid esophagus 27 (5.5%)

Distal esophagus 196 (39.9%)

Gastroesophageal junction 260 (53.0%)

Histology

Squamous cell carcinoma 73 (14.9%)

Adenocarcinoma 418 (85.1%)

ASA classification American Society of Anesthesiologists classification.aAge is given as median (range).

Impact of prehospital delay: time from onset of symptoms until first endoscopy

The majority of patients underwent endoscopy for investigation of obstructive symptoms suggestive

of cancer like dysphagia, odynophagia, and weight loss (N = 462, 94.1%). Other indications for

endoscopy encompassed investigation of hematemesis (N = 12, 2.4%), anemia (N = 9, 1.8%), or melena

(N = 8, 1.6%). Prehospital delay (from onset of symptoms until first endoscopy) lasted a median time

period of 3.0 months (range, 0–36 months). Patient’s short-term (morbidity, reoperation rate, and

in-hospital mortality) and long-term outcome (overall 5-year survival) after esophagectomy were

comparable for patients who experienced symptoms for a period of 3 months or less versus more

than 3 months until endoscopy was performed (Table 2).


75

Cha

pter

5

TABLE 2. Impact of prehospital delay from onset of symptoms to first endoscopy on short- and long-term outcome after esophagectomy; comparison of prehospital delay ≤3 months (N = 308) versus >3 months (N = 183).

Prehospital delay ≤3 months, N = 308

Prehospital delay >3 months, N = 183 p value

Morbidity 199 (64.6%) 104 (56.8%) 0.09

Reoperation 34 (11.0%) 16 (8.7%) 0.42

In-hospital mortality 18 (5.8%) 9 (4.9%) 0.66

Overall 5-year survival 24.0% 29.3% 0.10

Impact of hospital delay: time from endoscopic diagnosis until surgery

The hospital delay from establishing the diagnosis of esophageal cancer on endoscopy (either in the

referring hospital for group A or in Erasmus MC for group B) until surgery was 49 days (given as median,

range of 5–175 days). This delay encompassed a median time period of 28 days (range, 0–147 days)

from diagnosis on patient’s first endoscopy until the multidisciplinary oncology meeting (staging

delay), and a median time period of 15 days from this meeting until surgery (operative waiting list,

range of 1–67 days). Median hospital delay between diagnosis and surgery increased during the study

period (1991–2007): 3.9 weeks in 1991 toward 10.9 weeks in 2007 (Fig. 2). This increase in hospital

delay should rather be ascribed to the 3.4 times increase in length of the operative waiting list (1.6

weeks in 1991 towards 5.6 weeks in 2007) than to the 1.5 times increase in staging delay (3.3 weeks in

1991 towards 4.9 weeks in 2007).

3,9

4,7

6,6 6,6 7

7,9

6,9

10,6 10,9

0

2

4

6

8

10

12

1991-1992 1993-1994 1995-1996 1997-1998 1999-2000 2001-2002 2003-2004 2005-2006 2007

FIGURE 2. Median hospital delay (in weeks) between endoscopic diagnosis and surgery increased during the study period (1991–2007): 3.9 weeks in 1991 toward 10.9 weeks in 2007.

| Chapter 5

76

A shorter hospital delay between establishing the diagnosis of esophageal cancer on patient’s first

endoscopy and surgery was associated with significantly lower overall morbidity and mortality (Table

3). These associations appeared to be linear: morbidity (p = 0.001) and in-hospital mortality (p = 0.01)

increased with longer hospital delay. Patients of ASA classes I and II experienced a shorter hospital

delay than patients of ASA classes III and IV (hospital delay <5 weeks, 28.8%; 5–8 weeks, 36.9%; and >8

weeks, 34.4% versus <5 weeks, 15.3%; 5–8 weeks, 41.8%; and > 8 weeks, 42.9%, respectively; p = 0.02).

Length of hospital delay did not affect pTNM stage or R0-resection rate (Table 3).

TABLE 3. Impact of the hospital delay from diagnosis on patient’s first endoscopy until surgery: hospital delay <5 weeks (N = 128), 5–8 weeks (N = 186), and >8 weeks (N = 177).

Delay <5 weeks, N = 128

Delay 5–8 weeks, N = 186

Delay >8 weeks, N = 177 p value

Morbidity 62 (48.4%) 122 (65.6%) 119 (67.2%) <0.01

In-hospital mortality 2 (1.6%) 10 (5.4%) 15 (8.5%) 0.03

Reoperation 7 (5.5%) 20 (10.8%) 23 (13.0%) 0.10

pT stage

pT1–pT2 30 (23.4%) 57 (30.6%) 54 (30.5%) 0.31

pT3–pT4 98 (76.6%) 129 (69.4%) 123 (69.5%)

pN stage

pN0 42 (32.8%) 66 (35.5%) 62 (35.0%) 0.88

pN1 86 (67.2%) 120 (64.5%) 115 (65.0%)

pM stage

pM0 103 (80.5%) 150 (80.6%) 131 (74.0%) 0.24

pM1a–M1b 25 (19.5%) 36 (19.4%) 46 (26.0%)

Radicality of resection

R0 86 (67.2%) 124 (66.7%) 130 (73.4%) 0.32

R1–R2 42 (32.8%) 62 (33.3%) 47 (26.6%)

Longer hospital delay did not result in worse survival (Fig. 3): Overall 5-year survival was 24.7% in

patients with a hospital delay less than 5 weeks, 21.7% in patients with a hospital delay between 5

and 8 weeks and 32.3% in patients in whom the hospital delay was more than 8 weeks. Although

overall survival appeared to be longer in patients with a longer hospital delay, this difference was not

statistically significant (p = 0.12).


77

Cha

pter

5

TABLE 4. Univariable analyses of potential prognostic variables associated with overall survival after esophagectomy for cancer (N = 491).

Variable Five-year survival (%) p value

Age≤65 years 30.2 0.001

>65 years 21.4

SexMale 25.4 0.84

Female 28.5

ASA classificationI–II 27.0 0.12

III–IV 22.2

pT stagepT1–T2 53.3 <0.001

pT3–T4 15.0

pN stagepN0 50.3 <0.001

pN1 12.2

pM stagepM0 39.8 <0.001

pM1a–M1b 9.5

HistologySquamous cell carcinoma 27.1 0.98

Adenocarcinoma 25.8

Differentiation grade of tumorGood 69.1

Moderate 29.5 <0.001

Poor 16.0

Radicality of resectionR0 35.5 <0.001

R1–R2 5.5

Lymph node ratio≤0.24 36.0 <0.001

>0.24 12.0

ReferralBy another hospital (group A) 25.9 0.65

By GP (group B) 26.2

Prehospital delay≤3 months 24.0 0.10

>3 months 29.3

Hospital delay<5 weeks 24.7

5–8 weeks 21.7 0.12

>8 weeks 32.3

ASA classification American Society of Anesthesiologists classification, GP general practitioner.

http://link.springer.com/search?dc.title=Squamous+cell+carcinoma&facet-content-type=ReferenceWorkEntry&sortOrder=relevance

| Chapter 5

78

Parameters found to be associated with overall survival in univariate analyses are shown in Table 4:

age younger than 65 years, early pT stage (pT1 or pT2), no lymph node involvement (pN0), absence

of distant metastatic disease (pM0), good differentiation grade of the tumor, R0 resection, and lymph

node ratio smaller than 0.24 were favorable of improved overall survival. Survival analysis with regard

to 5-year disease-free survival paralleled the overall 5-year survival curves (27.0%, 27.7%, and 38.3%,

respectively; p = 0.09).

No at risk 0 months 12 months 24 months 36 months 48 months 60 months >8 weeks 177 107 68 48 38 24 5-8 weeks 186 122 78 54 39 34 <5 weeks 128 80 53 42 32 30

FIGURE 3. Overall 5-year survival for esophageal cancer patients appeared longer for patients with a hospital delay between diagnosis on first endoscopy and surgery >8 weeks (N = 177) versus patients with a hospital delay <5 weeks (N = 128) or 5–8 weeks (N = 186), although this difference did not reach statistical significance (p = 0.12).

Impact of specific time intervals between endoscopic diagnosis and surgery (Group A)

The median hospital delay was 53 days (range, 5–175 days) for patients in group A in whom the

diagnosis esophageal cancer had been established in another hospital and who were referred to

the Erasmus MC for surgical treatment (N = 365). The breakdown of this delay is shown in Table 5,

according to the different time intervals between diagnosis in the referring hospital, first visit to the

outpatient clinic in Erasmus MC, diagnosis on endoscopy in Erasmus MC, multidisciplinary oncology

meeting, and surgery.

When analyzing the impact of the separate time intervals, it appeared that the delay between the

multidisciplinary oncology meeting and surgery (median, 15 days; reflecting the length of the operative

waiting list) was the only time interval that influenced short-term outcome post-esophagectomy.


79

Cha

pter

5

Although in-hospital mortality was comparable between patients who had been on the waiting list

for 15 days or shorter versus patients who were waiting for more than 15 days (p = 0.14), length of

the operative waiting list did influence morbidity (55.7% versus 67.1%, p = 0.03), and a trend towards

an increased reoperation rate could be noted (7.8% versus 13.9%, p = 0.06). However, in contrast with

the hospital delay between endoscopic diagnosis and surgery, none of the separate time intervals

affected long-term survival.

TABLE 5. Delays encountered by esophageal cancer patients who have been referred from an other hospital to the Erasmus MC for surgical treatment (group A, N = 365).

Diagnosis on endoscopy elsewhere → first visit outpatient clinic Erasmus MC 17 days (1–138)

First visit outpatient clinic Erasmus MC → diagnosis on endoscopy Erasmus MC 6 days (0–36)

Diagnosis on endoscopy Erasmus MC → multidisciplinary oncology meeting 7 days (0–95)

Multidisciplinary oncology meeting → surgery 15 days (1–67)

Total hospital delay

Diagnosis on endoscopy elsewhere → surgery 53 days (5–175

Lengths of delays are given as a median values with the corresponding range in brackets.

DISCUSSION

When initiating the current study, we hypothesized that longer delays between onset of symptoms,

diagnosis, and surgical treatment are associated with worse short-term outcome (in terms of

morbidity, reoperation rate, and mortality) and worse long-term outcome (overall survival) following

esophagectomy for cancer. In the present series, it appeared that length of prehospital delay (from

onset of symptoms until endoscopic diagnosis) did not influence patient’s short-term outcome or

overall 5-year survival. Onset of symptoms is a subjective measurement, and it may be that patients

are not able to recall the exact moment that they first experienced discomfort. Furthermore, although

little information is known about the tumor doubling time of esophageal cancer, the period of time in

which a patient is symptomatic may be relatively short when compared to the total period between

the first presence of malignant cells in the esophagus and the diagnosis of esophageal cancer.

Unfortunately, we did not have information on delays caused by the GP (i.e., time between onset of

symptoms and referral for endoscopy). Nevertheless, we do want to emphasize the importance of

both patient and primary care education that will result in earlier notification of alarming symptoms

such as dysphagia and weight loss.

A longer hospital delay from endoscopic diagnosis until surgery was associated with higher overall

morbidity and mortality. This could be explained by a more thorough and time-consuming diagnostic

workup in patients with a poorer physical status. Indeed, in the present study, patients of ASA classes I

and II experienced a shorter hospital delay than patients of ASA classes III and IV. Alternatively, a longer

delay prior to surgery may also have caused a worse physical status in esophageal cancer patients by

| Chapter 5

80

means of malnutrition. However, this remains speculative, as our database did not provide detailed

information with regard to patients’ preoperative nutritional status (e.g., nutritional risk indices). When

analyzing the impact of the separate time intervals between patient’s first endoscopy and surgery, it

appeared that the length of the operative waiting list was the time interval that influenced short-term

outcome following esophagectomy the most. From the literature, it is also known that the quality of

life in newly diagnosed esophageal cancer patients who are waiting for surgery is seriously impaired.18

Hence, it should be aimed for to keep this time interval to a minimum.

Our second hypothesis was that patients with longer delays would generally present with more

advanced disease and that this relation between delay and stage would result in a poorer survival.

However, pTNM stages were comparable in patients with a hospital delay <5, 5-8, or >8 weeks

between endoscopy and surgery. Surprisingly, it appeared that overall survival was improved in

patients with a longer hospital delay, although this difference was not statistically significant. This is

in line with the results of Kötz et al.10 who showed that a longer delay between diagnosis and surgical

resection was associated with improved survival in esophageal cancer patients. However, the delay

between diagnosis and surgery was not an independent prognostic variable on multivariate analysis

in their study. Kötz et al.10 noted that patients with a longer delay had a higher rate of complete tumor

resection, suggesting that they were more appropriately selected for surgical treatment. In our series,

we could not find evidence that patients were selected more appropriately, as both pTNM stage and

R0-resection rate did not differ between patients with a shorter or longer hospital delay. However,

hospital delay substantially increased especially over the last few years in our hospital (Fig. 2). This

can probably explain the counter-intuitive correlation between longer hospital delay and improved

long-term survival, which is rather reflecting state-of-the-art staging modalities, refined surgical

techniques, and improved intensive care that have been introduced over the past years. Theoretically,

it could also be possible that, in our hospital, patients did not undergo surgery anymore after a longer

hospital delay in case the tumor progressed to a stage that was considered irresectable. However, in

our patient group, the increased hospital delay can rather be ascribed to an increase in length of the

operative waiting list than to an increased staging delay. As the decision on whether to operate or

not has been made during the multidisciplinary oncology meeting, it is unlikely that a longer hospital

delay led to a dropout of patients with irresectable tumors and, hence, a more selected patient group

that underwent esophagectomy.

It is evident that efforts are taken to minimize delays experienced by patients with esophageal cancer

between onset of symptoms, diagnosis, and surgical treatment. The National Health Service cancer

plan was implemented in 2000 in the UK, indicating that all patients with relevant symptoms and

suspected cancer should be able to see a specialist within 2 weeks of their GP referral. The introduction

of these guidelines was associated with reductions in times to first outpatient visit, endoscopy,

and diagnosis in patients with upper gastrointestinal cancer (esophageal or gastric).19,20 However,

the effectiveness of the NHS cancer plan is uncertain, as it can be questioned whether the slightly

improved survival rates after 2000 can be ascribed to this plan.21


81

Cha

pter

5

In our hospital, we recently introduced a new schedule of diagnostic services for patients with

suspected esophageal cancer. It is attempted to see patients at the outpatient clinic of the Department

of Surgery or Department of Gastroenterology within 1 week after referral. Furthermore, patients are

offered all imaging modalities in 1 week, including upper gastrointestinal endoscopy, endoscopic

ultrasonography, CT scanning of thorax and abdomen, and external ultrasound of the neck. The aim

of this schedule is to minimize the delay between referral to our hospital and establishment of a

definitive treatment plan for each individual patient.

In conclusion, length of prehospital delay (from onset of symptoms until diagnosis) did not affect

patient’s short- or long-term outcome. A longer hospital delay (between endoscopic diagnosis and

surgery) resulted in worse patient’s short-term outcome (higher overall morbidity and mortality

rates) but not in worse long-term outcome (overall survival). This may be explained by a more time-

consuming diagnostic workup in patients with a poorer physical status and not by tumor progression.

| Chapter 5

82

REFERENCES1. Lerut T, De Leyn P, Coosemans W et al. Surgical strategies in esophageal carcinoma with emphasis on radical

lymphadenectomy. Ann Surg 1992;216(5):583–590

2. Siewert JR, Stein HJ, Feith M et al. Histologic tumor type is an independent prognostic parameter in esophageal cancer: lessons from more than 1,000 consecutive resections at a single center in the Western world. Ann Surg 2001;234(3):360–367. (discussion 368–369).

3. Hulscher JB, van Sandick JW, de Boer AG et al. Extended transthoracic resection compared with limited transhiatal resection for adenocarcinoma of the esophagus. N Engl J Med 2002;347(21):1662–1669.

4. Eloubeidi MA, Desmond R, Arguedas MR et al. Prognostic factors for the survival of patients with esophageal carcinoma in the U.S.: the importance of tumor length and lymph node status. Cancer 2002;95(7):1434–1443.

5. Wijnhoven BP, Tran KT, Esterman A et al. An evaluation of prognostic factors and tumor staging of resected carcinoma of the esophagus. Ann Surg 2007;245(5):717–725.

6. Richards MA, Westcombe AM, Love SB et al. Influence of delay on survival in patients with breast cancer: a systematic review. Lancet 1999;353(9159):1119–1126.

7. Ramos M, Esteva M, Cabeza E et al. Relationship of diagnostic and therapeutic delay with survival in colorectal cancer: a review. Eur J Cancer 2007;43(17):2467–2478.

8. Ramos M, Esteva M, Cabeza E et al. Lack of association between diagnostic and therapeutic delay and stage of colorectal cancer. Eur J Cancer 2008;44(4):510–521.

9. Wang J, Liu F, Gao H et al. The symptom-to-treatment delay and stage at the time of treatment in cancer of esophagus. Jpn J Clin Oncol 2008;38(2):87–91.

10. Kotz BS, Croft S, Ferry DR. Do delays between diagnosis and surgery in resectable oesophageal cancer affect survival? A study based on West Midlands cancer registration data. Br J Cancer 2006;95(7):835–840.

11. Rothwell JF, Feehan E, Reid I et al. Delay in treatment for oesophageal cancer. Br J Surg 1997;84(5):690–693.

12. Martin IG, Young S, Sue-Ling H, Johnston D. Delays in the diagnosis of oesophagogastric cancer: a consecutive case series. BMJ 1997;314(7079):467–470.

13. Witzig R, Schonberger B, Fink U et al. Delays in diagnosis and therapy of gastric cancer and esophageal adenocarcinoma. Endoscopy 2006;38(11):1122–1126.

14. Kok TC, Van Lanschot JJB, Siersema PD et al. Neoadjuvant chemotherapy in operable esophageal squamous cell cancer: final report of a phase III multicenter randomized controlled trial. Proc Am Soc Clin Oncol 1997;17:984.

15. van Heijl M, van Lanschot JJ, Koppert LB et al. Neoadjuvant chemoradiation followed by surgery versus surgery alone for patients with adenocarcinoma or squamous cell carcinoma of the esophagus (CROSS). BMC Surg 2008;8:21.

16. Omloo JM, Lagarde SM, Hulscher JB et al. Extended transthoracic resection compared with limited transhiatal resection for adenocarcinoma of the mid/distal esophagus: five-year survival of a randomized clinical trial. Ann Surg 2007;246(6):992–1000. (discussion 1000–1001).

17. Sobin L. TNM classification of malignant tumors, 6th edn. Wiley: New York, 2002.

18. Visser MR, van Lanschot JJ, van der Velden J et al. Quality of life in newly diagnosed cancer patients waiting for surgery is seriously impaired. J Surg Oncol 2006;93(7):571–577.

19. Logan RF, Behan MW, Campion G, Keating NA. Impact of the two-week referral guideline on time to diagnosis and treatment in oesophago-gastric cancer. Clin Med 2003;3(4):386–387.


83

Cha

pter

5

20. Spahos T, Hindmarsh A, Cameron E et al. Endoscopy waiting times and impact of the two week wait scheme on diagnosis and outcome of upper gastrointestinal cancer. Postgrad Med J 2005;81(961):728–730.

21. Rachet B, Maringe C, Nur U et al. Population-based cancer survival trends in England and Wales up to 2007: an assessment of the NHS cancer plan for England. Lancet Oncol 2009;10(4):351–369.

P. van Hagen

M.C.W. Spaander

A. van der Gaast

C.M. van Rij

H.W. Tilanus

J.J.B. van Lanschot

B.P.L. Wijnhoven

International Journal of Clinical Oncology 2013;18(2):214-9.

Chapter 6Impact of a multidisciplinary tumor board meeting for

upper-GI malignancies on clinical decision making: a prospective cohort study

| Chapter 6

86

ABSTRACT

BACKGROUND

The Dutch guidelines for diagnosis and treatment of upper-GI malignancies recommend review of

patients by a multidisciplinary tumor board (MDT). The purpose of this study was to determine the

effect on clinical decision making of an MDT for patients with upper-GI malignancies.

METHODS

All physicians participating in the MDT completed an electronic standardized case form to delineate

their proposed treatment plan for the patients they presented, including the intent of treatment and

the modality of treatment. This therapeutic or diagnostic proposal was then compared with the plan

on which consensus was reached by the MDT.

RESULTS

A total of 252/280 (90.0%) forms were completed and suitable for analysis. In 87/252 (34.5%) of the

case presentations, the MDT altered the proposed plan of management. In 29/87 (33.3%) cases, a

more extensive diagnostic work-up was decided upon. In 8/87 (9.2%) cases the curative intent of the

proposed treatment was altered to palliation only. In 2/75 (2.7%) cases, however, it was decided that a

patient could be treated with curative intent instead of the proposed palliative intent.

CONCLUSION

In over 1/3 of cases, the diagnostic work-up or treatment plan is altered after evaluation by a

multidisciplinary tumor board. This study supports Dutch guidelines recommending discussion of

patients with upper-GI malignancies by a multidisciplinary tumor board.

Impact of a multidisciplinary tumor board meeting for upper-GI malignancies on clinical decision making |

87

Cha

pter

6

INTRODUCTION

Diagnosis and treatment of upper-GI malignancies can be challenging, and most cases require a

multidisciplinary approach. The wide variety of diagnostic tools, (neo)adjuvant treatment regimens,

surgical techniques, and palliative treatment options makes it almost impossible for an individual

specialist to stage and treat a patient with esophageal or gastric cancer1. Many Dutch University

and community hospitals have multidisciplinary tumor board (MDT) meetings in which patients are

discussed by a team of dedicated specialists. The need for multidisciplinary treatment of patients is

also acknowledged in the Dutch guidelines for treating esophageal and gastric malignancies2.

In many other fields of oncologic practice, it has been shown that discussing patients in an MDT can

improve survival3 and resection rates4 and alter treatment plans for many patients. These alterations

of treatment plans have been reported for 24% of patients with head and neck malignancies5 and for

up to 52% of breast cancer patients6 discussed in an MDT. Overall, the variety of decisions made by

individual practitioners can be greatly reduced by the process of multidisciplinary decision making7.

Especially nowadays, when most patients with upper-GI malignancies are offered multimodal

treatment and the number of treatment options is increasing, communication between the treating

specialists should be optimum, to decide on the best possible care. MDTs can substantially contribute

to efficient and clear communication between specialists8.

The exact clinical effect of decisions made by an MDT is hard to determine. In several studies, alteration

of treatment plans after discussion by an MDT is taken as a measure of clinical effect5,6,9. In contrast

with, for instance, breast, lung, and gynecologic malignancies, little is known about the value of MDTs

for patients with upper-GI malignancies. The purpose of this study was to determine the effect of the

MDT for upper-GI malignancies on clinical decision making in a tertiary referral center.

METHODS

In our tertiary referral center, a weekly MDT meeting for upper-GI malignancies is held in which an

upper-GI surgeon, medical oncologist, gastroenterologist, radiation oncologist, and pathologist

participate. For newly diagnosed GI malignancies, patients were discussed by the MDT after

standardized pretreatment work-up had been performed. Patients with endoscopically resectable

tumors (T1N0) were discussed only after failure of endoscopic treatment. The participating specialist

to whom the patient was initially referred presents the case to the MDT. After discussion by the MDT,

consensus is reached on a treatment plan or further diagnostic workup. The decision made by the

MDT is binding and is digitally recorded in the patients’ electronic medical records.

Over an 8-month period, participating physicians were asked to complete the first part of a

standardized digital case form. Detailed description of patient demographics, results of diagnostic

tests, and proposed treatment are obligatory before a patient is discussed by an MDT. After this

discussion, the second part of the case form is completed by the chairman of the MDT; this contains

| Chapter 6

88

a summary of the decision of the panel, and the definitive plan of management. Possible treatment

plans that could be proposed by the presenting specialist and then decided upon by the MDT are

shown in Fig. 1.

new diagnosis recurrence second opinion pathology results other

curative palliative not applicable

primary surgery transhiatal resection transthoracic resection (neo)adjuvant/induction therapy followed by transthoracic resection transhiatal resection definitive chemoradiotherapy other

chemotherapy radiotherapy external beam radiotherapy brachytherapy chemoradiotherapy endoluminal stenting other

follow-up further diagnostic work-up - radiologic - gastroduodenoscopy/ endo-ultrasonography - histologic investigation no therapeutic options

definitive treatment plan



Reason for presentation

Intent of treatment

Treatment modality

Other options

Definitive treatment plan

FIGURE 1. Options for filling out the standardised digital multidisciplinary tumour (MDT) board meeting case form. All options are valid for both proposed and definitive treatment plan.

For this analysis, the initially proposed treatment plans were compared with the definitive treatment

plans. If after discussion by the MDT the treatment plan was altered, the level at which this alteration

occurred was determined and scored. Alterations could occur at the level of intent of treatment (e.g.

curative intent was altered to “not applicable” because further diagnostic work-up had to take place),

at the level of treatment modality (e.g. primary surgery was altered to definitive chemoradiotherapy),

and within the level of treatment modality (e.g. transthoracic resection was altered to transhiatal


89

Cha

pter

6

resection). Alterations in treatment plans that were made after discussion by the MDT were not

assessed by a third party.

Because all data were anonymized, obtaining patients’ informed consent for this study was not

obligatory according to the medical ethics committee in our center. Alterations in treatment plan

were expressed as percentage of discussed cases, or as percentage of cases in which alteration of

a treatment plan occurred. Statistical analysis was performed by use of Statistical Package for Social

Sciences version 17.0 (SPSS, Chicago, IL, USA).

RESULTS

Over an 8-month period, 34 MDT meetings were held, in which 171 patients were discussed. A total

of 280 case forms were completed, one for each time a patient was discussed. Of these case forms,

252 were complete and suitable for analysis. For most patients the case was presented to the panel by

the upper-GI surgeon (54.0%), followed by the medical oncologist, gastroenterologist, and radiation

oncologist (23.8, 18.3 and 3.2%, respectively).

One hundred-and-eight (42.9%) case forms were completed for patients with newly diagnosed

esophagogastric malignancies whereas for 33 (13.1%) the reason for presentation was recurrent

disease. Fifteen (6.0%) patients were discussed as a second opinion and in 39 (15.5%) cases the

pathology results after resection were discussed. In 57 (22.6%) patients, there was another reason for

presentation, for example re-evaluation after chemo(radio)therapy or new diagnostic findings. The

clinical characteristics of patients, the number of patients per presenting specialty, and the reasons for

presentation are summarized in Table 1.

One hundred-and-eighteen (46.8%) of the 252 proposed treatment plans were initially curative, and

75/252 (29.8%) palliative. In 59/252 (23.4%) of the cases presented, a curative or palliative intent could

not be scored, for example when pathologic findings after resection were evaluated.

For 87/252 (34.5%) of the proposed treatment plans, the MDT altered the proposed treatment. In

8/118 (6.8%) cases for which a curative treatment plan was proposed, the treatment plan was altered

to palliative treatment. Only 2/75 (2.7%) cases with a proposed palliative treatment plan were altered

to curative treatment. Of the proposed treatment plans for which no intent of treatment could be

applied, because further diagnostic work up was proposed, 6/59 (10.2%) were altered to curative or

palliative treatment after discussion.

For 31/87 (35.6%) cases in which an alteration occurred, another treatment modality was decided

upon, without altering the intent of treatment. In 29/87 (33.3%) cases, a more extensive diagnostic

work-up was decided upon. In 2/87 (2.3%) cases, another treatment within the same modality

(another form of radiotherapy) was decided on. All proposed and definitive management plans are

summarized in Table 2.

| Chapter 6

90

TABLE 1. Characteristics of 171 patients presented to the multidisciplinary tumor board (MDT)

Characteristic n (%)

Age (n = 171)

Median 63

Range 39–86

Sex (n = 171)

Male 129 (75.4%)

Female 42 (24.6%)

Tumor histology (n = 171)

Squamous cell carcinoma 46 (26.9%)

Adenocarcinoma 113 (66.1%)

Other 12 (7.0%)

Tumor location (n = 171)

Esophagus 94 (55.0%)

EGJ 34 (19.9%)

Stomach 41 (24.0%)

Not applicable 2 (1.2%)

cT-stagea (n = 171)

0 1 (0.6%)

1 9 (5.3%)

2 21 (12.3%)

3 87 (50.9%)

4 17 (9.9%)

Unknown 36 (21.1%)

cN-stagea (n = 171)

0 29 (17.0%)

1 117 (68.4%)

Unknown 25 (14.6%)

cM-stageb (n = 171)

0 107 (62.6%)

1a 24 (14.0%)

1b 21 (12.3%)

Unknown 19 (11.1%)

Presenting specialty (n = 252)

Surgery 136 (54.0%)

Medical oncology 60 (23.8%)

Gastroenterology 46 (18.3%)


91

Cha

pter

6

TABLE 1. Continued

Characteristic n (%)

Radiotherapy 8 (3.2%)

Other 2 (0.8%)

Reason for presentation (n = 252)

Newly diagnosed 108 (42.9%)

Recurrent disease 33 (13.1%)

Second opinion 15 (6.0%)

Pathology results 39 (15.5%)

Other 57 (22.6%)

EGJ, esophago-gastric junctionaClinical T-stage and N-stage according to the 6th edition of the TNM classification of the Union International Contre le Cancer (UICC), determined by use of endo-ultrasonography (and/or computed tomography for T4 tumors)bClinical M-stage according to the 6th edition of the TNM classification of the Union International Contre le Cancer (UICC), determined by use of computed tomography and/or external ultrasonography of the neck.

http://link.springer.com/search?dc.title=UICC&facet-content-type=ReferenceWorkEntry&sortOrder=relevance

http://link.springer.com/search?dc.title=UICC&facet-content-type=ReferenceWorkEntry&sortOrder=relevance

| Chapter 6

92

TABL

E 2.

Pro

pose

d an

d de

finiti

ve m

anag

emen

t pl

ans

for

252

cons

ecut

ive

case

s di

scus

sed

by t

he m

ultid

isci

plin

ary

tum

or b

oard

(M

DT)

for

upp

er g

astr

oint

estin

al

mal

igna

ncie

s

Defi

nitiv

e m

anag

emen

t pla

n

Prop

osed

trea

tmen

t pla

nCu

rativ

e in

tent

prim

ary

surg

ery

tran

shia

tal

rese

ctio

ntr

anst

hora

cic

rese

ctio

not

her

(neo

)adj

uvan

t/in

duct

ion

chem

othe

rapy

defin

itive

ch

emo-

radi

othe

rapy

othe

r

Palli

ativ

e tr

eatm

ent

optio

nsC

hem

o-th

erap

yRa

dio-

ther

apy

Exte

rnal

be

am

radi

atio

nBr

achy

-th

erap

yC

hem

o-ra

diot

hera

pyEn

dolu

min

al

sten

ting

othe

r

no o

ptio

ns/

best

su

ppor

tive

care

Not

ap

plic

able

Follo

w-

up

Furt

her

diag

nost

ic

wor

k-up

118

Cura

tive

inte

nt92

818

35 p

rimar

y su

rger

y23

97

74

12

14

14 tr

ansh

iata

l res

ectio

n9

21

11

8 tr

anst

hora

cic

rese

ctio

n7

1

14 o

ther

72

14

64(n

eo)a

djuv

ant/

indu

ctio

n ch

emot

hera

py1

145

11

32

11

11

9de

finiti

ve

chem

orad

ioth

erap

y7

11

1

10ot

her

73

75Pa

lliat

ive

trea

tmen

t op

tions

262

11

32ch

emot

hera

py 1

192

11

42

4

16ra

diot

hera

py2

87

11

14

11 e

xter

nal b

eam

ra

diat

ion

16

61

3

5 b

rach

ythe

rapy

21

11

11

7ch

emor

adio

ther

apy

15

1

7en

dolu

min

al s

tent

ing

24

1

11ot

her

11

21

14

21

2no

opt

ions

/bes

t su

ppor

tive

care

2

59N

ot a

pplic

able

42

53

32fo

llow

-up

11

31

27fu

rthe

r dia

gnos

tic

wor

k-up

31

11

11

22

252

TOTA

L98

2811

89

509

1172

3015

610

47

8233

4


93

Cha

pter

6

DISCUSSION

The purpose of this study was to determine the clinical effect of discussing patients with upper-GI

malignancies in an MDT meeting. These data show that in 87/252 (34.5%) of cases, the initially proposed

management plan was altered after discussion by the MDT. In most cases, alterations concerned

specific details of the initially proposed treatment modality, although in 16/252 (6.4%) cases the intent

of treatment (curative or palliative) was modified.

Diagnosis and treatment of esophageal and gastric malignancies has long been a process in which

the different individual specialists were responsible for sequential steps in the process. Staging

was performed by the gastroenterologist; surgery was then performed by the surgeon and the

radiotherapeutic and medical oncologist were involved when palliation was the only option.

Much has changed, however, over recent decades, resulting in multi-modal treatment for upper-GI

malignancies1, 10, 11. Neoadjuvant and adjuvant therapies are now important in the treatment of upper-

GI malignancies. Moreover, diagnostic tools have evolved, necessitating the input of several specialists

for more accurate staging12. Communication and cooperation between these specialists are of great

importance8, justifying the need for multi-disciplinary discussion by MDTs.

For patients with gynecologic malignancies, Greer et al.9 showed that evaluation by an MDT altered

clinical management in 20.0% of cases that were presented for a review of pathology and in 10.0%

of cases presented for a review of radiology. Another retrospective study performed by Chang et al.13

for 75 patients with breast malignancies showed that the multidisciplinary panel disagreed with the

external presenting physician in 43% of cases. However, both studies were retrospective in nature, and

involved tumors for which pathology has a much greater effect on clinical decision making than for

upper-GI tumors (e.g. hormone receptor status in breast malignancies).

Wheless et al.5 published a prospective study of patients with head and neck tumors for which the

clinical effect of discussion by an MDT was described. Malignancies of the head and neck are similar to

upper-GI malignancies in that they are often diagnosed and treated by a multidisciplinary team. In that

study, in 27.0% of the cases discussed there was an alteration in primary tumor diagnosis, tumor stage,

or treatment plan. This percentage is comparable with the percentage found in our study.

A study performed by Bumm et al.14 involving the effect of an MDT for upper-GI malignancies showed

that the proposed treatment plan was altered in 36.0% of the cases discussed. In another 6% of cases,

further diagnostic work-up was decided upon. That is, as far as we are aware, the only study performed

so far for patients with an upper-GI malignancy. Unfortunately, that study does not describe the actual

alterations in treatment, so the effect on clinical decision making is hard to determine. One can imagine

that alterations within the same treatment modality have less effect than changing a curative intent to

a palliative intent of treatment (or vice versa).

The data for this study were collected prospectively in a high-volume center with special interest in

this type of patient. The clinical characteristics of the study population are sufficiently heterogeneous

to ensure a wide variety of possible treatments. Devitt et al.15 have already shown that health-care

professionals regard the MDT as a useful tool in the treatment of patients with malignancies and value

the opportunity for peer review that is provided by MDT meetings.

| Chapter 6

94

Case discussion by an MDT may cause a delay between diagnosis and start of treatment. Leo et al.16

found the median delay was 20 days for patients treated for lung malignancies. However, in a study

performed in our center by Grotenhuis et al.17 it was shown that a treatment delay longer than 5

weeks (including discussing patients in the MDT) resulted in more post-operative morbidity, but did

not affect long-term survival. However, as Grotenhuis et al. also stated, higher postoperative morbidity

and mortality for patients with a longer delay is also very likely to be caused by case selection.

Patients with severe co-morbidity usually have a longer preoperative work-up before undergoing

esophagectomy, and are more prone to suffer from post-operative morbidity and even mortality.

CONCLUSION

Discussing patients with upper-GI malignancies in a multidisciplinary tumor board meeting results in

an altered clinical decision in more than one third of cases. This study supports the recommendations

of the Dutch “Guidelines for the diagnosis and treatment of upper-GI malignancies” that all patients

should be discussed in the setting of an MDT before initiation of therapy.


95

Cha

pter

6

REFERENCES1. Kaifi JT, Gusani NJ, Jiang Y et al (2011) Multidisciplinary management of early and locally advanced esophageal

cancer. J Clin Gastroenterol 45:391–399

2. Siersema PD, Rosenbrand CJ, Bergman JJ et al (2006) Richtlijn ‘diagnostiek en behandeling oesofaguscarcinoom’ [guideline ‘diagnosis and treatment of oesophageal carcinoma’]. Ned Tijdschr Geneeskd 150:1877–1882

3. Junor EJ, Hole DJ, Gillis CR (1994) Management of ovarian cancer: referral to a multidisciplinary team matters. Br J Cancer 70:363–370

4. Davison AG, Eraut CD, Haque AS et al (2004) Telemedicine for multidisciplinary lung cancer meetings. J Telemed Telecare 10:140–143

5. Wheless SA, McKinney KA, Zanation AM (2010) A prospective study of the clinical impact of a multidisciplinary head and neck tumor board. Otolaryngol Head Neck Surg 143:650–654

6. Newman EA, Guest AB, Helvie MA et al (2006) Changes in surgical management resulting from case review at a breast cancer multidisciplinary tumor board. Cancer 107:2346–2351

7. Pfeiffer SI, Naglieri JA (1983) An investigation of multidisciplinary team decision-making. J Learn Disabil 16:588–590

8. Ruhstaller T, Roe H, Thurlimann B et al (2006) The multidisciplinary meeting: an indispensable aid to communication between different specialities. Eur J Cancer 42:2459–2462

9. Greer HO, Frederick PJ, Falls NM et al (2010) Impact of a weekly multidisciplinary tumor board conference on the management of women with gynecologic malignancies. Int J Gynecol Cancer 20:1321–1325

10. Menges M (2011) Gastric cancer: where is the place for the surgeon, the oncologist and the endoscopist today? World J Gastrointest Oncol 3:10–13

11. Quiros RM, Bui CL(2009) Multidisciplinary approach to esophageal and gastric cancer. Surg Clin North Am 89:79–96, viii

12. Low DE (2011) Update on staging and surgical treatment options for esophageal cancer. J Gastrointest Surg 15:719–729

13. Chang JH, Vines E, Bertsch H et al (2001) The impact of a multidisciplinary breast cancer center on recommendations for patient management: the University of Pennsylvania experience. Cancer 91:1231–1237

14. Bumm R, Feith M, Lordick F et al (2009) Impact of multidisciplinary tumor boards on diagnosis and treatment of esophageal cancer. Eur Surg 39:136–140

15. Devitt B, Philip J, McLachlan SA (2010) Team dynamics, decision making, and attitudes toward multidisciplinary cancer meetings: health professionals’ perspectives. J Oncol Pract 6:e17–e20

16. Leo F, Venissac N, Poudenx M et al (2007) Multidisciplinary management of lung cancer: how to test its efficacy? J Thorac Oncol 2:69–72

17. Grotenhuis BA, van Hagen P, Wijnhoven BP et al (2010) Delay in diagnostic workup and treatment of esophageal cancer. J Gastrointest Surg 14:476–483

P. van Hagen

K. Biermann

J.E. Boers

O. Stoss

H.F. Sleddens

J.J.B. van Lanschot

W.N.M. Dinjens

J. Rueschoff

B.P.L. Wijnhoven

Diseases of the Esophagus 2015, 28(4):380-5

This study was financially supported by an unrestricted grant from Roche b.v., Woerden, the Netherlands

Chapter 7Human epidermal growth factor receptor 2

overexpression and amplification in endoscopic biopsies and resection specimens in esophageal

and junctional adenocarcinoma

| Chapter 7

98

ABSTRACT

AIMS

Human Epidermal growth factor Receptor 2 (HER2) is overexpressed in a subset of esophageal

adenocarcinomas. Frequently, biopsy material is used for evaluation of HER2 status. The aim of the

study was to determine if HER2 expression in preoperative endoscopic biopsies is representative for

the entire tumor.

METHODS

Preoperative endoscopic biopsies and matched resection specimens were collected from 75 patients

who underwent esophagectomy for esophageal adenocarcinoma. Immunohistochemical staining

(IHC) on HER2 and dual-color in-situ hybridization (ISH) were performed. HER2 status was determined

by following a clinical algorithm, first determining HER2 overexpression on IHC, and when equivocal

(2+), determining HER2 amplification on ISH.

RESULTS

71/75 (95%) biopsies and 69/75 (92%) resection specimens could be analyzed, due to technical failure.

HER2 positivity was seen in 18/71 (25%) biopsies and in 15/69 (22%) resection specimens. Overall,

HER2 status in the biopsy was concordant with HER2 status in the resection specimen in 94% of

cases. Interobserver agreement on IHC-scoring for all three observers was 83% in biopsies and 85%

in resection specimens.

CONCLUSION

HER2 positivity was detected in 22% of esophageal adenocarcinomas. Although inter-observer

agreement was moderate, HER2 status of a primary tumor can be reliably determined based on the

endoscopically obtained pretreatment biopsy.

HER2/neu overexpression in biopsies and resection specimens of esophageal adenocarcinoma |

99

Cha

pter

7

INTRODUCTION

Cancer of the esophagus is the sixth cause of cancer death worldwide. It is a highly lethal disease in

which only one third of patients have a resectable tumor at the time of presentation. Surgery with

neoadjuvant chemo(radio)therapy offers the best chances for long-term survival.1 However, still half

of patients suffer from recurrence of disease after multimodality therapy.2-4 Clearly, new treatments are

needed to improve the outcome of patients with esophageal cancer.

The Human Epidermal growth factor Receptor 2 (HER2) regulates cell growth and proliferation.

Overexpression of HER2 has been identified in a wide variety of carcinomas, including esophageal

cancer. HER2 overexpression correlates with advanced tumor stage and poor survival.5-9 Trastuzumab,

a monoclonal antibody against the HER2 receptor, inhibits the receptor activity and oncogenic

downstream targets. 10 In HER2-positive breast cancer, trastuzumab improves overall and disease-free

survival in early and metastatic disease. 11 Overexpression of HER2 is not uncommon in esophageal

cancer with up to 43% of the cancers being HER2 positive.10, 12-17 Hence, trastuzumab could well be an

effective drug in the multimodal treatment of esophageal cancer.18 In metastatic gastric cancer, the

addition of Trastuzumab to standard palliative chemotherapy has proved to be of benefit. 19

Trastuzumab is only effective in cancer cells overexpressing HER2 and/or cells that show HER2 gene

amplification 6 The HER2 status is evaluated by immunohistochemical staining (IHC) and HER2 gene

amplification by In Situ Hybridization (ISH).20 Abundant tumor tissue is available for analysis from

the resection specimen. However, preoperative biopsies from esophageal cancer are limited in size.

Moreover, expression of oncogenes and tumor suppressor genes in adenocarcinomas of the upper

gastrointestinal tract are well known to be heterogeneous which questions the representativeness of

a small tissue fragment taken at endoscopy.21 This might lead to under-treatment of patients in whom

HER2 overexpression/amplification was falsely scored negative. On the other hand, false positive

biopsies may lead to an increase of toxicity and costs, and moreover, inadequate prognostication with

a potential negative impact on clinical decision making.

The aims of the present study were to determine if the HER2 status of an esophageal adenocarcinoma

can be reliably determined based on endoscopic biopsies, and to assess the inter-observer agreement

in determining HER status.

MATERIALS AND METHODS

Patients and samples

Patients who had undergone an esophagectomy for adenocarcinoma of the esophagus or OGJ

between January 2000 to January 2010 in the Erasmus MC, Rotterdam, the Netherlands were

identified from the institutional database. Seventy-five patients with a cT2-cT3 tumor were selected.

Excluded were cT1 and cT4 tumors and patients who underwent any form of neoadjuvant or adjuvant

chemo(radio)therapy. cT1 tumors were excluded because of the limited additional effect of neo-

| Chapter 7

100

adjuvant therapy targeted at HER2 that can be expected after resection in this group of tumors which

already has a relatively high survival rate after resection. cT4 tumors were excluded because in our

center, cT4 tumors are not resected but are treated with definitive chemoradiotherapy, after which no

resection specimen is available to compare to the pre-treatment biopsies.

Also patients with incomplete or inadequate archived tumor tissue (either in biopsies or the surgical

resection specimen) were excluded. Patients were selected consecutively whenever possible.

Tissue blocks of the endoscopic biopsies, routinely taken from multiple areas of the tumor, as well as

the resection specimen for each patient were retrieved. All tissue samples had been fixed using 4%

buffered formalin and embedded in paraffin after dehydration. All biopsies and a representative area

of invasive tumor of the resection specimen that was identified using a Haematoxylin and Eosin (H&E)

stain. Subsequently, 4-μm sections were cut from the tissue blocks and were used for IHC and ISH.

Several patient and tumor characteristics were identified including patients’ age (at the time of

diagnosis), sex and tumor stage according to the AICC TNM 6th edition (as determined by endo-

ultrasonography and CT-scanning of the chest and abdomen, or from pathology report when clinical

investigations were not conclusive).22 The study was reviewed and approved by the local medical

ethics committee and the requirement of obtaining patients’ informed consent was waived.

HER2 immunohistochemistry (IHC) and dual-colour In-situ Hybridization (ISH)

The 4-μm sections were mounted on starfrost slides (Knittel Gläser, Braunschweig, Germany) and

dried. The immunohistochemical staining was performed using a BenchMark ULTRA automated

slide staining instrument (Ventana Medical Systems, Tuscon, AZ, USA) and using the PATHWAY HER2

4B5 antibody, according to the manufacturer’s instructions. ISH staining was also performed fully

automated with the BenchMark ULTRA, with the INFORM HER2 DNA probe and the Red ISH V probe

(both Ventana Medical Systems).

Evaluation of HER2 status

HER2 expression and amplification was independently assessed by three experienced pathologists

(KB, JEB and JR) from three institutions (Erasmus MC, University Medical Centre Rotterdam, Rotterdam,

The Netherlands; Isala Clinics Zwolle, The Netherlands, and TARGOS Molecular Pathology, Kassel,

Germany). The pathologists were blinded for the patients. Biopsies and resection specimens were

rotating amongst the centers in three batches making sure that matching biopsies and resection

specimens were never present at the same location simultaneously. The H&E staining, IHC and ISH

results of the tissues from a single patient’s biopsy or resection specimens were at the disposal of the

pathologist.


101

Cha

pter

7

Positive

2+ 1+ 3+

Negative

Dc-SISH

HER2:CEP17 <2.2 HER2:CEP17 >2.2

IHC staining

FIGURE 1. Algorithm for the assessment of HER2 status IHC: ImmunohistochemistryIntensity of reactivity: 0-1+ = absent - faint, 2+ = moderate, 3+ = strongISH: dual color-silver in situ hybridizationHER2:CEP17 : HER2 and centromeric probe 17 ratio

For determining the HER2 status an algorithm was used (Fig 1.). In case of no or weak HER2 expression

(1+) the tumor is scored as HER2 negative, and in case of strong expression (3+), the tumor is scored

as HER2 positive. In case of moderate expression (2+), the results of ISH are evaluated, and the HER2

status is determined based on the absence (negative) or presence (positive) of HER2 amplification

(Fig 2.).

Evaluation of the IHC for HER2 was done according to the scoring system for gastric and OGJ cancer.23 In

brief: HER2 expression in the resection specimen was scored as 2+ when ≥10% of tumor cells showed

a weak to moderate complete or basolateral membranous reactivity. For preoperative biopsies, the

presence of cohesive IHC 3+ and/or ISH+ clones resulted in a positive HER2 status irrespective of

size.23 For ISH evaluation, the HER2: CEP17 (centromeric probe 17) ratio was determined. A sample was

considered positive for HER2 amplification when this ratio was >2.2.


HER2 status was defined as positive or negative.23 The statistical analysis was performed using

Statistical Package for Social Sciences version 17.0 (SPSS Inc., Chicago, IL, USA).

| Chapter 7

102

RESULTS

Patients

A total of 75 patients were selected who met the inclusion criteria. The male: female ratio was 3:1 and

the mean age was 63 years. Forty-eight patients (64%) had a cT3 tumor and 45 (60%) patients were

staged as node positive. (Table 1)

TABLE 1. Patient and tumor characteristics of 75 patients with esophageal or esophagogastric junction adenocarcinoma

Characteristics

Gender (n, %)malefemale

5619

(75%)(25%)

Age [years] mean (SD) 63 (11%)

Tumor differentiation (n, %)goodmoderatepoormissing

32942 1

( 4%)(39%)(56%)( 1%)

Clinical T-stage (n, %)†cT2cT3no endoscopic pass

2048 7

(27%)(64%)( 9%)

Clinical N-stage (n, %)‡N0N1missing

2345 7

(31%)(60%)( 9%)

SD: Standard deviation† Clinical T-stage: T-stage as defined by endo-ultrasonography and/or CT-scanning according to UICC TNM Cancer Staging, 6th edition. ‡Clinical N-stage: clinical N-stage as defined by endo-ultrasonography and/or CT-scanning and/or FDG-PET according to UICC TNM Cancer Staging, 6th edition.

Biopsies

A mean of 5 biopsies (standard deviation ± 2) per patient were collected. The biopsies of four (5%)

patients could not be evaluated for HER2 expression by IHC. One biopsy was identified as squamous

cell carcinoma and in the three remaining biopsies the IHC failed. In 16/71 (23%) patients, expression

of HER2 by IHC was 3+, in 3/71 (4%) patients this was 2+. Two of these biopsies showed HER

amplification. Hence, 18/71 (25%) biopsies were HER2 positive. (Table 2)


103

Cha

pter

7

Resection specimens

Six (8%) patients were excluded: there was one squamous cell carcinoma, two samples contained no

representative tumor cells and in three IHC staining failed. Thirteen patients (19%) were 3+, whilst 3/69

(4%) were 2+. Two of these showed HER2 gene amplification, resulting in a total of 15/69 (22%) HER2

positive resection specimens. In the third 2+ specimen the ISH failed, resulting in one specimen which

could not be classified as positive or negative after using the clinical algorithm. (Table 2)

TABLE 2. Conventional immunohistochemistry scoring and results with the application of a clinical algorithm† of 71 evaluable biopsies and 69 evaluable resection specimens of esophageal adenocarcinoma

IHC After following algorithm

0-1+ 2+ 3+ Neg Pos NE

Biopsies (n=71) 52 (73%) 3 (4%) 16 (23%) 53 (75%) 18 (25%) 0 (0%)

Resection specimens (n=69) 53 (78%) 3 (4%) 13 (19%) 53 (77%) 15 (22%) 1 (1%)

IHC: ImmunohistochemistryNeg: negativePos: positiveNE: not evaluableIntensity of reactivity: 0-1+: absent - faint, 2+: moderate, 3+: strong† Clinical algorithm:IHC 0-1+: HER2 negative; IHC 2+ à HER2/centromere17 ratio on ISH <2.2: HER2 negative; >2.2: HER2 positive IHC3+: HER2 positive

Concordance, sensitivity and specificity

Concordance of HER2 positivity between the biopsies and matching resection specimens was found

in 14/18 (78%) patients. In 3 (16%) patients the matching resection specimens were HER2 negative,

and in one patient the resection specimen could not be evaluated due to technical failure.

For the positive resection specimens, 14/15 (93%) biopsies were concordant with the positive

status. One of the 53 (2%) HER2 negative biopsies was collected from a tumor that showed HER2

overexpression in the resection specimen. (Table 3) Sensitivity for determining HER2 status of a tumor

in an endoscopic biopsy was 93%, specificity was 94%. The positive predictive value of HER2 testing in

biopsies was 82%, the negative predictive value was 98%.

Inter-observer agreement

There was agreement amongst the pathologists for the assessment of HER2 expression in 83% of

biopsies and in 85% of the resection specimens. Agreement on HER2 gene amplification was found in

55% of the biopsies and 47% of the resection specimens.

| Chapter 7

104

TABLE 3. Concordance in HER2 status (based on clinical algorithm† for HER2 status) in 75 matched pairs of biopsies and resection specimen of esophageal adenocarcinoma

Resection specimen

negative positive NE total

Biop

sy

negative 47 1 5 53

positive 3 14 1 18

NE 3 0 1 4

total 53 15 7 75

NE: not evaluable

† Clinical algorithm:IHC 0-1+: HER2 negative; IHC 2+ à HER2/centromeric probe 17 ratio by ISH <2.2: HER2 negative; >2.2: HER2 positive IHC3+: HER2 positive

FIGURE 2. HE (a1, b1), IHC (a2, b2) and ISH (a3, b3) in two different invasive esophageal adenocarcinomas. In (a), the adenocarcinoma is HER2 negative in IHC, and correspondingly there is no amplification of HER2 gene in ISH. In contrast, adenocarcinoma in example two shows moderate HER2 expression (2+) and the corresponding ISH revealsHER2 amplification in some, but not all tumor cells. Magnification a1, b1, a2: 1:200. b2: 1:400, a3 and b3: 1:600


105

Cha

pter

7

DISCUSSION

The treatment of esophageal cancer with monoclonal antibodies targeted against the HER2 receptor

is only indicated when HER2 is overexpressed.6 Hence, careful selection of HER2-positive tumors is

essential to prevent unnecessary (neo-) adjuvant treatment of patients with HER2-negative tumors,

who are not likely to benefit. This study found that 22% of the esophageal adenocarcinomas are HER2

positive. Most importantly, we showed that HER2 expression can be evaluated in small diagnostic

tissue fragments taken at the time of upper GI endoscopy and HER2 status in these biopsies matches

the HER2 status of the resection specimen available after surgery in 94% of the cases.

The findings of this study are of importance for clinical practice. In many patients only a formalin-

fixed, paraffin-embedded endoscopic biopsy is available in which this HER2 status can be determined.

Hence, the indication for cancer treatment with Trastuzumab relies on accurate determination of HER

2 expression in these small tissue fragments. Nowadays, neoadjuvant therapy is given to most patients

with esophageal cancer which induces complete tumor regression in 25% of patients and no vital

tumor cells are present for HER2 evaluation. Also genetic modifications may take place during chemo-

and radiotherapy which can make determination of the HER2 status unreliable or impossible. 24

It has been questioned that small pretreatment biopsies are not representative for the genetic profile

of the primary tumor. Great heterogeneity was observed in HER2 expression throughout esophageal

adenocarcinomas.25 Several other studies were performed using tissue arrays for determining

HER2 status, risking false-positive or false-negative results due to sampling error in heterogeneous

tumors.26,27 By evaluating all biopsies taken during the first endoscopy from multiple areas of the

tumor and representative sections of the resection specimens, we feel that this possible sampling

error was reduced to a minimum in the present study. However, since we still encountered three false

positive biopsies, sampling error might still be an issue, even when using multiple biopsies.

Problems were encountered with the quality of the ISH. Because IHC already provided sufficient

information to determine HER2 status according to the algorithm, it was only necessary to repeat ISH

in two patients that had been scored as HER 2+ by IHC. Also the use of archival material has not led to

a deviating proportion of HER positive tumors. In fact, 22% positivity for HER2 is well within the range

of other studies.12,13,16,17

A concordance rate of 89-98% for HER2 expression between core biopsies and excision specimens

has been demonstrated for breast cancer.28-30 This is comparable to the 94% concordance rate in the

present study. However, it is also known that after neoadjuvant treatment the HER2-status of a tumor

can differ from the pretreatment status, which might make it more difficult to show concordance in

patients who were neoadjuvantly treated.12

When considering both neoadjuvant and adjuvant treatment, the concordance in HER2-status

between the primary tumor and metastases (both regional and distant) is of importance. A study by

Reichelt et al. reported 100% concordance in HER2 status between the primary tumor and lymph-

node metastases as well as distant metastatic deposits of esophageal cancer when analysed using

a FISH technique.26 A more recent study by Bozetti et al. also found high levels of concordance

| Chapter 7

106

between primary gastric tumors and metastasis.31 Fassan et al found a concordance between primary

esophagogastric junction tumors and lymph-node metastases of more than 98%, using a CISH

technique.32 These results suggest that both neoadjuvant and adjuvant treatment targeted at HER2

in patients with tumors of the upper gastro-intestinal tract will affect the primary tumor as well as

possible metastases in the majority of cases.

The IHC scoring algorithm complicates a comparison in HER2 expression between biopsies and

resection specimens. Discrepant HER2 expression between biopsies and resection specimens could

be explained by different cut-off criteria for HER2 positivity or be true differences based on biology.

However, the different cut-off criteria would only lead to a false positive result in biopsies (due to the

10% rule in resection specimens) and this also only holds true for IHC 1+ or IHC 0 resection specimens

with a focus of less than 10% of strong membrane staining. These cases are rare.

In the present study, the interobserver agreement was moderate with respect to the assessment of

ISH. There are several possible explanations for this finding. First, experience in evaluating ISH results

differed between the observers. Secondly, the equipment used for the analyses was likely not the

same in the three centers influencing the quality of the staining. However, the disagreement amongst

the pathologists is likely reduced to a minimum when using the algorithm, not affecting the finally

determined HER2 status of a specimen.

The developments in HER2 targeted therapy for cancers of the upper gastro-intestinal tract raise

questions about the accuracy of determining the HER2 status of the tumor in small tissue fragments.

The present study shows that this can be reliably performed by following an algorithm, when using

multiple endoscopically obtained pretreatment biopsies.


107

Cha

pter

7

REFERENCES1. van Hagen P, Hulshof MC, van Lanschot JJ, Steyerberg EW, van Berge Henegouwen MI, Wijnhoven BP, et al.

Preoperative chemoradiotherapy for esophageal or junctional cancer. N Engl J Med. 2012 May 31;366(22):2074-84.

2. Burmeister BH, Smithers BM, Gebski V, Fitzgerald L, Simes RJ, Devitt P, et al. Surgery alone versus chemoradiotherapy followed by surgery for resectable cancer of the oesophagus: a randomised controlled phase III trial. Lancet Oncol. 2005 Sep;6(9):659-68.

3. Tepper J, Krasna MJ, Niedzwiecki D, Hollis D, Reed CE, Goldberg R, et al. Phase III trial of trimodality therapy with cisplatin, fluorouracil, radiotherapy, and surgery compared with surgery alone for esophageal cancer: CALGB 9781. J Clin Oncol. 2008 Mar 1;26(7):1086-92.

4. Urba SG, Orringer MB, Turrisi A, Iannettoni M, Forastiere A, Strawderman M. Randomized trial of preoperative chemoradiation versus surgery alone in patients with locoregional esophageal carcinoma. J Clin Oncol. 2001 Jan 15;19(2):305-13.

5. Dreilich M, Wanders A, Brattstrom D, Bergstrom S, Hesselius P, Wagenius G, et al. HER-2 overexpression (3+) in patients with squamous cell esophageal carcinoma correlates with poorer survival. Dis Esophagus. 2006;19(4):224-31.

6. Hudis CA. Trastuzumab--mechanism of action and use in clinical practice. N Engl J Med. 2007 Jul 5;357(1):39-51.

7. Polkowski W, van Sandick JW, Offerhaus GJ, ten Kate FJ, Mulder J, Obertop H, et al. Prognostic value of Lauren classification and c-erbB-2 oncogene overexpression in adenocarcinoma of the esophagus and gastroesophageal junction. Ann Surg Oncol. 1999 Apr-May;6(3):290-7.

8. Sato-Kuwabara Y, Neves JI, Fregnani JH, Sallum RA, Soares FA. Evaluation of gene amplification and protein expression of HER-2/neu in esophageal squamous cell carcinoma using Fluorescence in situ Hybridization (FISH) and immunohistochemistry. BMC Cancer. 2009;9:6.

9. Tanner M, Hollmen M, Junttila TT, Kapanen AI, Tommola S, Soini Y, et al. Amplification of HER-2 in gastric carcinoma: association with Topoisomerase IIalpha gene amplification, intestinal type, poor prognosis and sensitivity to trastuzumab. Ann Oncol. 2005 Feb;16(2):273-8.

10. al-Kasspooles M, Moore JH, Orringer MB, Beer DG. Amplification and over-expression of the EGFR and erbB-2 genes in human esophageal adenocarcinomas. Int J Cancer. 1993 May 8;54(2):213-9.

11. Slamon DJ, Godolphin W, Jones LA, Holt JA, Wong SG, Keith DE, et al. Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science. 1989 May 12;244(4905):707-12.

12. Schoppmann SF, Jesch B, Zacherl J, Wrba F, Hejna M, Maresch J, et al. HER-2 status in primary oesophageal cancer, lymph nodes and distant metastases. Br J Surg. 2011 May 25.

13. Radu OM, Foxwell T, Cieply K, Navina S, Dacic S, Nason KS, et al. HER2 amplification in gastroesophageal adenocarcinoma: correlation of two antibodies using gastric cancer scoring criteria, H score, and digital image analysis with fluorescence in situ hybridization. Am J Clin Pathol. 2012 Apr;137(4):583-94.

14. Jankowski J, Coghill G, Hopwood D, Wormsley KG. Oncogenes and onco-suppressor gene in adenocarcinoma of the oesophagus. Gut. 1992 Aug;33(8):1033-8.

15. Hardwick RH, Shepherd NA, Moorghen M, Newcomb PV, Alderson D. c-erbB-2 overexpression in the dysplasia/carcinoma sequence of Barrett’s oesophagus. J Clin Pathol. 1995 Feb;48(2):129-32.

16. Schoppmann SF, Jesch B, Friedrich J, Wrba F, Schultheis A, Pluschnig U, et al. Expression of Her-2 in carcinomas of the esophagus. Am J Surg Pathol. 2010 Dec;34(12):1868-73.

17. Yoon HH, Shi Q, Sukov WR, Wiktor AE, Khan M, Sattler CA, et al. Association of HER2/ErbB2 expression and gene amplification with pathologic features and prognosis in esophageal adenocarcinomas. Clin Cancer Res. 2012 Jan 15;18(2):546-54.

| Chapter 7

108

18. Safran H, Dipetrillo T, Akerman P, Ng T, Evans D, Steinhoff M, et al. Phase I/II study of trastuzumab, paclitaxel, cisplatin and radiation for locally advanced, HER2 overexpressing, esophageal adenocarcinoma. Int J Radiat Oncol Biol Phys. 2007 Feb 1;67(2):405-9.

19. Bang YJ, Van Cutsem E, Feyereislova A, Chung HC, Shen L, Sawaki A, et al. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010 Aug 28;376(9742):687-97.

20. Schneider S, Uchida K, Brabender J, Baldus SE, Yochim J, Danenberg KD, et al. Downregulation of TS, DPD, ERCC1, GST-Pi, EGFR, and HER2 gene expression after neoadjuvant three-modality treatment in patients with esophageal cancer. J Am Coll Surg. 2005 Mar;200(3):336-44.

21. Grabsch H, Sivakumar S, Gray S, Gabbert HE, Muller W. HER2 expression in gastric cancer: Rare, heterogeneous and of no prognostic value - conclusions from 924 cases of two independent series. Cell Oncol. 2010;32(1-2):57-65.

22. Sobin LH, Wittekind C, International Union against Cancer. TNM : classification of malignant tumours. 6th ed. New York: Wiley-Liss; 2002.

23. Hofmann M, Stoss O, Shi D, Buttner R, van de Vijver M, Kim W, et al. Assessment of a HER2 scoring system for gastric cancer: results from a validation study. Histopathology. 2008 Jun;52(7):797-805.

24. Adams AL, Eltoum I, Krontiras H, Wang W, Chhieng DC. The effect of neoadjuvant chemotherapy on histologic grade, hormone receptor status, and HER2/neu status in breast carcinoma. Breast J. 2008 Mar-Apr;14(2):141-6.

25. Boers JE, Meeuwissen H, Methorst N. HER2 status in gastro-oesophageal adenocarcinomas assessed by two rabbit monoclonal antibodies (SP3 and 4B5) and two in situ hybridization methods (FISH and SISH). Histopathology. 2011 Feb;58(3):383-94.

26. Reichelt U, Duesedau P, Tsourlakis M, Quaas A, Link BC, Schurr PG, et al. Frequent homogeneous HER-2 amplification in primary and metastatic adenocarcinoma of the esophagus. Mod Pathol. 2007 Jan;20(1):120-9.

27. Thompson SK, Sullivan TR, Davies R, Ruszkiewicz AR. HER-2/neu Gene Amplification in Esophageal Adenocarcinoma and Its Influence on Survival. Ann Surg Oncol. 2011 Jan 26.

28. Shousha S, Peston D, Amo-Takyi B, Morgan M, Jasani B. Evaluation of automated silver-enhanced in situ hybridization (SISH) for detection of HER2 gene amplification in breast carcinoma excision and core biopsy specimens. Histopathology. 2009 Jan;54(2):248-53.

29. Sutela A, Vanninen R, Sudah M, Berg M, Kiviniemi V, Rummukainen J, et al. Surgical specimen can be replaced by core samples in assessment of ER, PR and HER-2 for invasive breast cancer. Acta Oncol. 2008;47(1):38-46.

30. Apple SK, Lowe AC, Rao PN, Shintaku IP, Moatamed NA. Comparison of fluorescent in situ hybridization HER-2/neu results on core needle biopsy and excisional biopsy in primary breast cancer. Mod Pathol. 2009 Sep;22(9):1151-9.

31. Bozzetti C, Negri FV, Lagrasta CA, Crafa P, Bassano C, Tamagnini I, et al. Comparison of HER2 status in primary and paired metastatic sites of gastric carcinoma. Br J Cancer. 2011 Apr 26;104(9):1372-6.

32. Fassan M, Ludwig K, Pizzi M, Castoro C, Guzzardo V, Balistreri M, et al. Human epithelial growth factor receptor 2 (HER2) status in primary and metastatic esophagogastric junction adenocarcinomas. Hum Pathol. 2012 Aug;43(8):1206-12.

PART IIPostoperative outcome and prognostication

P. van Hagen

M. van Heijl


R. Boellaard

P.M.M. Bossuyt

F.J.W. ten Kate

H. van Dekken

O.S. Hoekstra

G.W. Sloof

J.J.B. van Lanschot

Diseases of the Esophagus (accepted for publication)

Chapter 8Prediction of disease-free survival using

relative change in FDG-uptake early during neoadjuvant chemoradiotherapy for potentially

curable esophageal cancer: A prospective cohort study

| Chapter 8

112

ABSTRACT

BACKGROUND18F-Fluorodeoxyglucose positron emission tomography (FDG-PET) has been investigated as a tool for

monitoring response to neoadjuvant chemo- and chemoradiotherapy (CT and CRT respectively) and

as a predictor for survival in patients with esophageal cancer. In contrast to patients who undergo

neoadjuvant CT, it is not known whether patients who are clinically identified as responders after

neoadjuvant CRT show better disease-free survival (DFS) than patients identified as nonresponders.

The aim of the study was to determine the predictive value of FDG-uptake measured prior to and early

during neoadjuvant CRT.

METHODS

Patients treated with neoadjuvant CRT between 2004 and 2009 within a randomized trial were

included. FDG-uptake was measured at baseline and after 14 days of CRT. According to the PERCIST-

criteria, patients were allocated to have metabolic response, stable disease or progression. Patients

were followed until recurrence of disease or death. The predictive value of FDG-PET was determined

with uni- and multivariable analysis in patients who underwent potentially curative surgery.

RESULTS

One-hundred and six patients were included in the analysis. Minimal follow-up for surviving patients

was 60 months. No significant differences in DFS were found between patients with metabolic

response, stable disease or progression, with five-year DFS rates of 66%, 53% and 67% respectively

(p=0.39).

CONCLUSION

Relative change in FDG uptake after 14 days of CRT is not associated with disease-free survival in

patients with esophageal cancer undergoing neoadjuvant chemoradiotherapy followed by surgery.

These measurements should not be used for prognostication in this specific group of patients.

Prediction of disease-free survival using relative change in FDG-uptake early during neoadjuvant chemoradiotherapy |

113

Cha

pter

8

INTRODUCTION

In the Western world, the overall incidence of esophageal cancer is rising.1 Despite many improvements

in diagnosis and treatment, five-year survival rates rarely exceed 40% even after careful preoperative

selection.2,3

Evidence in favor of neoadjuvant chemoradiotherapy (CRT) followed by surgery versus surgery alone

is growing steadily, in terms of improved long-term survival.4,5 It is however well established that only a

proportion of patients treated with neoadjuvant CRT shows a (substantial) histopathological response

in the resection specimen.5-9 The non-responding patients do not benefit from neoadjuvant CRT, but

are still exposed to its toxicity.

If it would be possible to identify non-responders early during neoadjuvant CRT, these patients

would discontinue this therapy and undergo surgery directly. This strategy was designed and already

successfully executed by the Munich group, in patients undergoing FDG-PET monitoring during

neoadjuvant chemotherapy for esophagogastric cancer in the MUNICON-1 and -2 trial.13-14

18F-Fluorodeoxyglucose positron emission tomography (FDG-PET) is a non-invasive imaging

technique which enables quantification of tumor activity based on tissue glucose uptake, which is

generally increased in malignant tissue.10-12 FDG-PET has been investigated as a tool to monitor (early)

response to neoadjuvant therapy and as a predictor for survival.13-16

Definitive evidence for the usefulness of FDG-PET in distinguishing between responding and non-

responding patients would be a difference in (disease-free) long-term survival between patients

identified by FDG-PET as responders compared to non-responders.

The aim of the present study was, therefore, to evaluate the predictive value of pretreatment FDG-

uptake and relative change in FDG-uptake measured early during neoadjuvant chemoradiotherapy.


Study design

In September 2004, a multicenter randomized phase-III trial was initiated in the Netherlands,

comparing neoadjuvant CRT followed by surgery with surgery alone in patients with potentially

curable esophageal cancer.5 Details of the study design and the final results have been reported

previously.4,5,17,18

The present report concerns an ancillary study within that randomized trial, evaluating the value of

FDG-PET scanning before and two weeks after the start of neoadjuvant CRT in the multimodality

treatment arm.19

Patients

Inclusion criteria for patients were previously described.5 Patients with histologically proven carcinoma

of the esophagus or esophago-gastric junction underwent extensive preoperative staging. Only

| Chapter 8

114

patients with cT1N1 and cT2-3N0-1 tumors were eligible. Eligible patients who had provided written

informed consent were randomly allocated to neoadjuvant CRT followed by surgery or to surgery

alone. Patients allocated to the neoadjuvant arm were subsequently invited to undergo the ancillary

FDG-PET protocol.

Neoadjuvant concurrent chemoradiotherapy

Over a five week period, concurrent chemoradiotherapy was applied on an outpatient basis. Paclitaxel

(50 mg/m2) and Carboplatin (area-under-curve=2) were administered by i.v. infusion once weekly

during the radiation period. External beam radiation with a total dose of 41.4 Gy was given in 23

fractions of 1.8 Gy, five fractions a week.

Surgery

Either transhiatal or transthoracic esophagectomy was performed, preferably within 6 weeks after

completion of the CRT. Only centers performing at least ten esophagectomies per year were allowed

to participate in this trial.20

Evaluation of tumor response by FDG-PET

A tumor-focused static FDG-PET scan was performed after injection of 5 MBq/kg 18F-FDG at baseline

and at day 14 after the start of CRT. Acquisition was started 90 minutes after injection. FDG-PET

scans were performed in five centers, using standardized acquisition and image reconstruction, with

centralized data-analysis (VUmc PET data-analysis center) and after interinstitutional calibration of

scanners.21,22 For every patient, the two scans were performed in the same institution on the same

scanner.

For quantitative analysis of FDG uptake, the maximum standardized uptake value (SUV) was used

according to previously validated methodology and corrected for body weight and serum glucose

(SUV-bw-g).22-27

Histopathological response

The disease stage after CRT was evaluated based on the histology in the surgical resection specimen.

Tumor regression grade (TRG) was classified according to the modified Mandard-score which is

based on the percentage of viable residual tumor cells in relation to the area of fibrosis/necrosis.28

The classification was performed by one of two experienced gastrointestinal pathologists (FJWtK and

HvD).


115

Cha

pter

8

Follow-up

After completion of the treatment protocol, patients were seen every three months for the first year,

every six months for the second year, and then each year until five years after treatment, to document

recurrent disease.

Analysis

Disease-free survival was used as primary outcome, counted from the start of CRT up to the time of a

first relapse or death, when no relapse occurred. Only patients who completed the PET protocol and

underwent an operation were analyzed. Follow-up was performed to the time of first relapse, the last

hospital visit or until death. For patients with tumors found to be irresectable during surgery, the date

of operation was used as cut-off for disease-free survival. For patients who showed progression of

disease prior to surgery, censoring was performed on the date on which progression was discovered.

Disease-free and overall survival curves were constructed by the Kaplan–Meier method, and the

log-rank test was used to determine significance (P<0.05 [two-sided] was considered statistically

significant). For the relative change in SUV, patients were allocated to three groups: patients with

metabolic response, those with metabolically stable disease, and those with metabolic progression of

disease. Metabolic response was defined as a relative decrease of SUV of 30% or more, or a reduction

of absolute SUV of >1 unit. Metabolic progression was defined as a relative increase of SUV of 30%

or more, or an increase of absolute SUV with >1 unit. All other patients were defined as having

metabolically stable disease.29

Univariable Cox regression analysis for disease-free survival was used to determine the predictive

significance of pretreatment SUV measurements and of relative change in SUV as continuous

variables. Relative change in SUV (%) was defined as: (100 * (SUV 14 days – SUV baseline)/SUV baseline).

Multivariable Cox regression analyses were used to determine the predictive significance of FDG-PET,

corrected for possible confounding factors.

Furthermore, a time dependent receiver operating characteristic (ROCt) analysis was performed to

estimate the discriminating value of the relative SUV change for cumulative disease free survival

at 5 years. The ROCt-curve and the area under the curve (AUC) were computed according to the

method by Heagerthy and colleagues, and the accompanying R package (The R-project for statistical

computing, http://www.r-project.org/ 10 October 2014, date last accessed).30, 31

RESULTS

Of the 180 patients that were included in the multimodality arm of the randomized trial (parent

study), 145 patients could be included in the current study. One hundred and forty patients

underwent the first FDG-PET, due to withdrawal of two patients and disease progression in 3 patients.

After the first FDG-PET, another patient withdrew from the study. Of the remaining 139 patients, 33

http://www.r-project.org/

| Chapter 8

116

had to be excluded from further analysis for the following reasons: no FDG uptake on the first PET

(n=6), progression during chemoradiotherapy (n=4), no fasting before FDG-PET (n=3), second FDG-

PET was performed on a different scanner (n=5), nausea during PET (n=1), no FDG available (n=1),

PET technically not analyzable (n=3), no second FDG-PET performed due to logistic reasons (n=10).

Baseline characteristics of the 106 analyzed patients were comparable to the characteristics of all

patients included in the CRT-arm of the randomized CROSS trial (data not shown).

TABLE 1. Patient, PET and tumor characteristics for 106 included patients.

All patientsn = 106

Female / male (%) 26 / 80 (25% / 76%)

Age (y) median [range] 59 [37 to 76]

Tumor type: Adenocarcinoma / Squamous cell carcinoma (%) 87 / 19 (82% / 18%)

Pretreatment SUV (median) [IQR] 10.8 [7.2 to 17.2]

SUV 14 days after start of chemoradiotherapy (median) [IQR] 8.2 [5.4 to 11.0]

%SUV change 14 days after start of chemoradiotherapy (median) [IQR] -26.9 [-47.3 to 0.73]

ypT-stage (n, %)

T0 (complete regression) 34 (34%)

T1 16 (15%)

T2 17 (16%)

T3 33 (31%)

Tx (not resected) 6 ( 6%)

pN-stage (n, %)

N0 72 (68%)

N1 27 (26%)

Nx (not resected or no LN in resection specimen) 7 ( 7%)

Histopathological response* 64 (60%)

Resectability (n, %)

R0 94 (89%)

R1 6 ( 6%)

Irresectable 6 ( 6%)

Median follow-up time in months [IQR]# 76.6 [68.56 to 91.31]

Patients with recurrent / residual disease (n, %) 62 (58%)

SUV: standardized uptake value, IQR: interquartile rangepT stage: pathological T-stage (UICC TNM Cancer Staging, 6th edition)pN stage: pathological N-stage (UICC TNM Cancer Staging, 6th edition)R0: no viable tumor cells within 1mm of all resection marginsR1: viable tumor cells within 1 mm of one of the resection margins *Histopathological response: grade of 1 or 2 on the modified Mandard score # after exclusion of deceased patients


117

Cha

pter

8

Eighty men and 26 women with a median age of 59 years (range 37 to 76) completed the PET

protocol and were available for further analysis. Eighty-two percent had an adenocarcinoma and 18%

a squamous cell carcinoma. (Table 1) Disease progression occurred in 5 patients, detected either

during neoadjuvant CRT or during surgery, and one patient died before surgical resection due to a

bleeding from the tumor. Pathological assessment of the resection specimens revealed complete

regression due to CRT in 34 patients, a ypT1 tumor in 16 patients, ypT2 in 17 patients and ypT3 in

33 patients. A microscopically radical resection (no viable tumor cells within 1mm of the resection

margins, R0) was achieved in 94/100 (94%) patients.

One patient was lost to follow-up before the minimal time to follow-up was reached. For all other

patients, the minimal time of follow-up for surviving patients was 60 months, median follow-up for

surviving patients was 76.6 months (interquartile range 68.56-91.31 months). During follow-up, 62

patients deceased, of whom 5 did not have recurrence of disease and died from other causes. In 62

patients (58%), recurrent disease (locoregional, metastatic or both) was found during follow-up.

The log-rank test showed no significant difference in the five year disease-free survival between

patients with a pretreatment SUV above the median of 10.8 and those with a SUV below the median:

51% versus 62% respectively (p=0.32). (Figure 1)

No. at risk

SUV > median 53 40 32 27 25 22

SUV < median 53 40 35 31 28 28

FIGURE 1. Kaplan-Meier curves for estimated five year disease-free survival in 106 patients. (According to pretreatment SUV-uptake above (green line) or below (blue line) the median uptake. Error bars represent 95% confidence intervals)

Metabolic response was found in 49 patients. In 48 patients, metabolically stable disease was found,

while nine patients showed metabolic progression.

The estimated disease-free survival after five years was comparable between patients with

metabolically progressive disease and patients with metabolic response (67% vs. 53% respectively),

| Chapter 8

118

who both showed better survival than patients with metabolically stable disease (40%). These

differences in five year DFS between the groups were not statistically significant (P=0.39 by log rank

test).(Figure 2)

Metabolic response 49 38 34 30 27 26Metabolically stable disease 48 34 27 22 20 18Metabolic progression 9 8 6 6 6 6

FIGURE 2. Kaplan-Meier curves for estimated five year disease-free survival in 106 patients.Metabolic response: a relative SUV decrease of 30% of more (and an absolute reduction of SUV of >1 unit). Metabolically stable disease: SUV decrease of less than 30%, or SUV increase of less than 30%. Metabolic progression was defined as more than 30% increase in SUV, (and an absolute increase of SUV with >1 unit). Dotted lines represent the 95% confidence intervals.

FIGURE 3. Time dependent receiver operating characteristic curve for the prediction of five year disease free survival using relative change in SUV 14 days after the start of chemoradiotherapy in 106 patients. [AUC = 0.58 (95% CI: 0.44 to 0.67)]


119

Cha

pter

8

When compared with patients with metabolic progression, 5-year DFS in univariate analysis was

lower in patients with metabolic response and those with metabolically stable disease (HRs 2.02

and 2.72 respectively), however these differences were not statistically significant (p-values 0.25 and

0.10 respectively). When corrected for possible confounding baseline covariates, HRs for metabolic

response and metabolically stable disease were 3.89 and 3.70 respectively, both not statistically

significant (p-values 0.07 and 0.09 respectively).

Subgroup analysis for patients having adenocarcinoma or squamous cell carcinoma, and for patients

who underwent a microscopically radical resection showed no significant correlations between SUV-

max and DFS (data not shown).

When using a cut off of 0% (increased or decreased SUV after 14 days), no significant differences were

found in five year DFS between patients with an increase in SUV and patients with a decrease in SUV

(data not shown).

The time dependent ROC-curve that was computed is shown in Figure 3. The corresponding

estimated AUC at 5 years was 0.57 (95% CI: 0.444 to 0.671).

DISCUSSION

The data analyzed in this prospective trial show that the relative change in FDG-uptake 14 days after

start of neoadjuvant chemoradiotherapy or pretreatment PET measurements is not a significant

predictor of disease-free survival in patients with esophageal cancer undergoing potentially curative

esophagectomy.

Most studies describing FDG-PET as a tool for assessment of response to neoadjuvant therapy use

histopathological response as primary outcome. This is probably because, compared to survival, this

is a (dichotomous) outcome for which one can more easily determine optimal cut-off values. A more

relevant and convincing demonstration for the usefulness of FDG-PET in distinguishing responding

from non-responding patients is whether the patients identified by PET as responders also show

better disease-free survival compared to patients identified by PET as non-responders.

In a previous paper from our group, it was found that FDG-PET response had a positive predictive

value of 76% for histopathological response, while the negative predictive value was found to be

too low (75%) to justify the use of FDG-PET response for early discontinuation of neoadjuvant CRT. 17 As histopathological response determined after neoadjuvant CRT has an association with survival

in several studies,2,24,28,29 it might be expected that there is a (significant) difference in disease-free

survival between patients with a metabolic response on FDG-PET compared to patients in whom

metabolically stable disease or metabolic progression was observed.

FDG-PET has proven to be relatively successful in differentiating between histopathologic responders

and non-responders early during neoadjuvant chemotherapy without radiation.13,34,35 According to

these studies, patients identified by FDG-PET as responder had a significantly better survival compared

to patients identified as non-responder. Previous studies show limited evidence for a similar potential

| Chapter 8

120

in patients treated with concurrent neoadjuvant chemoradiotherapy, but specific data on the relation

between response according to early FDG PET monitoring and long-term survival after combination

therapy are scarcely available.15,36,37 Two of these studies suggested that FDG-PET could actually be

used for early response assessment.15,37

In accordance with our previously presented data, Gillham et al. also concluded that FDG-PET was

not able to predict histopathological response, based on an analysis of 32 patients with esophageal

cancer treated with neoadjuvant chemoradiotherapy.21,36

The most plausible explanation for the limited value of FDG-PET for early response assessment in

patients treated with neoadjuvant chemoradiotherapy is probably the local (inflammatory) response

to radiation, which will cause increased glucose uptake and hence interfere with the FDG-uptake

measurement associated with the treatment effect.38 The fact that nearly all studies investigating early

response to chemotherapy (without concurrent radiotherapy) show more promising results supports

this hypothesis. 13,34,35

Also, the observed metabolic progression in patients in the current study might be caused by an

increase in SUV due to inflammation rather than by tumor progression, which would explain why

this group had survival comparable with the responders, and why time-dependent ROC analysis only

showed a moderate AUC of 0.58.

A recent systematic review by Schollaert et al. shows that metabolic response on FDG-PET after

chemoradiotherapy is significantly associated with better disease-free survival (HR 0.47, 95%CI 0.38-

0.57).39 However, the heterogeneity in timing of the second PET-scan is great, with several studies

performing the second FDG-PET scan even 5 or 6 weeks after the completion of chemoradiotherapy.

Unfortunately, no subgroup analysis was performed for the studies in which the second FDG-PET scan

was performed during CRT.

The sample-size for this ancillary study was based on the sensitivity and specificity of FDG-PET in

predicting histopathologic response on CRT. Disease-free survival was a secondary endpoint, which

leaves room for a type-I error. This could be an explanation for the fairly wide confidence intervals

that accompany the HR’s for DFS according to the PERCIST criteria. This problem might have been

overcome with a larger sample-size.

Whereas measuring relative change in FDG-uptake early during neoadjuvant therapy might have

direct therapeutic consequences, the role of using pretreatment FDG-PET measurements is limited

to the purpose of prognostication which has been described extensively in previous papers.40 Most

of these studies show pretreatment FDG-uptake to be a significant predictor of survival in univariable

analysis. Most impressive are the results of the relatively large but retrospective study by Cerfolio

et al.(N = 89), in which a high four year survival rate was described (89%) for patients with a low

pretreatment SUV value.41 This was significantly higher compared to the survival rate in patients with

a high pretreatment SUV value (31%). In the present analysis the 5-year disease-free survival rate for

patients with a low versus a high pretreatment SUV value was 40% and 56% respectively. The reason

why the results of Cerfolio et al. could not be duplicated in the present study remains unclear.


121

Cha

pter

8

In conclusion, the change in FDG-uptake as measured by FDG-PET before and 14 days after the start of

neoadjuvant chemoradiotherapy was found not to be associated with disease-free survival in patients

with potentially curable esophageal cancer. Therefore, the present study indicates, in accordance with

our previous findings, that FDG-PET should not be applied for early response assessment in patients

with potentially curable esophageal cancer who undergo neoadjuvant chemoradiotherapy followed

by surgery.

| Chapter 8

122

REFERENCES1. Kamangar F, Dores GM, Anderson WF. Patterns of cancer incidence, mortality, and prevalence across five

continents: defining priorities to reduce cancer disparities in different geographic regions of the world. J Clin Oncol. 2006 May 10;24(14):2137-50.

2. Berger AC, Farma J, Scott WJ, et al. Complete response to neoadjuvant chemoradiotherapy in esophageal carcinoma is associated with significantly improved survival. J Clin Oncol. 2005 Jul 1;23(19):4330-7.

3. Omloo JM, Lagarde SM, Hulscher JB, et al. Extended transthoracic resection compared with limited transhiatal resection for adenocarcinoma of the mid/distal esophagus: five-year survival of a randomized clinical trial. Ann Surg. 2007 Dec;246(6):992-1000; discussion -1.

4. Shapiro J, van Lanschot JJ, Hulshof MC et al. Neoadjuvant chemoradiotherapy plus surgery versus surgery alone for oesophageal or junctional cancer (CROSS): long-term results of a randomised controlled trial. Lancet Oncol. 2015 Sep, 16(9):1090-1098

5. van Hagen P, Hulshof MC, van Lanschot JJ, et al. Preoperative chemoradiotherapy for esophageal or junctional cancer. N Engl J Med. 2012 May 31;366(22):2074-84.

6. Bosset JF, Gignoux M, Triboulet JP, et al. Chemoradiotherapy followed by surgery compared with surgery alone in squamous-cell cancer of the esophagus. N Engl J Med. 1997 Jul 17;337(3):161-7.

7. Burmeister BH, Smithers BM, Gebski V, et al. Surgery alone versus chemoradiotherapy followed by surgery for resectable cancer of the oesophagus: a randomised controlled phase III trial. Lancet Oncol. 2005 Sep;6(9):659-68.

8. Urba SG, Orringer MB, Turrisi A, Iannettoni M, Forastiere A, Strawderman M. Randomized trial of preoperative chemoradiation versus surgery alone in patients with locoregional esophageal carcinoma. J Clin Oncol. 2001 Jan 15;19(2):305-13.

9. Walsh TN, Noonan N, Hollywood D, Kelly A, Keeling N, Hennessy TP. A comparison of multimodal therapy and surgery for esophageal adenocarcinoma. N Engl J Med. 1996 Aug 15;335(7):462-7.

10. Czernin J, Phelps ME. Positron emission tomography scanning: current and future applications. Annu Rev Med. 2002;53:89-112.

11. Gambhir SS, Czernin J, Schwimmer J, Silverman DH, Coleman RE, Phelps ME. A tabulated summary of the FDG PET literature. J Nucl Med. 2001 May;42(5 Suppl):1S-93S.

12. Rohren EM, Turkington TG, Coleman RE. Clinical applications of PET in oncology. Radiology. 2004 May;231(2):305-32.

13. Lordick F, Ott K, Krause BJ, et al. PET to assess early metabolic response and to guide treatment of adenocarcinoma of the oesophagogastric junction: the MUNICON phase II trial. Lancet Oncol. 2007 Sep;8(9):797-805.

14. zum Büschenfelde CM, Herrmann K, Schuster T, et al. (18)F-FDG PET-guided salvage neoadjuvant radiochemotherapy of adenocarcinoma of the esophagogastric junction: the MUNICON II trial. J Nucl Med. 2011;52(8):1189-96

15. Westerterp M, Omloo JM, Sloof GW, et al. Monitoring of response to pre-operative chemoradiation in combination with hyperthermia in oesophageal cancer by FDG-PET. Int J Hyperthermia. 2006 Mar;22(2):149-60.

16. Westerterp M, van Westreenen HL, Reitsma JB, et al. Esophageal cancer: CT, endoscopic US, and FDG PET for assessment of response to neoadjuvant therapy--systematic review. Radiology. 2005 Sep;236(3):841-51.

17. van Heijl M, Omloo JM, van Berge Henegouwen MI, et al. Fluorodeoxyglucose positron emission tomography for evaluating early response during neoadjuvant chemoradiotherapy in patients with potentially curable esophageal cancer. Ann Surg. 2011 Jan;253(1):56-63.


123

Cha

pter

8

18. van Heijl M, van Lanschot JJ, Koppert LB, et al. Neoadjuvant chemoradiation followed by surgery versus surgery alone for patients with adenocarcinoma or squamous cell carcinoma of the esophagus (CROSS). BMC Surg. 2008;8:21.

19. van Heijl M, Omloo JM, van Berge Henegouwen MI, et al. NEOadjuvant therapy monitoring with PET and CT in Esophageal Cancer (NEOPEC-trial). BMC Med Phys. 2008;8:3.

20. van Lanschot JJ, Hulscher JB, Buskens CJ, Tilanus HW, ten Kate FJ, Obertop H. Hospital volume and hospital mortality for esophagectomy. Cancer. 2001 Apr 15;91(8):1574-8.

21. Boellaard R, Oyen WJ, Hoekstra CJ, et al. The Netherlands protocol for standardisation and quantification of FDG whole body PET studies in multi-centre trials. Eur J Nucl Med Mol Imaging. 2008 Dec;35(12):2320-33.

22. Young H, Baum R, Cremerius U, et al. Measurement of clinical and subclinical tumour response using [18F]-fluorodeoxyglucose and positron emission tomography: review and 1999 EORTC recommendations. European Organization for Research and Treatment of Cancer (EORTC) PET Study Group. Eur J Cancer. 1999 Dec;35(13):1773-82.

23. Hoekstra CJ, Hoekstra OS, Stroobants SG, et al. Methods to monitor response to chemotherapy in non-small cell lung cancer with 18F-FDG PET. J Nucl Med. 2002 Oct;43(10):1304-9.

24. Hoekstra CJ, Paglianiti I, Hoekstra OS, et al. Monitoring response to therapy in cancer using [18F]-2-fluoro-2-deoxy-D-glucose and positron emission tomography: an overview of different analytical methods. Eur J Nucl Med. 2000 Jun;27(6):731-43.

25. Krak NC, Hoekstra OS, Lammertsma AA. Measuring response to chemotherapy in locally advanced breast cancer: methodological considerations. Eur J Nucl Med Mol Imaging. 2004 Jun;31 Suppl 1:S103-11.

26. Kroep JR, Van Groeningen CJ, Cuesta MA, et al. Positron emission tomography using 2-deoxy-2-[18F]-fluoro-D-glucose for response monitoring in locally advanced gastroesophageal cancer; a comparison of different analytical methods. Mol Imaging Biol. 2003 Sep-Oct;5(5):337-46.

27. Westerterp M, Pruim J, Oyen W, et al. Quantification of FDG PET studies using standardised uptake values in multi-centre trials: effects of image reconstruction, resolution and ROI definition parameters. Eur J Nucl Med Mol Imaging. 2007 Mar;34(3):392-404.

28. Mandard AM, Dalibard F, Mandard JC, et al. Pathologic assessment of tumor regression after preoperative chemoradiotherapy of esophageal carcinoma. Clinicopathologic correlations. Cancer. 1994 Jun 1;73(11):2680-6.

29. Wahl RL, Jacene H, Kasamon Y, Lodge MA. From RECIST to PERCIST: Evolving Considerations for PET response criteria in solid tumors. J Nucl Med. 2009 May;50 Suppl 1:122S-50S.

30. Heagerty PJ, Lumley T, Pepe MS. Time-dependent ROC curves for censored survival data and a diagnostic marker. Biometrics. 2000 Jun;56(2):337-44.

31. Heagerty PJ, Zheng Y. Survival model predictive accuracy and ROC curves. Biometrics. 2005 Mar;61(1):92-105.

32. Brucher BL, Becker K, Lordick F, et al. The clinical impact of histopathologic response assessment by residual tumor cell quantification in esophageal squamous cell carcinomas. Cancer. 2006 May 15;106(10):2119-27.

33. Schneider PM, Baldus SE, Metzger R, et al. Histomorphologic tumor regression and lymph node metastases determine prognosis following neoadjuvant radiochemotherapy for esophageal cancer: implications for response classification. Ann Surg. 2005 Nov;242(5):684-92.

34. Ott K, Weber WA, Lordick F, et al. Metabolic imaging predicts response, survival, and recurrence in adenocarcinomas of the esophagogastric junction. J Clin Oncol. 2006 Oct 10;24(29):4692-8.

| Chapter 8

124

35. Weber WA, Ott K, Becker K, et al. Prediction of response to preoperative chemotherapy in adenocarcinomas of the esophagogastric junction by metabolic imaging. J Clin Oncol. 2001 Jun 15;19(12):3058-65.

36. Gillham CM, Lucey JA, Keogan M, et al. (18)FDG uptake during induction chemoradiation for oesophageal cancer fails to predict histomorphological tumour response. Br J Cancer. 2006 Nov 6;95(9):1174-9.

37. Wieder HA, Brucher BL, Zimmermann F, et al. Time course of tumor metabolic activity during chemoradiotherapy of esophageal squamous cell carcinoma and response to treatment. J Clin Oncol. 2004 Mar 1;22(5):900-8.

38. Hautzel H, Muller-Gartner HW. Early changes in fluorine-18-FDG uptake during radiotherapy. J Nucl Med. 1997 Sep;38(9):1384-6.

39. Schollaert P, Crott R, Bertrand C, D’Hondt L,Borght TV, Krug B. A systematic review of the predictive value of (18)FDG-PET in esophageal and esophagogastric junction cancer after neoadjuvant chemoradiation on the survival outcome stratification. J Gastrointest Surg 2014 May;18(5):894-905.

40. Omloo JM, van Heijl M, Hoekstra OS, van Berge Henegouwen MI, van Lanschot JJ, Sloof GW. FDG-PET parameters as prognostic factor in esophageal cancer patients: a review. Ann Surg Oncol. 2011 Nov;18(12):3338-52.

41. Cerfolio RJ, Bryant AS. Maximum standardized uptake values on positron emission tomography of esophageal cancer predicts stage, tumor biology, and survival. Ann Thorac Surg. 2006 Aug;82(2):391-4; discussion 4-5.

V. Oppedijk

A. van der Gaast

J.J.B. van Lanschot

P. van Hagen

R. van Os

C.M. van Rij

M.J. van der Sangen

J.C. Beukema

H. Rütten

P.H. Spruit

J.G. Reinders

D.J. Richel


M.C.C.M. Hulshof

Journal of Clinical Oncology 2014 Feb 10;32(5):385-91.

Chapter 9Patterns of recurrence after surgery alone

versus preoperative chemoradiotherapy and surgery in the CROSS trials

| Chapter 9

128

ABSTRACT

PURPOSE

To analyze recurrence patterns in patients with cancer of the esophagus or gastroesophageal junction

treated with either preoperative chemoradiotherapy (CRT) plus surgery or surgery alone.


Recurrence pattern was analyzed in patients from the previously published CROSS I and II trials in

relation to radiation target volumes. CRT consisted of five weekly courses of paclitaxel and carboplatin

combined with a concurrent radiation dose of 41.4 Gy in 1.8-Gy fractions to the tumor and pathologic

lymph nodes with margin.

RESULTS

Of the 422 patients included from 2001 to 2008, 418 were available for analysis. Histology was mostly

adenocarcinoma (75%). Of the 374 patients who underwent resection, 86% were allocated to surgery

and 92% to CRT plus surgery. On January 1, 2011, after a minimum follow-up of 24 months (median,

45 months), the overall recurrence rate in the surgery arm was 58% versus 35% in the CRT plus

surgery arm. Preoperative CRT reduced locoregional recurrence (LRR) from 34% to 14% (P < .001)

and peritoneal carcinomatosis from 14% to 4% (P < .001). There was a small but significant effect on

hematogenous dissemination in favor of the CRT group (35% v 29%; P = .025). LRR occurred in 5%

within the target volume, in 2% in the margins, and in 6% outside the radiation target volume. In 1%,

the exact site in relation to the target volume was unclear. Only 1% had an isolated infield recurrence

after CRT plus surgery.

CONCLUSION

Preoperative CRT in patients with esophageal cancer reduced LRR and peritoneal carcinomatosis.

Recurrence within the radiation target volume occurred in only 5%, mostly combined with outfield

failures.

Patterns of recurrence in the CROSS trials |

129

Cha

pter

9

INTRODUCTION

Patients with esophageal cancer have poor prognosis; at the time of diagnosis; ≥ 50% present with

distant metastasis or irresectable disease.1 For potentially curable patients, for decades, surgical

resection had been the main treatment. However, incomplete resections occurred in up to 25%2

and locoregional recurrence (LRR) in 30% to 40%, with 5-year survival rarely exceeding 25%.3 Most

randomized controlled trials (RCTs) investigating the role of preoperative chemoradiotherapy versus

surgery alone failed to show a significant survival benefit, mostly because of a lack of statistical power.

However, a recent meta-analysis showed a survival benefit for patients treated with preoperative

chemoradiotherapy (CRT) or chemotherapy compared with surgery alone.4

The results of the CROSS (Chemoradiotherapy for Oesophageal Cancer Followed by Surgery Study)

trial have recently been published. This was an RCT comparing preoperative CRT followed by surgery

with surgery alone.5 CRT consisted of 41.4 Gy in 1.8-Gy fractions combined with weekly concurrent

carboplatin and paclitaxel. After a minimum follow-up of 24 months, there was a significant estimated

5-year overall survival benefit of 13% in favor of the CRT plus surgery arm. The CRT regimen was well

tolerated, with little added toxicity.5

Patterns of recurrence of esophageal cancer after surgery compared with CRT plus surgery are

infrequently reported in the literature. Meguid et al6 and Denham et al7 describe relapse patterns after

CRT plus surgery, and CRT plus surgery and definitive CRT, respectively; however, those patient groups

were not compared with surgery alone. Understanding relapse patterns provides insight into the

effectiveness of the combined treatment and may lead to improvements. Therefore, we analyzed the

recurrence pattern of patients treated in the CROSS trial and the preceding phase II trial investigating

the same preoperative regimen.8 In particular, we related the site of recurrence to the radiation fields

employed.


Patient Population

The patient population consisted of patients enrolled onto the CROSS trial, an RCT in which eligible

patients were randomly assigned between CRT plus surgery and surgery alone,5 and patients included

in the preceding phase II trial8 investigating the same preoperative regimen followed by surgery.

All patients had histologically proven and resectable squamous cell carcinoma (SCC) or

adenocarcinoma (AC) of the esophagus, stage cT1N1M0 or cT2-3N0-1M0 according to the Union

International Contre Cancer (sixth edition, 2002). The upper border of the tumor had to be ≥ 3 cm

below the upper esophageal sphincter. Those with tumors of the gastroesophageal junction were

also eligible, provided that the primary tumor did not extend ≥ 4 cm into the stomach.

Patients had to be age 18 to 75 years with a WHO performance score ≤ 2. Weight loss had to be ≤

10%. No past or current history of malignancy other than the entry diagnosis was allowed, except for

nonmelanomatous skin cancer, curatively treated carcinoma in situ of the cervix, or a nonrecurred

http://jco.ascopubs.org/content/32/5/385.long#ref-1











| Chapter 9

130

malignancy treated ≥ 5 years before enrollment. No previous radiotherapy or chemotherapy was

allowed. Written informed consent was required from all patients before random assignment. The

medical ethics committees of all eight participating centers approved the study protocol.

Staging

Pretreatment staging included elaborate history taking, physical examination, routine blood workup

and pulmonary function tests, an upper GI endoscopy, endoultrasonography, and computed

tomography (CT) of neck, chest, and upper abdomen. On indication, ultrasound of the neck was

performed with fine-needle aspiration.

Chemotherapy

Chemotherapy consisted of five cycles of concurrent paclitaxel 50 mg/m2 and carboplatin targeted at

area under the curve of 2, starting on days 1, 8, 15, 22, and 29. Toxicity of CRT was closely monitored

using the National Cancer Institute Common Terminology Criteria for Adverse Events (version 3.0).9

Radiotherapy

A total radiation dose of 41.4 Gy was administered in 23 fractions of 1.8 Gy, five fractions per week,

starting on the first day of chemotherapy. All patients were treated with external-beam radiation

using a three-dimensional conformal radiation technique. Gross tumor volume was drawn on each

relevant slice of the planning CT and was defined by the primary tumor and any enlarged regional

lymph nodes. The planning target volume (PTV) provided a proximal and distal margin of 4 cm and

a radial margin of 1.5 cm around the gross tumor volume. A distal margin of 3 cm was chosen in

case the tumor extended into the gastric cardia. Individually shaped beams were used in each field

by either cerrobend blocks or multileaf collimators to ensure optimal sparing of normal tissue. The

daily prescription dose of 1.8 Gy was specified at the International Commission on Radiation Units

and Measurement 50/62 reference point, and the 95% isodose had to encompass the entire PTV.

The maximum dose to the PTV was not to exceed the prescription dose by > 7%. Tissue density

inhomogeneity correction was used.

Surgery

Patients randomly assigned to the surgery arm were treated as soon as possible after random

assignment. Patients in the CRT plus surgery arm preferably underwent surgery at 6 weeks after

completion of CRT; surgery consisted of a transthoracic approach with a two-field lymph node

dissection or transhiatal approach, depending on tumor localization, patient characteristics, and

local expertise. A wide local excision of the N1 lymph nodes, including standard excision of the celiac

nodes, was carried out in both techniques. Continuity of the digestive tract was restored by gastric

tube reconstruction or colonic interposition procedure with cervical anastomosis.



131

Cha

pter

9

Pathologic Analysis

For grading of the therapy response, the degree of histomorphologic regression was classified

into four modified categories, as first described by Mandard et al.10 All resection margins, including

circumferential margins, were evaluated for vital tumor, with a cutoff point of 1 mm. If vital tumor was

present at ≤ 1 mm from a resection margin, that margin was considered to be positive.

Follow-Up

In the first year after completion of the protocol, patients were seen every 3 months. In the second

year, follow-up took place every 6 months and, thereafter, yearly until 5 years after treatment. If

applicable, late toxic effects and recurrence of disease or death were documented. During follow-up,

additional diagnostics were only performed on indication.

Recurrences

Relapses were classified as locoregional or distant. LRRs were defined as recurrences at the site of the

primary tumor or locoregional lymph nodes. Lymph node recurrences at the celiac trunk or in the

supraclavicular region were also considered to be locoregional. Distant recurrences were defined as

nonregional lymph node recurrences, systemic metastases, malignant pleural effusions, or peritoneal

metastases. Most patients suspected of experiencing recurrence underwent a CT scan of thorax and

abdomen or an endoscopy. If necessary, cytology or histology was obtained. If a second recurrence

was detected within 4 weeks after the first occurrence, it was considered to be synchronous.

Localization and date of identification of all locoregional and distant recurrences were scored.

Radiation Target Volumes

In patients with recurrent disease who were treated with CRT plus surgery, radiation target volumes

were analyzed in relation to the site of recurrence. Treatment failures were classified as infield when

relapse occurred within the PTV, outfield when relapse occurred outside the PTV, and borderline

when adjacent to the PTV or field edge. We compared the exact site of recurrence with the treatment

volume on the planning CT scan. When a recurrence was detected endoscopically, the location was

compared with the results of the staging endoscopy. In case of a relapse at the anastomotic site,

endoscopy results, histology reports of the esophageal resection specimen, and planning CT scans

were used to reconstruct the proximal and distal ends of the resection specimen in relation to the

irradiated volume.

Statistical Analysis

Duration of follow-up was defined as the interval between the day of random assignment and death or

the last date of hospital visit or telephone call. The Kaplan-Meier method was used to calculate survival

probabilities. The influence of prognostic factors was analyzed using univariable and multivariable


| Chapter 9

132

Cox regression analyses. The backward-step method was used to optimize the multivariable model.

A univariable Cox regression model was also used to analyze the difference per treatment arm for

each separate location of recurrence. We used one-way analysis of variance test to investigate the

differences between both treatment arms. Analyses were performed using SPSS software (version

18.0; SPSS, Chicago, IL) and the R statistical program (http://www.r-project.org).

CROSS phase II trial: Patients receiving CRT+S

(n=54)

CROSS phase III trial: Patients with esophageal or

GEJ cancer randomly assigned (n=368)

Died after first cycle of CRT (n=1)

Refused surgery (n=1)

Allocated to CRT+S (n=180)

Allocated to S alone (n=188)

Withdrew consent (n=1) Did not receive CRT (n=7)

Underwent resection after completion of CRT

(n=52)

Received CRT (n=171) Underwent surgery (n=168)

Underwent resection (n=161)

Underwent surgery (n=186) Underwent resection

(n=161)

Included in the analysis (n=52)




FIGURE 1. CONSORT diagram. CROSS, Chemoradiotherapy for Oesophageal Cancer Followed by Surgery Study; CRT, chemoradiotherapy; GEJ, gastroesophageal junction; S, surgery.

http://www.r-project.org


133

Cha

pter

9

RESULTS

Patients

A total of 422 patients were included in both trials (Fig 1). Of the 368 patients in the phase III CROSS

trial, two patients were ineligible: one because of withdrawal, and one because of distant metastases

at the time of diagnosis. Of the remaining 366 patients, 188 were randomly assigned to the surgery

arm and 178 to the CRT plus surgery arm. Of the 54 patients included in the phase II trial, 52 completed

the protocol, one patient died after the first course of chemotherapy (probably because of cardiac

arrest), and one patient refused surgery after CRT. The 52 patients who underwent resection were

included in the analysis of the CRT plus surgery arm. Finally, 418 patients were available for analysis.

In the surgery arm, 161 (85.6%) of 188 patients underwent an esophageal resection versus 213 (92.2%)

of 230 in the CRT plus surgery group. A microscopically radical (R0) resection was achieved in 68% of

patients in the surgery arm and in 93% of patients in the CRT plus surgery arm. In the CRT plus surgery

arm, 28% had a pathologic complete response (ypT0N0). One or more pathologically positive lymph

nodes were found in 74% of patients in the surgery arm and in 31% of those in the CRT plus surgery

arm (P < .001).

Patterns of Recurrence

After a minimum follow-up of 24 months and a median survival of 45 months for surviving patients,

57.1% of the resected patients in the surgery group had recurrent disease versus 34.7% in the CRT plus

surgery group. Most patients had distant failure (22%) or combined locoregional and distant failure

(16.5%). Only 9.3% of patients in the surgery arm had an isolated LRR without distant metastasis versus

3.3% in the CRT plus surgery arm. Also, 24.2% versus 10.8% of patients in the surgery and CRT plus

surgery arms, respectively, had concurrent locoregional and distant relapses, and 23.6% versus 20.7%

of patients had distant relapse only in the surgery and CRT plus surgery arms, respectively. The majority

of LRRs occurred within 2 years of follow-up. In the CRT plus surgery arm, no LRRs were observed after

30 months. Figures 2A, 2B, and 2C show the differences between both arms for disease-free survival

(DFS), locoregional DFS, and distant metastasis–free survival, respectively.

Site of Recurrence

Further analysis showed that recurrences at the anastomosis occurred in 8.7% versus 2.8% (P = .008)

of patients in the surgery and CRT plus surgery arms, respectively (Table 2). LRRs at the anastomosis

occurred more often after R1 resections (11%) than after R0 resections (4%) and more often in

patients with pN1 disease (7%) than in those with pN0 disease (3%). Mediastinal relapses occurred

in 20.5% versus 7.0% (P < .001) of patients in the surgery and CRT plus surgery arms, respectively.

Peritoneal carcinomatosis occurred in 13.7% versus 4.2% (P < .001) of patients and hematogenous

metastasis occurred in 35.4% versus 28.6% (P = .025) of patients in the surgery and CRT plus surgery

arms, respectively. There were no significant differences between both arms in recurrence rates at

http://jco.ascopubs.org/content/32/5/385.long#F1




http://jco.ascopubs.org/content/32/5/385.long#T2

| Chapter 9

134

the supraclavicular or celiac axis levels (Table 2). Generally, these latter areas were not included in the

radiation target volume.

TABLE 1. Patient demographic and clinical characteristics at diagnosis

Characteristic S Arm (n = 161) CRT + S Arm (n = 213) P*

No. % No. %

Age, years 0.54

Median 60 60

Range 36-73 37-79

Male sex 129 80 169 81 0.85

T stage

T1 1 1 1 0 0.81

T2 35 22 31 15 0.12

T3 122 76 180 85 0.15

Unknown 1 1 1 0 0.40

Nodal stage

N0 50 31 80 38 0.21

N1 106 66 125 61 0.96

Unknown 3 3 3 1 0.96

Nodal status

Positive supraclavicular nodes 0 0 0 0 NA

Positive celiac nodes 6 4 8 4 0.63

Tumor length, cm 0.62

Median 5 5

Range 1-13 1-12

Histology

Adenocarcinoma 122 76 160 75 0.87

SCC 38 24 52 24 0.88

Other 1 1 1 1 0.84

Abbreviations: CRT, chemoradiotherapy; NA, not applicable; S, surgery; SCC, squamous cell carcinoma.* Analysis of variance test.



135

Cha

pter

9

A B

C

FIGURE 2. (A) Disease-free survival for patients undergoing surgery alone (S) or chemoradiotherapy (CRT) followed by S (CRT + S; hazard ratio [HR], 0.47; 95% CI, 0.35 to 0.64). (B) Locoregional recurrence–free survival; recurrences at anastomotic site, mediastinum, celiac trunk, or supraclavicular lymph nodes (HR, 0.37; 95% CI, 0.23 to 0.59). (C) Distant metastasis–free survival; systemic metastases including nodal metastases other than regional, peritoneal deposits, and malignant pleural effusion (HR, 0.52; 95% CI, 0.38 to 0.73).

| Chapter 9

136

TABLE 2. Results of Univariable Cox Regression Analysis of RFS Time per Treatment Arm in Patients Undergoing Resection (n = 374)

Site of Recurrence S Arm (n = 161) CRT+ S Arm (n=213)HR 95% CI P

No. % No. %

Anastomosis 14 8.7 6 2.8 0.28 0.11 to 0.72 0.008

Mediastinum 33 20.5 15 7.0 0.29 0.16 to 0.53 <0.001

Supraclavicular 7 4.3 9 4.2 0.83 0.31 to 2.2 0.71

Celiac axis 11 6.9 8 3.8 0.42 0.17 to 1.04 0.06

Para-aortic 17 10.6 14 6.6 0.53 0.26 to 1.1 0.08

Peritoneal carcinomatosis

22 13.7 9 4.2 0.27 0.12 to 0.58 0.01

Hematogenous 57 35.4 61 28.6 0.67 0.46 to 0.96 0.03

NOTE. Bold font indicates significance.Abbreviations: CRT, chemoradiotherapy; HR, hazard ratio; RFS, recurrence-free survival; S, surgery.

Site of Recurrence in Relation to the Radiation Target VolumeIn the 74 patients with recurrences after CRT plus surgery, the precise localization of relapse was

determined and correlated to the irradiated field volume (Table 3). Infield recurrences occurred in

11 (5.2%) of 213 patients, of whom only two patients experienced an infield recurrence without

synchronous distant failure. Recurrences at the borders of the treatment volume occurred in five

(2.3%) of 213 patients; three of these occurred at the site of the celiac axis. In two of the borderline

recurrences, the site of relapse was in the anterior-posterior beams but not in the lateral beams.

Regional outfield recurrences occurred in 13 (6.1%) of 213 patients; two of these were solitary LRRs.

Two patients were scored as unknown; one had a relapse at the site of the anastomosis, and for the

other, the diagnostic CT scan of the recurrence could not be retrieved.

TABLE 3. Tumor Recurrences in Relation to Radiation Target Volumes in Patients Undergoing CRT Plus Surgery (n = 213)

Recurrence Infield Outfield Borderline Unknown Total

LRR only 2 2 2 1 7

Distant only 0 43 0 1 44

LRR plus distant 9 11 3 0 23

Total 11 56 5 2 74

Abbreviations: CRT, chemoradiotherapy; LRR, locoregional recurrence.



137

Cha

pter

9

Potential Prognostic Factors for Developing an LRR

Table 4 lists the results of the analyses per treatment arm. Prognostic factors predicting LRRs in

univariable analysis were surgery alone, pathologically positive lymph nodes (pN1), and R1 resection.

In the multivariable analysis, the backward method showed that surgery alone, pathologic nodal

stage N1, and histology of SCC significantly increased the risk of developing an LRR. After multivariable

analysis, surgery alone, pN1, and SCC remained independent prognostic factors.

In the surgery arm, 47% of patients with SCC developed an LRR compared with 30% of patients with

AC. In the CRT plus surgery arm, there was no significant difference between SCC and AC (15% and

14%, respectively).

Of the 59 patients with a pathologic complete response (pCR) after CRT, 17% developed any recurrent

disease; only one patient (1.7%) had a solitary LRR. Of the 154 patients with no pCR, 42% experienced

a recurrence: LRR ± distant recurrence in 17% and a solitary LRR in 4%. After R1 resection, there was no

significant difference in LRRs between treatment arms, although a trend was present (36% v 29% for

surgery and CRT plus surgery, respectively).

TABLE 4. Univariable and multivariable cox regression analyses for LRRs in patients undergoing resection (n = 374)

Factor LRR Incidence Univariable Multivariable

S Arm CRT + S Arm HR 95% CI HR 95% CI

Method of resection (TTE v THE) 20 v 17 6 v 8 0.83 0.54 to 1.29 NA

Tumor length (≤ 5.0 v > 5.0 cm) 23 v 39 16 v 11 0.89 0.54 to 1.46 NA

Clinical T stage (T1-2 v T3-4) 31 v 35 5 v 17 1.32 0.76 to 2.29 NA

Clinical nodal stage (N0 v N1) 31 v 35 10 v 18 1.50 0.93 to 2.41 NA

Pathologic nodal stage (N0 v N1) 22 v 38 10 v 23 3.66 2.20 to 5.85 2.85 1.59 to 5.11

Involved margins (R0 v R1) 34 v 36 13 v 29 2.29 1.38 to 3.76 NA

Histology (SCC v AC) 47 v 30 15 v 14 0.70 0.44 to 1.12 0.49 0.29 to 0.82

Sex (male v female) 33 v 34 12 v 20 1.12 0.67 to 1.87 NA

Treatment arm (S v CRT + S) 27 14 0.37 0.23 to 0.59 0.50 0.29 to 0.86

pCR after CRT (no v yes)* NA 7 v 17 0.36 0.13 to 1.05 NA

NOTE. Bold font indicates significance.Abbreviations: AC, adenocarcinoma; CRT, chemoradiotherapy; HR, hazard ratio; LRR, locoregional recurrence; NA, not applicable; pCR, pathologic complete response; S, surgery; SCC, squamous cell carcinoma; THE, transhiatal esophagectomy; TTE, transthoracic esophagectomy.* Factor only available in the CRT + S arm and therefore not suitable for multivariable analysis.


| Chapter 9

138

DISCUSSION

In the CROSS phase III trial, preoperative CRT followed by surgery compared with surgery alone

improved DFS, with an absolute difference of 22% at 5 years and an improved overall survival of 13%.

Most patients diagnosed with LRRs also developed synchronous distant metastases. Of the patients

undergoing resection, 24% and 11% had concurrent LRRs and distant relapses and only 9.3% and

3.3% had an isolated LRR in the surgery and CRT plus surgery arms, respectively. Few data are available

on relapse patterns after CRT plus surgery for esophageal and gastroesophageal cancers. In most

RCTs comparing CRT plus surgery with surgery alone, the sites of recurrence are either imprecise

or not reported. LRR rates of 13% to 25% and 12% to 42% after CRT plus surgery and surgery alone,

respectively, have been reported.11–13 Most studies have shown a reduction in LRRs after preoperative

CRT.11,13

In our study, patients with a pCR after CRT had a significantly lower LRR rate compared with patients

with a partial (tumor regression grade 2 to 3) or no response after CRT (tumor regression grade 4 to 5).

Of patients with a pCR, 17% had recurrent disease, of whom only one patient had a solitary LRR. The

only patient experiencing LRR after pCR had no lymph nodes examined in the resection specimen

and should probably be considered as having experienced inadequately staged pCR. These data

compare favorably with those of Meguid et al,6 who in their retrospective series reported five solitary

infield recurrences in 82 patients (6%) achieving pCR after CRT.

A marked difference was seen in the occurrence of peritoneal carcinomatosis in favor of the CRT plus

surgery arm (13.7% v 4.2%; P < .001). This might be explained by a reduction of microscopic residual

disease, because of patients achieving pCR, 1.7% had peritoneal metastasis compared with 5.2% with

no pCR. The reduction in recurrences at the site of anastomosis in the CRT arm might be an effect of

a reduction of microscopically positive surgical margins. Unfortunately, in case of R1 resection, the

site of irradicality was not always reported. This is supposedly more likely at the lateral borders of the

specimen than at the cranial or caudal borders. Irradicality and subsequent tumor spill could also be

considered a cause of recurrence at the anastomosis or in the abdominal cavity.

In both the multivariable and univariable analyses, patients with SCC had a higher probability of

developing LRR; however, this was significant only in the multivariable analysis. Patients with SCC are

known to have a higher risk of LRR after surgery alone,14 which is confirmed by the current data. Of

patients with SCC undergoing resection in the surgery arm, 47% experienced LRRs compared with

30% of those with AC. However, because SCC histology has a higher response rate to CRT, this was not

a prognostic factor in the univariable analysis. After CRT, there was no difference between SCC and

AC regarding LRR. Therefore, in the surgery arm, histology was an independent negative prognostic

factor for LRR, which disappeared after preoperative CRT.

Recurrences of only 5% within the radiation target volumes confirms the hypothesis that preoperative

CRT reduces the LRR rate. Recurrences at the supraclavicular fossae, generally not included in the

radiation target volumes of midesophageal and distal tumors, were similar in both groups (4%), which

further confirms this conclusion. Huang et al15 described supraclavicular lymph node recurrences in









139

Cha

pter

9

16.7% of 54 patients with SCC of the proximal esophagus after surgery alone, which included removal

of pathologic supraclavicular lymph nodes. In the small group of patients with proximal tumors, no

supraclavicular recurrences were seen, probably because of the proximity of the supraclavicular fossae

to the radiation treatment volume.

Of the 20 patients with a recurrence near the celiac axis, 18 had a primary tumor located in the distal

esophagus, and most of them had synchronous distant metastases. On the basis of these data, elective

inclusion in the radiation target volume of the supraclavicular fossae for mid or distally located tumors

or celiac nodes for mid or proximal tumors would probably not have a large effect on survival. The idea

behind preoperative CRT in the treatment of esophageal cancer and cancer of the gastroesophageal

junction was to improve survival by reducing locoregional failure. However, we also observed a small

but significant effect on the development of hematogenous metastasis. From the current data, it

cannot be concluded whether this was a systemic effect of the chemotherapy or an indirect effect of

reducing LRRs. However, the short interval and frequently occurring synchronous recurrences argue

in favor of the first hypothesis.

In conclusion, preoperative CRT in patients with esopheageal or junctional cancer improves locoregional

control and has an effect on both hematogenous metastasis and peritoneal carcinomatosis. A pCR

after CRT was a favorable prognostic factor for both locoregional and systemic recurrences.

| Chapter 9

140

REFERENCES1. Enzinger PC, Mayer RJ: Esophageal cancer. N Engl J Med 349:2241-2252, 2003

2. Kelsen DP, Ginsberg R, Pajak TF, et al: Chemotherapy followed by surgery compared with surgery alone for localized esophageal cancer. N Engl J Med 339:1979-1984, 1998

3. Bosset JF, Gignoux M, Triboulet JP, et al: Chemoradiotherapy followed by surgery compared with surgery alone in squamous-cell cancer of the esophagus. N Engl J Med 337:161-167, 1997

4. Sjoquist KM, Burmeister BH, Smithers BM, et al: Survival after neoadjuvant chemotherapy or chemoradiotherapy for resectable oesophageal carcinoma: An updated meta-analysis. Lancet Oncol 12: 681-692, 2011

5. van Hagen P, Hulshof MC, van Lanschot JJ, et al: Preoperative chemoradiotherapy for esophageal or junctional cancer. N Engl J Med 366:2074-2084, 2012

6. Meguid RA, Hooker CM, Taylor JT, et al: Recurrence after neoadjuvant chemoradiation and surgery for esophageal cancer: Does the pattern of recurrence differ for patients with complete response and those with partial or no response? J Thorac Cardiovasc Surg 138:1309-1317, 2009

7. Denham JW, Steigler A, Kilmurray J, et al: Relapse patterns after chemo-radiation for carcinoma of the oesophagus. Clin Oncol (R Coll Radiol) 15:98-108, 2003

8. van Meerten E, van der Gaast A, Tilanus HW, et al: Pathological analysis after neoadjuvant chemoradiotherapy for esophageal carcinoma: The Rotterdam experience. J Surg Oncol 100:32-37, 2009

9. Trotti A, Colevas AD, Setser A, et al: CTCAE v3.0: Development of a comprehensive grading system for the adverse effects of cancer treatment. Semin Radiat Oncol 13:176-181, 2003

10. Mandard AM, Dalibard F, Mandard JC, et al: Pathologic assessment of tumor regression after preoperative chemoradiotherapy of esophageal carcinoma: Clinicopathologic correlations. Cancer 73:2680-2686, 1994

11. Burmeister BH, Smithers BM, Gebski V, et al: Surgery alone versus chemoradiotherapy followed by surgery for resectable cancer of the oesophagus: A randomised controlled phase III trial. Lancet Oncol 6:659-668, 2005

12. Tepper J, Krasna MJ, Niedzwiecki D, et al: Phase III trial of trimodality therapy with cisplatin, fluorouracil, radiotherapy, and surgery compared with surgery alone for esophageal cancer: CALGB 9781. J Clin Oncol 26:1086-1092, 2008

13. Urba SG, Orringer MB, Turrisi A, et al: Randomized trial of preoperative chemoradiation versus surgery alone in patients with locoregional esophageal carcinoma. J Clin Oncol 19:305-313, 2001

14. Siewert JR, Stein HJ, Feith M, et al: Histologic tumor type is an independent prognostic parameter in esophageal cancer: Lessons from more than 1,000 consecutive resections at a single center in the Western world. Ann Surg 234:360-367, 2001

15. Huang W, Li B, Gong H, et al: Pattern of lymph node metastases and its implication in radiotherapeutic clinical target volume in patients with thoracic esophageal squamous cell carcinoma: A report of 1077 cases. Radiother Oncol 95:229-233, 2010

P. van Hagen

B.P.L. Wijnhoven

P. Nafteux

J. Moons

K. Haustermans

G. DeHertogh

J.J.B. van Lanschot

T. Lerut

British Journal of Surgery 2013;100(2):267-73

Chapter 10Recurrence pattern in patients with a pathologically complete response

after neoadjuvant chemoradiotherapy and surgery for esophageal cancer

| Chapter 10

144

ABSTRACT

BACKGROUND

Little is known about recurrence patterns in patients with a pathologically complete response (pCR)

or an incomplete response after neoadjuvant chemoradiotherapy (CRT) followed by resection for

oesophageal cancer. This study was performed to determine the pattern of recurrence in patients

with a pCR after neoadjuvant CRT followed by surgery.

METHODS

All patients who received neoadjuvant CRT followed by esophagectomy between 1993 and 2009 were

identified from a database, and categorized according to pathological tumor response. Recurrences

were classified as locoregional or distant.

RESULTS

One hundred and eighty-eight patients were included. Median potential follow-up was 71·6 months.

A pCR was achieved in 62 (33·0 per cent) of 188 patients. Recurrence developed in 24 (39 per cent) of

62 patients with a pCR and 70 (55·6 per cent) of 126 without a pCR (P = 0·044). Locoregional recurrence

with or without synchronous distant metastases occurred in eight patients (13 per cent) in the pCR

group and 31 (24·6 per cent) in the non-pCR group (P = 0·095). Locoregional recurrences without

synchronous distant metastases occurred four (6 per cent) and ten (7·9 per cent) patients respectively

(P = 0·945). The overall 5-year survival rate was significantly higher in the pCR group than in the non-

pCR group (52 versus 33·9 per cent respectively; P = 0·019).

CONCLUSION

Of patients with a pCR, 13 per cent still developed a locoregional recurrence. Although pCR is more

favorable for survival, it is not synonymous with cure or complete locoregional disease control.

Recurrence pattern in patients with pCR after nCRT and surgery for esophageal cancer |

145

Cha

pter

10

INTRODUCTION

Esophageal cancer is the eighth most common cancer worldwide1, causing more than 400 000

deaths per year.2 Esophagectomy is generally considered the treatment of choice for potentially

curable cancers. Neoadjuvant chemoradiotherapy (CRT) is now used widely because it is believed to

facilitate radical resection, improve locoregional control and prolong survival in patients with locally

advanced oesophageal cancer.3–5 Disease recurrence is still seen, however, in approximately half of

patients within 2–3 years of combined modality treatment.6–8

Up to 57 per cent of patients have no tumor left in the resection specimen after preoperative

chemoradiotherapy.3 They are classified as having obtained a pathologically complete response (pCR;

pathological category after neoadjuvant therapy ypT0 N0) and generally have longer survival than

patients with incomplete or poor responses.3, 9, 10

Understanding the pattern and timing of recurrent disease in patients treated with neoadjuvant

CRT followed by surgery is important. If locoregional recurrence in patients who have a pCR is still

frequently encountered, modifying surgery by extended lymphadenectomy or intensifying the

locoregional effects of CRT by widening irradiation fields or increasing radiotherapy doses might be

worthwhile. If distant metastasis is the most common type of recurrence in patients with a pCR, then

new systemic treatments would seem more logical.

The aim of the present study was to define the pattern and timing of disease recurrence in patients

with a pCR compared with those in patients without a pCR after neoadjuvant CRT followed by surgery

for esophageal cancer.

METHODS

Patients were selected from a prospectively assembled database at the Department of Thoracic

Surgery, University Hospital of Leuven, Belgium. All patients who underwent neoadjuvant CRT

followed by esophagectomy for locally advanced cancer (stages T2–4 N0–1 M0) of the esophagus

or esophagogastric junction between January 1993 and January 2009 were eligible for inclusion. All

patients were followed until death or January 2010.

Patients with disease progression during neoadjuvant treatment, unresectable tumors during

exploratory thoracotomy or macroscopically incomplete resection (R2), those who also received

adjuvant chemotherapy, those who died in hospital after surgery, and those lost to follow-up within

12 months after surgery were excluded.

Patients were eligible for neoadjuvant CRT if they were thought to have lymph node metastases (clinical

cN+) on preoperative computed tomography (CT) with or without positron emission tomography

(PET) and/or endoscopic ultrasonography, and/or had tumors of questionable resectability (cT4). CRT

consisted of 80 mg/m2 cisplatin (Platinol®; Bristol-Meyers Squibb, Braine-l’Alleud, Belgium) on days

1 and 21, and 800 mg/m2 5-fluorouracil (5-FU) (Roche, Brussels, Belgium) by continuous infusion

on days 1–4 and 21–24, with concurrent radiotherapy. Patients were treated in the supine position

http://onlinelibrary.wiley.com/doi/10.1002/bjs.8968/full#bib1








| Chapter 10

146

with an immobilizing mask. All patients underwent CT in the treatment position with intravenous

contrast. The clinical target volume was delineated based on the results of PET–CT and endoscopic

ultrasound examination. The clinical target volume included the primary tumor with a 5-cm margin

in the craniocaudal direction, the suspected lymph nodes and all regional lymph nodes. A margin of

1 cm was added around the planning target volume. Three-dimensional conformal radiotherapy was

used with at least four beams (anteroposterior–posteroanterior (AP-PA) and 2 oblique beams) and

multileaf collimation. To make the dose distribution more uniform, half beams were used in the AP-PA

direction with wedges and energies varying between 6 and 18 MV in the majority of patients.

Patients underwent surgery within 2 months of completion of CRT, via a left-sided thoracophrenotomy11

for infracarinal tumors or as a two-phase procedure, combining a laparotomy with a thoracotomy12,

when the tumor was located above the carina. In both techniques, a two- or three-field

lymphadenectomy was performed, including resection of the thoracic duct whenever possible.

Gastric tube reconstruction with a cervical anastomosis was the preferred method of reconstruction.

Resection specimens were fixed routinely in formalin. Representative tissue blocks were taken from

macroscopically visible residual tumor, all resection margins, lymph nodes and other lesions if present.

When no tumor was observed macroscopically, the entire tumor bed, identified by the presence of

scar tissue, was submitted for microscopic investigation. Tissue blocks were routinely embedded

in paraffin and a single 5-µm thick section of each block was stained with haematoxylin and eosin.

All slides were examined systematically by dedicated gastro-intestinal pathologists. Lymph nodes

identified macroscopically were halved and a single 5-µm thick section of each node was evaluated.

If there were reasons to assume that any vital tumor could be present in the evaluated lymph

node (such as signs of therapy effects), more sections of that lymph node were cut and evaluated.

Immunohistochemical staining was not performed routinely.

For more rigorous verification of a pCR in patients who subsequently developed locoregional

recurrence, additional immunohistological pankeratin staining was performed on freshly cut 5-µm

thick slides of the tumor bed and lymph node blocks of the archival resection specimens (monoclonal

mouse antihuman cytokeratin antibody, clone AE1/AE3; Dako, Heverlee, Belgium). A skin biopsy was

used as an external positive control, and mesenchymal or lymphoid tissue within the specimen as

an internal negative control. Whenever at least one group (lesions measuring 2·0 mm or less) of

pankeratin-positive, atypical epithelial cells was present within the specimen, immunohistochemical

staining was classified as positive.13 In order to reflect standard clinical practice (where pankeratin

staining is not standard), patients with positive pankeratin staining for isolated primary tumor cells

were still classified as having a pCR.

When no tumor cells were seen within 1 mm of the proximal, distal and circumferential resection

margins, the resection was classified as R0. If tumor cells were present within 1 mm of the surgical

resection margin on microscopic evaluation, the resection was classified as R1.

According to the observed tumour regression in the resection specimen, two groups were identified:

patients who had a pCR and those who did not. A pCR was defined as absence of residual tumor cells





147

Cha

pter

10

in the resection specimen, either at the primary tumor site or in any of the resected lymph nodes.

If any vital tumor cells were seen on conventional histopathological examination, the patient was

allocated to the non-pCR group.

Patients were followed for up to 5 years in the outpatient clinic. CT of the neck, chest and abdomen

was carried out routinely every 6 months for at least the first 3 years after surgery. The diagnosis

of recurrence was confirmed by biopsy-positive histopathology or cytopathology. If this was not

possible, clinical deterioration was considered as evidence of recurrence.

In patients with no signs of recurrence, the interval between CT was lengthened to a year in the

fourth and fifth years after operation or until diagnosis of recurrence was established. The last date

of follow-up, date of recurrence (first failure), as well as site of this recurrence (as applicable) were

recorded. Recurrences were defined as locoregional (at the previous site of the primary tumor and/

or in regional lymph nodes within the field of surgical resection and/or the irradiated field), distant

(haematogenous metastases or distant lymph node metastases) or both (synchronous locoregional

and distant recurrence).


The χ2 test was used to detect statistical differences between subgroups. Hazard ratios were calculated

using a Cox regression model. Multivariable analysis for prognostic factors was performed using a Cox

proportional hazard model. Only parameters with P < 0·100 in univariable analysis were added to

the multivariable model. Overall and disease-free survival rates were calculated using the Kaplan-

Meier method, and compared by means of the log rank test. P < 0·050 (two-sided) was considered

statistically significant. Statistical analysis was performed using SPSS® version 17.0 (IBM, Armonk, New

York, USA).

RESULTS

Two hundred and sixty-five patients who were scheduled to receive neoadjuvant CRT followed by

surgery between January 1993 and January 2009 were identified. Of these, 77 were excluded, because

of: progressive disease during neoadjuvant CRT (18), unresectability during exploration (3), adjuvant

chemotherapy (10), R2 resection (35), refusal to undergo surgery (4), considered medically unfit for

surgery (2), or they died after surgery before discharge from hospital (5). Of the five patients who died

after operation, three had a pCR in the resection specimen.

Among the 188 included patients (147 men, 41 women), median age was 60 (range 37-80) years. On

clinical staging, the majority of patients (93·6 per cent) had a tumor invading the adventitia (cT3) or

suspicion of invasion of a surrounding structure (cT4). Most (91·5 per cent) had locoregional lymph

node metastases (cN+). Patient and tumor characteristics are summarized in Table 1.

http://onlinelibrary.wiley.com/doi/10.1002/bjs.8968/full#tbl1

| Chapter 10

148

TABLE 1. Baseline characteristics of 188 patients who underwent neoadjuvant CRT followed by macroscopically radical resection for advanced esophageal cancer.¥

All patients (n=188)

Age [years]medianrange

6037-80

Sex ratio (M:F) 147:41

ASA grade* (n, %)ASA IASA IIASA III

3112928

(17%)(69%)(15%)

Histological type (n, %)SCCACsmall cell carcinoma

107 801

(57%)(43%)( 1%)

cT category^ (n, %)cT2cT3cT4cTx

9112643

( 5%)(60%)(34%)( 2%)

cN grade& (n, %)cN0cN+cNx

15172

1

( 8%)(92%)( 1%)

Tumour location (n, %)proximal thirdmiddle thirddistal thirdGEJ

22638221

(12%)(34%)(44%)(11%)

SCC: Squamous cell carcinomaAC: Adenocarcinoma¥ Percentages may not total 100 because of rounding*ASA grade: American Society of Anesthesiologists Physical Status Classification System^cT grade: clinical T-stage as defined by endo-ultrasonography and/or CT-scanning according to AJCC TNM Cancer Staging, 6th edition.&cN grade: clinical N-stage as defined by endo-ultrasonography and/or CT-scanning and/or FDG-PET-scanning according to AJCC TNM Cancer Staging, 6th edition.

All patients received cisplatin'5-FU-based chemotherapy and concurrent radiotherapy, with a median

dose of 40 (range 36-60) Gy in 20 (range 18-36) fractions. The radiotherapy dose varied during the

study period. In 91 patients (48·4 per cent), a total of 40 Gy was administered in 20 fractions of 2 Gy.

Another 61 patients (32·4 per cent) received a total of 36 Gy in 20 fractions of 1·8 Gy, and 25 (13·3 per

cent) were treated with 45 Gy in 25 fractions of 1·8 Gy. In the remaining 11 patients (5·9 per cent), the

radiotherapy dose varied between 43 and 60 Gy, in fractions of 1·8 Gy.

Ninety-seven patients (51·6 per cent) underwent a left-sided thoracophrenotomy, 89 (47·3 per cent) a

right-sided transthoracic resection and two (1·1 per cent) a left-sided transthoracic resection. A two-


149

Cha

pter

10

field lymphadenectomy (chest and upper abdomen) was performed in 100 patients (53·2 per cent)

and a cervical lymph node dissection was also carried out in 75 (39·9 per cent)15. Thirteen patients

(6·9 per cent) had no formal lymphadenectomy.

A pCR was observed in 62 (33·0 per cent) of 188 patients. A further nine patients had a complete

response at the primary tumor site (ypT0), but were allocated to the non-pCR group because of the

presence of lymph node metastases (ypN1). R0 resection was achieved in 183 patients (97·3 per cent).

The results of histopathological examination are summarized in Table 2.

TABLE 2. Results of conventional histopathological examination of resection specimens

All patients (n=188)

ypT-stage (n, %)ypT0ypTis/ypT1ypT2ypT3ypT4

731930624

(39%)(10%)(16%)(33%)( 2%)

ypN-stage (n, %)ypN0ypN+

11969

(63%)(37%)

No. of LNs removed (n)*

medianp25-p75

2414-32

No. of pos LNs (n)*Ω

medianp25-75

20-2

Radicality (n, %)R0R1

1835

(97%)( 3%)

Tumour response (n, %)pCRnon-pCR

62126

(33%)(67%)

Values in parentheses are percentages unless indicated otherwise;*values are median (interquartile range).ΩNumber of positive lymph nodes in patients with positive lymph nodes in the resection specimen. ypT/N, pathological tumor/node category after neoadjuvant therapy in American Joint Committee on Cancer tumor node metastasis staging system (6th edition)14; Tis, tumor in situ; pCR, pathologically complete response.

Median potential follow-up was 71·6 months, 60·0 (range 15·5-190·3) months for the pCR group and

73·0 (13·6-121·7) months for the non-pCR group (P = 0·323). Recurrence was identified in 94 patients

(50·0 per cent). There were 24 recurrences in the pCR group and 70 in the non-pCR group (P = 0·044).

Patterns of recurrence in the two groups are shown in Table 3. There were no significant differences

between the two groups in rates of locoregional recurrence alone (P = 0·945) or locoregional

recurrence with or without synchronous distant spread (P = 0·095). For patients who developed any

locoregional recurrence, the median interval to that recurrence was 8·7 (5·3-100·8) months in the





| Chapter 10

150

pCR group and 12·0 (1·9-69·2) months in the non-pCR group (P = 0·320). For patients who developed

distant metastases, the median interval to their occurrence (with or without synchronous locoregional

recurrence) was 8·7 (2·4-100·8) and 8·5 (0·6-51·9) months in the pCR and non-pCR groups respectively

(P = 0·257).

TABLE 3. Recurrence patterns

nNo recurrence

Locoregional recurrence alone*

Distant metastases alone† Both‡ Site

unknown§

All patients 188 94 (50.0) 14 (7.4) 53 (28.2) 25 (13.3) 2 (1.1)

pCR 62 38 (61) 4 (6) 16 (26) 4 (6) 0 (0)

Non-pCR 126 56 (44.4) 10 (7.9) 37 (29.4) 21 (16.7) 2 (1.6)

Values in parentheses are percentages; percentages may not total 100 because of rounding.*Recurrence of primary tumor and/or tumor in locoregional lymph nodes.†Hematogenous metastases or distant lymph node metastases.‡Synchronous locoregional recurrence and distant metastases.§Insufficient information to determine exact site of recurrence. pCR, pathologically complete response.

Pankeratin staining of the resection specimens of patients who developed recurrent disease after a

pCR revealed small clusters of vital tumour cells (lesions measuring no more than 0·2 mm)13 in the

peri-esophageal lymph nodes in two of eight patients.

Follow up in months3624120

Prop

ortio

nal s

urvi

val

1,0

0,8

0,6

0,4

0,2

0,0

pCR

non-pCR

Page 1

pCR 62 49 41 26non-pCR 126 91 62 47total 188 139 103 73

No. at risk

FIGURE 1. Estimated overall survival among 188 patients who underwent neoadjuvant chemoradiotherapy followed by surgery according to completeness of pathological response in the resection specimen. pCR, pathologically complete response. P=0.019 (log rank test

http://onlinelibrary.wiley.com/doi/10.1002/bjs.8968/full#fn8






151

Cha

pter

10

Follow up in months3624120

Pro

porti

onal

surv

ival

1,0

0,8

0,6

0,4

0,2

0,0

pCR

non-pCR

Page 1

pCR 62 48 40 25non-pCR 126 91 62 47total 188 139 102 72

No. at risk

FIGURE 2. Estimated disease-specific survival among 188 patients who underwent neoadjuvant chemoradiotherapy followed by surgery according to completeness of pathological response in the resection specimen. pCR, pathologically complete response. P=0.016 (log rank test)

The estimated 5-year overall survival rate for the pCR group was 52 per cent compared with 33·9

per cent for the non-pCR group (P = 0·019) (Fig. 1). Median overall survival was 60·0 and 26·6 months

respectively (hazard ratio 0·59, 95 per cent confidence interval 0·38 to 0·92). The estimated disease-

specific survival rate at 5 years was also significantly better in the pCR group than in the non-pCR

group (70 versus 47·8 per cent respectively; P = 0·016) (Fig. 2). In a multivariable Cox regression analysis,

only ypN category was found to be an independent predictor of survival (Table 4).

http://onlinelibrary.wiley.com/doi/10.1002/bjs.8968/full#fig1

http://onlinelibrary.wiley.com/doi/10.1002/bjs.8968/full#fig2


| Chapter 10

152

TABLE 4. Univariable and multivariable analysis to identify predictors of survival

Univariable analysis† Multivariable analysis‡

Hazard ratio P Hazard ratio P

Sex 1·05 (0·66, 1·69) 0·828

Age 1·00 (0·99, 1·03) 0·398

ASA grade 0·97 (0·69, 1·36) 0·862

Tumour type 1·08 (0·77, 1·53) 0·655

cT category 0·94 (0·74, 1·12) 0·618

cN category 0·90 (0·62, 1·32) 0·597

ypT category 1·06 (0·99, 1·13) 0·122

Tumour location 1·22 (0·97, 1·53) 0·850

ypN category 2·03 (1·38, 2·98) < 0·001 1·70 (1·02, 2·83) 0·043

Lymph node ratio* 3·98 (1·53, 10·36) 0·005 1·75 (0·52, 5·90) 0·370

pCR 0·59 (0·38, 0·92) 0·021 0·84 (0·50, 1·44) 0·532

Values in parentheses are 95 per cent confidence intervals.*Ratio of positive lymph nodes to total number of lymph nodes in resection specimen. ASA, American Society of Anesthesiologists physical status; cT/N, clinical tumour/node category according to the American Joint Commit-tee on Cancer (AJCC) tumour node metastasis (TNM) staging system (6th edition); ypT/N, pathological tumour/node category after neoadjuvant therapy in AJCC TNM staging system (6th edition); pCR, pathologically com-plete response.† Cox regression;‡ Cox proportional hazards model.

DISCUSSION

A pCR was observed after neoadjuvant chemoradiotherapy in one-third of the patients in this study.

The overall recurrence rate in the pCR group was significantly lower than in the non-pCR group,

although 13 per cent of these patients still developed locoregional recurrence, with or without

synchronous distant metastases. No significant difference was found in the time to locoregional

recurrence or distant dissemination between complete and incomplete responders.

Patterns of recurrence after esophagectomy have been studied in patients undergoing surgery

without neoadjuvant CRT. The locoregional recurrence rate was 13 per cent in 202 patients who

underwent primary, microscopically complete (R0) resection for stage II–IV esophageal carcinoma16,

a similar figure to that in the present pCR group. In a series of 100 consecutive patients with

oesophageal adenocarcinoma who underwent en bloc resection with extended lymphadenectomy

but without neoadjuvant therapy, the local recurrence rate was only 1 per cent, although 32 per

cent of these patients had a pT1 tumour, only 40 per cent a T3 tumour, those who underwent R1–2

resection were excluded and lymph node recurrences outside the surgical field were not scored

as locoregional recurrences.17 Lerut and colleagues15 reported a 5·2 per cent isolated locoregional

recurrence rate and another 9 per cent with synchronous locoregional and distant recurrence after





153

Cha

pter

10

three-field lymphadenectomy in a series of 174 patients.15 These data indicate that low locoregional

recurrence rates can also be achieved with appropriate primary surgery alone, raising the question of

how neoadjuvant CRT contributes to locoregional control. The commonly accepted rationale is that

preoperative CRT improves R0 resection rates and decreases the risk of leaving behind both known

and unknown nodal disease.

There have been several reports on recurrence patterns after neoadjuvant CRT and surgery. In a study

of 243 patients treated with cisplatin/5-FU and a radiotherapy dose of 40·0 Gy, 39 (16.0 per cent)

had a pCR. Overall 5-year survival rates in the pCR and non-pCR groups were 50 and 28 per cent

respectively, comparable with rates in the present study.18 In the pCR group, seven patients (18 per

cent) developed locoregional recurrence, with or without distant metastases, compared with 13 per

cent in the present study. In a group of 267 patients, 82 (30.7 per cent) had a pCR, of whom 22 per cent

developed a locoregional and/or distant recurrence, compared with 36 per cent of those without a

pCR19. Within the pCR group, 27 per cent of all these recurrences were located in the mediastinum or

(remnant) esophagus, again similar to the present study. These relatively high in-field recurrence rates

indicate that pCR is not synonymous with complete locoregional control.

It is acknowledged that immunohistochemical staining was not performed on sections from

patients with a pCR who developed distant metastases in the present study. The likelihood of distant

metastases in patients described as having a pCR might be greater if immunohistochemistry were

carried out routinely on all primary tumors and lymph nodes thought not to contain tumor deposits

on conventional histology.

Although significantly fewer patients developed recurrent disease in the pCR group, the time to

recurrence, either locoregional recurrence or distant metastasis, was comparable for patients with and

without a pCR. The disease-free interval in patients who had a pCR was not significantly different from

that in patients without a pCR (for both locoregional recurrences and distant metastases). This is in

contrast to the findings of a study involving 235 patients who underwent CRT followed by surgery, in

which three response types were described: complete (29 per cent), partial (46 per cent) and none (24

per cent); the disease-free interval was significantly longer in complete and partial responders than in

non-responders.20 The significantly longer disease-free interval in that study may be explained by the

different chemoradiation protocols used in the different groups, as well as differences in tumor stage.

One study has specifically examined the impact of surgical radicality (en bloc versus transhiatal

resection) on outcome in the context of preoperative CRT.21 Overall 5-year survival was significantly

better in the en bloc group (51 versus 22 per cent). Better overall 5-year survival in favor of the en bloc

group was also seen both in the presence of a pCR (70 versus 43 per cent) and with no pCR (48 versus

9 per cent). There were no locoregional recurrences in the pCR group suggesting that more radical

surgery, even when allied to neoadjuvant CRT, improved survival as well as decreasing the risk of

locoregional recurrence.

In the present study, after diagnosis of the first recurrence (either locoregional, distant metastases or a

synchronous combination of both), further imaging was driven by clinical need. Further locoregional






| Chapter 10

154

and/or distant failures may well have occurred. This implies that more patients assigned to the pCR

group may have had locoregional failure than were recorded.

Some confounding is inevitable in observational studies of this nature. Patient selection (only patients

with suspicious lymph nodes or questionable resectability were treated with neoadjuvant therapy)

resulted in a less favorable pretreatment tumor stage that might have led to worse outcomes than

reported in series dominated by early-stage tumors. It is acknowledged that the small numbers in

several subgroups of patients invites statistical misinterpretation.

Although it is clear that locoregional control is not always achieved by CRT and surgery, the high

percentage of patients who presented with distant metastases during follow-up was striking. The

radiosensitizing effect of chemotherapy is probably important for the improved locoregional tumor

control, but its systemic impact is clearly far from adequate.22 The lack of systemic chemotherapeutic

efficacy has also been suggested in studies using preoperative chemotherapy without radiotherapy.23

Although locoregional tumor control remains suboptimal after neoadjuvant CRT plus surgery (even in

patients with a pCR), preventing distant dissemination is the major clinical challenge.




155

Cha

pter

10

REFERENCES1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008:

GLOBOCAN 2008. Int J Cancer 2010;127(12): 2893-2917.

2. Parkin DM, Bray FI, Devesa SS. Cancer burden in the year 2000. The global picture. Eur J Cancer 2001;37 Suppl 8: S4-66.

3. Gannett DE, Wolf RF, Takahashi GW, Louie J, Wagner RC, Ey FS, Owens MM, Johnson WE, Cook DW, Alberty RE. Neoadjuvant chemoradiotherapy for esophageal cancer using weekly Paclitaxel and Carboplatin plus infusional 5-Fluorouracil. Gastrointest Cancer Res 2007;1(4): 132-138.

4. Lin CC, Hsu CH, Cheng JC, Wang HP, Lee JM, Yeh KH, Yang CH, Lin JT, Cheng AL, Lee YC. Concurrent chemoradiotherapy with twice weekly paclitaxel and cisplatin followed by esophagectomy for locally advanced esophageal cancer. Ann Oncol 2007;18(1): 93-98.

5. van Hagen P, Hulshof MC, van Lanschot JJ, et al. Preoperative chemoradiotherapy for esophageal or junctional cancer. N Engl J Med 2012;366(22): 2074-2084.

6. Burmeister BH, Smithers BM, Gebski V, Fitzgerald L, Simes RJ, Devitt P, Ackland S, Gotley DC, Joseph D, Millar J, North J, Walpole ET, Denham JW, Trans-Tasman Radiation Oncology G, Australasian Gastro-Intestinal Trials G. Surgery alone versus chemoradiotherapy followed by surgery for resectable cancer of the oesophagus: a randomised controlled phase III trial. Lancet Oncol 2005;6(9): 659-668.

7. Tepper J, Krasna MJ, Niedzwiecki D, Hollis D, Reed CE, Goldberg R, Kiel K, Willett C, Sugarbaker D, Mayer R. Phase III trial of trimodality therapy with cisplatin, fluorouracil, radiotherapy, and surgery compared with surgery alone for esophageal cancer: CALGB 9781. J Clin Oncol 2008;26(7): 1086-1092.

8. Urba SG, Orringer MB, Turrisi A, Iannettoni M, Forastiere A, Strawderman M. Randomized trial of preoperative chemoradiation versus surgery alone in patients with locoregional esophageal carcinoma. J Clin Oncol 2001;19(2): 305-313.

9. Schneider PM, Baldus SE, Metzger R, Kocher M, Bongartz R, Bollschweiler E, Schaefer H, Thiele J, Dienes HP, Mueller RP, Hoelscher AH. Histomorphologic tumor regression and lymph node metastases determine prognosis following neoadjuvant radiochemotherapy for esophageal cancer: implications for response classification. Ann Surg 2005;242(5): 684-692.

10. Mandard AM, Dalibard F, Mandard JC, Marnay J, Henry-Amar M, Petiot JF, Roussel A, Jacob JH, Segol P, Samama G, et al. Pathologic assessment of tumor regression after preoperative chemoradiotherapy of esophageal carcinoma. Clinicopathologic correlations. Cancer 1994;73(11): 2680-2686.

11. Page RD, Khalil JF, Whyte RI, Kaplan DK, Donnelly RJ. Esophagogastrectomy via left thoracophrenotomy. Ann Thorac Surg 1990;49(5): 763-766.

12. Muller JM, Erasmi H, Stelzner M, Zieren U, Pichlmaier H. Surgical therapy of oesophageal carcinoma. Br J Surg 1990;77(8): 845-857.

13. Singletary SE, Greene FL, Sobin LH. Classification of isolated tumor cells: clarification of the 6th edition of the American Joint Committee on Cancer Staging Manual. Cancer 2003;98(12): 2740-2741.

14. Lerut T, Nafteux P, Moons J, Coosemans W, Decker G, De Leyn P, Van Raemdonck D, Ectors N. Three-field lymphadenectomy for carcinoma of the esophagus and gastroesophageal junction in 174 R0 resections: impact on staging, disease-free survival, and outcome: a plea for adaptation of TNM classification in upper-half esophageal carcinoma. Ann Surg 2004;240(6): 962-972; discussion 972-964.

15. Lee PC, Mirza FM, Port JL, Stiles BM, Paul S, Christos P, Altorki NK. Predictors of recurrence and disease-free survival in patients with completely resected esophageal carcinoma. J Thorac Cardiovasc Surg 2011;141(5): 1196-1206.

| Chapter 10

156

16. Hagen JA, DeMeester SR, Peters JH, Chandrasoma P, DeMeester TR. Curative resection for esophageal adenocarcinoma: analysis of 100 en bloc esophagectomies. Ann Surg 2001;234(4): 520-530; discussion 530-521.

17. Reynolds JV, Muldoon C, Hollywood D, Ravi N, Rowley S, O’Byrne K, Kennedy J, Murphy TJ. Long-term outcomes following neoadjuvant chemoradiotherapy for esophageal cancer. Ann Surg 2007;245(5): 707-716.

18. Meguid RA, Hooker CM, Taylor JT, Kleinberg LR, Cattaneo SM, 2nd, Sussman MS, Yang SC, Heitmiller RF, Forastiere AA, Brock MV. Recurrence after neoadjuvant chemoradiation and surgery for esophageal cancer: does the pattern of recurrence differ for patients with complete response and those with partial or no response? J Thorac Cardiovasc Surg 2009;138(6): 1309-1317.

19. Rohatgi PR, Swisher SG, Correa AM, Wu TT, Liao Z, Komaki R, Walsh G, Vaporciyan A, Lynch PM, Rice DC, Roth JA, Ajani JA. Failure patterns correlate with the proportion of residual carcinoma after preoperative chemoradiotherapy for carcinoma of the esophagus. Cancer 2005;104(7): 1349-1355.

20. Rizzetto C, DeMeester SR, Hagen JA, Peyre CG, Lipham JC, DeMeester TR. En bloc esophagectomy reduces local recurrence and improves survival compared with transhiatal resection after neoadjuvant therapy for esophageal adenocarcinoma. J Thorac Cardiovasc Surg 2008;135(6): 1228-1236.

21. Hennequin C, Favaudon V. Biological basis for chemo-radiotherapy interactions. Eur J Cancer 2002;38(2): 223-230.

22. Allum WH, Stenning SP, Bancewicz J, Clark PI, Langley RE. Long-term results of a randomized trial of surgery with or without preoperative chemotherapy in esophageal cancer. J Clin Oncol 2009;27(30): 5062-5067.

A.K. Talsma

P. van Hagen

B.A. Grotenhuis

E.W. Steyerberg

H.W. Tilanus

J.J.B. van Lanschot

B.P.L. Wijnhoven

Annals of Surgical Oncology 2012; 19(7): 2142–2148

Chapter 11Comparison of the 6th and 7th editions

of the UICC-AJCC TNM classification for esophageal cancer

| Chapter 11

160

ABSTRACT

BACKGROUND

The new 7th edition of the Union for International Cancer Control–American Joint Committee on

Cancer (UICC-AJCC) tumor, node, metastasis (TNM) staging system is the ratification of data-driven

recommendations from the Worldwide Esophageal Cancer Collaboration database. Generalizability

remains questionable for single institutions. The present study serves as a validation of the 7th edition

of the TNM system in a prospective cohort of patients with predominantly adenocarcinomas from a

single institution.

METHODS

Included were patients who underwent transhiatal esophagectomy with curative intent between

1991 and 2008 for invasive carcinoma of the esophagus or gastroesophageal junction. Excluded were

patients who had received neoadjuvant chemo(radio)therapy, patients after a noncurative resection

and patients who died in the hospital. Tumors were staged according to both the 6th and the 7th

editions of the UICC-AJCC staging systems. Survival was calculated by the Kaplan–Meier method, and

multivariate analysis was performed with a Cox regression model. The likelihood ratio chi-square test

related to the Cox regression model and the Akaike information criterion were used for measuring

goodness of fit.

RESULTS

A study population of 358 patients was identified. All patients underwent transhiatal esophagectomy

for adenocarcinoma. Overall 5-year survival rate was 38%. Univariate analysis revealed that pT stage,

pN stage, and pM stage significantly predicted overall survival. Prediction was best for the 7th edition,

stratifying for all substages.

CONCLUSIONS

The application of the 7th UICC-AJCC staging system results in a better prognostic stratification

of overall survival compared to the 6th edition. The fact that the 7th edition performs better

predominantly in patients with adenocarcinomas who underwent a transhiatal surgical approach,

in addition to findings from earlier research in other cohorts, supports its generalizability for different

esophageal cancer practices.

Comparison of the 6th and 7th editions of the UICC-AJCC TNM classification for esophageal cancer |

161

Cha

pter

11

INTRODUCTION

Accurate staging of cancer is important for stage-specific treatment, thus minimizing inappropriate

treatment. Moreover, it allows for interinstitutional comparisons and disclosure of prognosis to patients.1

The staging system for cancer in the esophagus and esophagogastric junction has been revised as

outlined in the 7th edition of the Union for International Cancer Control/Union Internationale Contre

le Cancer (UICC) and the American Joint Committee on Cancer (AJCC), Cancer Staging Manual.2

Retrospective studies suggested that the number of involved lymph nodes is a better predictor of

outcome than classifying lymph node involvement as either present or absent.3,4 Peyre et al. showed

that patients with ≥3 lymph nodes involved have a risk of systemic disease that exceeds 50%. When >8

nodes are involved, the risk of dying is almost 100%.5 Indeed, the latest 7th edition of the UICC-AJCC

esophageal tumor, node, metastasis (TNM) staging system has acknowledged the importance of the

number of involved nodes by revising the N category from site-dependent staging to a numerically

based classification into N0 to N3. Another major change is the definition of regional lymph nodes.

The new UICC-AJCC staging system is the ratification of data-driven recommendations from a database

of >7800 esophageal cancer patients created from a large multi-institutional collaboration involving

13 institutions.6,7 This Worldwide Esophageal Cancer Collaboration (WECC) database overcomes

problems of rarity of this cancer, but generalizability remains questionable for single institutions. WECC

incorporates high-volume centers both from the West (where adenocarcinomas prevail) and from the

East (where most tumors are squamous cell carcinomas). Moreover, the extent of intrathoracic lymph

node dissection can vary greatly between different institutions, leading to potential bias.

The present study serves as a validation of the WECC-based 7th edition of the TNM system in a

cohort of patients with both squamous cell carcinomas and adenocarcinomas from a single Western

high-volume institution. Two studies already showed that the 7th edition criteria resulted in better

prognostic stratification than the 6th edition.8,9 However, both study cohorts consisted of squamous

cell carcinomas or junctional tumors, respectively. Moreover, Gaur et al. included patients who

received (neo)adjuvant therapy.9

The aim of this study was to assess the predictive ability of the 7th edition of the AJCC TNM staging

system for overall survival and to compare this with the 6th edition in a cohort of patients who

underwent transhiatal esophagectomy for adenocarcinomas without (neo)adjuvant therapy.


Study Population

Included were all patients who underwent a transhiatal esophagectomy with curative intent between

January 1991 and September 2008 at the Erasmus Medical Center (Rotterdam, The Netherlands) for

invasive squamous cell carcinoma and adenocarcinoma of the esophagus or gastroesophageal

junction. Excluded were patients who had received neoadjuvant chemo(radio)therapy, patients

https://vpn-gw.erasmusmc.nl/+CSCO+1h756767633A2F2F6A6A6A2E61706F762E61797A2E6176752E746269++/pmc/articles/PMC3381120/#CR1









| Chapter 11

162

after a noncurative (R1) resection (tumor-free margin <1 mm) and patients who died in the hospital.

Clinicopathologic data of all patients had been routinely collected in an ongoing prospective registry.

Surgery

Transhiatal esophagectomy with cervical anastomosis was the chosen surgical approach in the

present study. This encompasses the en-bloc dissection of the primary tumor and its adjacent lymph

nodes under direct vision through the widened hiatus of the diaphragm up to the level of the

inferior pulmonary vein. Subsequently, a 3–4-cm-wide gastric tube is created. The left gastric artery is

transected at its origin with resection of celiac trunk lymph nodes. After mobilization and transection

of the cervical esophagus, the intrathoracic middle and upper esophagus is bluntly dissected in an

antegrade fashion with a vein stripper. Esophagogastrostomy is performed in the neck without a

formal cervical lymphadenectomy.

Follow-up

Surviving patients were followed at regular intervals at the outpatient clinic until 5 years after surgery.

Outpatient clinic visits encompassed history taking and physical examination. No routine imaging

was performed. Recurrences were sought afterward, only when clinically indicated, by CT scan or

ultrasound and proven by histology and cytology whenever possible. Overall survival was defined

as the time between date of operation and date of death. Surviving patients were censored on the

day of last follow-up. Patient survival status was calculated after contacting the general practitioners

(performed by a trained data manager). The last follow-up checkpoint was July 2010. If follow-up was

incomplete, survival was verified in the municipal mortality registers.

Statistical Analysis

Tumors were staged according to both the 6th and 7th editions of the UICC-AJCC staging systems.

Survival was calculated by the Kaplan–Meier method, and differences between curves were assessed

by the log rank test.

Two multivariable models were built, one with the 6th edition and one with the 7th edition of the

TNM staging system as categorical variables. The performance was tested for the model in which the

stages were combined into four categories (I–IV) as well as for the model with all substages included

(IA, IB, IIA, IIB, IIIA, IIIB, IIIC, IV). A multivariable model with both 6th and 7th edition criteria included was

used to assess the remaining value of the 6th edition when the 7th edition information was known.

The likelihood ratio chi-square test related to the Cox regression model was used for measuring

goodness of fit. The Akaike information criterion (AIC) was applied to correct for the potential bias

in comparing prognostic systems with different number of stages.10,11 The −2 log likelihood (which is

the parameter in the Cox regression) of the 6th edition was compared to that of the 7th edition; the

smaller the value of this statistic, the better the model.




163

Cha

pter

11

AIC was defined as: AIC = −2 log maximum likelihood + 2 × (the number of parameters in the model).

A smaller AIC value indicates a more desirable model for predicting outcome. A value of P < 0.05 was

considered statistically significant. Statistical analysis was performed with SPSS 10 for Windows (SPSS,

Chicago, IL).

RESULTS

Patient Characteristics

A consecutive series of 766 patients underwent esophagectomy with curative intent. In total, 221

patients were excluded because they had received neoadjuvant chemo(radio)therapy in the context

of a randomized, controlled trial.12 Another 165 patients were excluded because of a noncurative (R1)

resection, and 20 patients were excluded because of in-hospital mortality. Two patients had an in situ

carcinoma and were also excluded from the current analysis. This resulted in a final study population

of 358 patients.

Mean follow-up was 51 months (median 37 months). Overall 5-year survival rate was 38%. Most

recurrences of disease occurred within 2 years after surgery. Patient characteristics and overall

survival rates are summarized in Table 1. All patients underwent transhiatal esophagectomy for

adenocarcinoma. Eight patients seemed to have distant metastasis during the operation; their disease

was scored as M1.

Univariate analysis revealed that parameters pT stage, pN stage, and pM stage all significantly

predicted overall survival. Except for histologic grade, no other significant predictors of survival were

detected in this univariate analysis. The median number of dissected nodes per patient was 11. In

patients with negative lymph nodes (pN0), the survival rates did not differ between patients with ≤11

nodes and >11 nodes dissected: 65% vs. 69%, respectively; P = 0.65; data not shown).

Stratification of Prognosis According to 6th and 7th Editions of TNM Staging Systems

The overall survival curves according to the N classifications of the 6th and 7th editions are shown in

Fig. 1a and b, respectively.

Patient stage migration for reclassifying patients from the 6th to the 7th staging system and their

survival rates are listed in Table 2. In 58% of the 358 esophageal cancer patients, stage did not differ in

these two classification systems. Reassignment of disease stage occurred in all other patients, either to

a higher or to a lower tier. According to the 6th edition staging system, 56 (87%) of 64 stage IV patients

were staged as such because of a celiac lymph node metastasis. These patients were reclassified to a

lower tier in the 7th edition: 6 of 64 were staged as stage IIB, 15 as stage IIIA, 19 as IIIB, and 16 as IIIC

(Table 2).


https://vpn-gw.erasmusmc.nl/+CSCO+1h756767633A2F2F6A6A6A2E61706F762E61797A2E6176752E746269++/pmc/articles/PMC3381120/table/Tab1/

https://vpn-gw.erasmusmc.nl/+CSCO+1h756767633A2F2F6A6A6A2E61706F762E61797A2E6176752E746269++/pmc/articles/PMC3381120/figure/Fig1/



| Chapter 11

164

TABLE 1. Patient demographics and results of univariate analysis for overall survival (N = 358)

Characteristic Value 5-y survival, % P

No. of patients 358

Age, year, mean (range) 62.6 (28–83) 38.8

Gender

Male 293 (82%) 37.2 0.664

Female 65 (18%) 45.9

pT

1 78 (22%) 68.7 <0.001

2 79 (22%) 51.1 <0.001

3 201 (56%) 22.7

pN

0 146 (41%) 65.9 <0.001

1 90 (25%) 28.4 <0.001

2 81 (23%) 17.5 <0.001

3 41 (11%) 3.0

pM

0 350 39.7<0.001

1 8 0.0

Grade

Well differentiated (G1) 31 (9%) 75.3 <0.001

Moderately differentiated (G2) 177 (49%) 39.4 <0.053

Poorly differentiated (G3) 150 (42%) 30.9

Histology

Squamous cell carcinoma 47 (13%) 41.9 0.752

Adenocarcinoma 311 (87%) 38.3

Location

Upper third 6 (2%) 30.4 0.352

Middle third 14 (4%) 42.6 0.325

Lower third (distal + EGJ) 338 (94%) 36.9

Type of surgical approach

Transhiatal esophagectomy 358 (100%)

Transthoracic esophagectomy

T tumor stage (depth of invasion), N lymphatic dissemination stage (according to 7th edition of UICC-AJCC TNM staging system: N0 no positive lymph nodes, N1 1–2 positive lymph nodes, N2 3–6 positive lymph nodes, N3 ≥6 positive lymph nodes), M distant metastasis stage (according to 7th edition of UICC-AJCC TNM staging system: M0 no metastasis, M1 distant metastasis present), EGJ esophagogastric junction.


165

Cha

pter

11

FIGURE 1. Kaplan–Meier overall survival curves for 358 patients stratified by N stage according to a 7th edition and b 6th edition UICC-AJCC TNM staging systems (overall log rank P < 0.01)

TABLE 2. Cross table of staging esophageal cancer patients according to the 6th and 7th editions of UICC-AJCC TNM staging

6th editiona5 year-survival according to

7th edition (%)I IIA IIB III IV

7th editionb

IA 43 0 0 0 0 87.7

IB 13 28 0 0 0 73.3

IIA 0 19 0 0 0 55.3

IIB 0 41 24 0 6 40.1

IIIA 0 0 21 50 15 24.3

IIIB 0 0 0 31 19 11.9

IIIC 0 0 4 20 16 3.1

IV 0 0 0 0 8 0.0

5 year-survival according to 6th edition (%) 81.9 56.8 38.3 14.1 12.4

M1a celiac nodes involved in lower esophageal cancer or cervical nodes involved in upper esophageal cancer, M1b beyond locoregional node involvement (i.e., cervical nodes in lower esophageal cancer and celiac nodes in upper esophageal cancer; metastatic involvement of visceral organs, pleura, peritoneum)aThe 6th edition AJCC-UICC TNM staging system: stage I T

1N

0, stage IIA T

2,3N

0, stage IIB T

1,2N

1, stage III T

3N

1 or T

4N

0,

stage IVA Tany

Nany

M1a, stage IVB Tany

Nany

M1b. The 7th edition AJCC-UICC TNM staging system (for adenocarcinoma): stage IA T

1N

0G

1,2, stage IB T

1N

0G

3 or T

2N

0G

1,2, stage IIA T

2N

0, stage IIB T

3N

0 or T

1,2N

1, stage IIIA T

4N

0 or T

3N

1 or T

1,2N

2, stage

IIIB T3N

2, stage IIIC T

anyN

3 or T

4aN

1–3 or T

4bN

any, stage IV T

any,N

any,M

1.

| Chapter 11

166

FIGURE 2. Kaplan–Meier curves of overall survival for 358 patients stratified according to a 6th edition and b 7th edition UICC-AJCC TNM staging systems

FIGURE 3. Kaplan–Meier overall survival curves for 64 UICC-AJCC 6th stage IV patients who were reclassified according to UICC-AJCC 7th edition TNM staging (log rank P = 0.43).

The Kaplan–Meier curves of esophageal cancer patients based on the 6th and 7th editions of the TNM

staging systems are depicted in Fig. 2. Both systems show a relatively ordered monotone distribution

of survival. However, according to the 6th edition staging system, the Kaplan–Meier plot shows



167

Cha

pter

11

overlapping curves for stage III and IV. In the 7th edition, no important overlapping occurs among

stages I through IV.

Subgroup analysis among selected patients who had been considered to have stage IV disease

according to the UICC-AJCC 6th edition scoring system showed that patients reclassified from stage

IV disease to a lower tier in the UICC-AJCC 7th edition had a significantly better survival compared to

patients still classified as stage IV according to the UICC-AJCC 7th edition. Moreover, the UICC-AJCC

7th edition was able to make further significant stratification of survival rates of these reclassified

patients (Fig. 3; log rank P = 0.43).

The UICC-AJCC 7th edition staging system defines patients with positive paraesophageal cervical

lymph nodes (n = 10) as having stage IIIA or IIIB disease. These patients, however, had a prognosis as

bad as that of patients with distant metastasis (1-year overall survival rate 30% vs. 33%).

The performance of the 6th and 7th edition staging systems won quantified by the likelihood ratio

chi-square and AIC (Table 3). Predictive ability was best for the full 7th edition criteria stratifying for

all substages (highest likelihood ratio χ2). AIC value was smaller for the 7th edition compared to the

6th edition staging system, indicating that it has a better prognostic stratification. The AIC value was

lowest when patients with cervical lymph node metastasis at a large distance from the primary tumor

(i.e., the lower third of the esophagus) were also classified as having stage IV disease. When the 6th

and 7th edition staging systems are both included in one Cox regression model, the 6th edition no

longer significantly predicted survival, whereas the 7th edition remained a significant stratifier of

prognosis (data not shown).

TABLE 3. Prognostic stratification of the 6th and 7th editions of the UICC-AJCC TNM staging systems

Model Figure Subgroups LR χ2 AIC valuea

6th edition 2a I, II, III, IV 96.9 2607.1

7th edition, full 2b IA, IB, IIA, IIB, IIIA, IIIB, IIIC, IV 128.6 2592.9

7th edition, collapsed I, II, III, IV 99.0 2605.4

AIC Akaike information criteria, LR likelihood ratioaA lower AIC value represents a better discriminatory model

DISCUSSION

This study shows that both the 6th and 7th UICC-AJCC TNM staging systems have a distinctive and

monotone (ordered) relationship of stage group to overall survival for esophageal cancer patients

who have undergone potentially curative surgery without (neo)adjuvant therapy. Distribution of

patients among different stages is in line with that described in the literature. All groups are large

enough for proper statistical analysis, except for stage IIA in the 7th edition.


| Chapter 11

168

Further testing of both systems on the present data shows that the 7th edition has the best

performance because of the lowest AIC (i.e., a better fit) when Cox regression models are used. Survival

curves stratified according to the UICC-AJCC 7th edition TNM staging system did not overlap, which

is in contrast to the curves of the 6th edition. Moreover, further stratification of N stage according to

number of positive lymph nodes in the 7th edition is indeed valuable, as shown in Fig. 1.

A major change in the new TNM staging system is the definition of regional lymph nodes. There

has always been debate regarding the prognostic importance of positive celiac nodes, which were

considered distant metastases in earlier editions.13 In the 6th edition staging system, the Kaplan–Meier

plot showed overlapping curves for stage III and IV. According to the UICC-AJCC 7th edition, only

patients with distant metastasis can be categorized as having stage IV disease. In contrast, according

to the 6th edition, most stage IV disease was due to nonregional celiac lymph node metastasis,

whereas stage IIB and III consisted of regional lymph node metastasis. Hence, 87% (56 of 64) of the

patients with stage IV disease who were assessed according to the 6th edition criteria were reclassified

as having stage IIB, IIIA, IIIB, and IIIC disease according to 7th edition criteria. Because these stages all

had different survivals (Fig. 3), the present results support the new concept that it is unnecessary to

identify nonregional lymph node metastasis and to label these nodes as M1A or M1B.

Two previous studies have compared the performance of 6th with the 7th editions of the TNM staging

system in predicting survival. Hsu et al. evaluated 392 patients who underwent primary surgical

resection through a tri-incisional approach in Taiwan during 1995–2006.8 In the other study, nearly

two-thirds of the patients received neoadjuvant therapy.9

Both Hsu et al. and Gaur et al. concluded that the 7th edition of the staging system was a better model

for predicting outcome.8,9 The most important difference with the present study is tumor histology;

the vast majority of our patients had an adenocarcinoma, and almost all patients underwent a

transhiatal resection.

The WECC-based 7th edition of the TNM staging system was built on data from patients without

neoadjuvant treatment in a squamous cell carcinoma predominant database. Our sample population

from a single institution is of course small compared with the worldwide esophageal cancer

collaboration database, but the surgical procedures were highly uniform throughout the entire

study period. The previous studies of Hsu et al. and Gaur et al., as well as the present study, underline

the generalizability of the 7th edition and make it broadly applicable for daily clinical practice of

esophageal cancer surgery around the world.8,9

The 7th edition of the UICC-AJCC esophageal TNM staging system has acknowledged the importance

of the number of involved nodes by subdividing the N classification into N0 to N3. The transhiatal

approach may profoundly affect the completeness of lymph node dissection and, accordingly, proper

nodal staging. On the basis of data from a Dutch trial, nowadays, tumors proximal of esophagogastric

junction (Siewert type 1) are preferably offered a transthoracic approach in our institution.14,15 The

latter approach will result in the collection of more lymph nodes and might give a more valid node

sampling for staging. To which extent lymph nodes should be sampled for proper staging remains













169

Cha

pter

11

an important issue.16 In a study performed by Peyre et al., the number of lymph nodes removed

was an independent predictor of survival and a minimum number of 23 regional lymph nodes was

proposed.17 In the present study, the median number of nodes removed in a transhiatal approach was

11. This relatively scarce lymph node collection result can be seen as a drawback of our study, but it

also gives rise to a remarkable finding. Although all patients underwent a transhiatal esophagectomy,

the survival curves of different N stages (N0–N3; Fig. 1) do not overlap in our data, which probably

indicates that there has been a valid and robust node sampling. On the other hand, there seems

to be a relatively large difference in survival rate between N0 and N1. We know from previous

studies that there is a dichotomy in survival rate between tumors that did and did not lymphatically

disseminate.18 Early tumors (pT1) with lymph node invasion have prognosis comparable to tumors

with more advanced T stage. Lymphatic dissemination is an independent indication of the biological

aggressiveness of the tumor.

However, the large step in survival rate between N0 and N1 might also be due to a stage migrational

effect. This, the so-called Will Rogers effect, means that stage N1 disease might actually include N2

or even N3 disease as a result of invalid node sampling.19 The WECC group has indicated a resection

of a minimum of 10 nodes for T1, 20 for T2, and ≥30 nodes for T3–4 to be resected to obtain optimal

results.20 In N0 patients, such an effect does not occur; we found no significant difference in survival

rates according to the number of resected lymph nodes in lymph node–negative patients. However,

a median of 11 nodes definitely entails the risk of a stage migration effect in the patient group with

positive nodes.

Finally, an important question remains: does a better predictive staging system have consequences

for preoperative decision making? Medical decision making in terms of administering neoadjuvant

chemotherapy and choosing the optimal surgical approach for esophagectomy is often based on

clinical N staging. Lack of accurate preoperative staging is a major problem in allocating treatment

modalities in these patients. It has been recently shown that further stratification according to the

position of the positive node relative to the diaphragm can effectively discriminate between node-

positive patients.21 The overall accuracy for endoscopic ultrasound and CT in predicting the N stage

per station is moderate, however. When the therapeutic approach depends on the status of a specific

lymph node station, a more objective and reliable assessment of lymph nodal involvement (e.g.,

endoscopic ultrasound–fine-needle aspiration) should be considered.22

This study indicates that the application of the 7th UICC-AJCC staging system results in a better

prognostic stratification of overall survival compared to the 6th edition. The fact that the 7th edition

also has a superior prognostic ability in this study population from a single high-volume institution with

predominantly adenocarcinomas and a two-incisional surgical approach supports its generalizability

for different esophageal cancer practices.









| Chapter 11

170

REFERENCES1. Lagarde SM, Franssen SJ, Werven JR, et al. Patient preferences for the disclosure of prognosis after esophagectomy

for cancer with curative intent. Ann Surg Oncol. 2008;15:3289–3298

2. Sobin LH, Gospodarowicz MK, Wittekind C. TNM classification of malignant tumors. 7. Oxford: Wiley-Blackwell; 2010.

3. Thompson SK, Ruszkiewicz AR, Jamieson GG, et al. Improving the accuracy of TNM staging in esophageal cancer: a pathological review of resected specimens. Ann Surg Oncol. 2008;15:3447–3458

4. Kato H, Tachimori Y, Watanabe H, Iizuka T. Evaluation of the new (1987) TNM classification for thoracic esophageal tumors. Int J Cancer. 1993;53:220–223

5. Peyre CG, Hagen JA, DeMeester SR, et al. Predicting systemic disease in patients with esophageal cancer after esophagectomy: a multinational study on the significance of the number of involved lymph nodes. Ann Surg. 2008;248:979–985

6. Rusch VW, Rice TW, Crowley J, Blackstone EH, Rami-Porta R, Goldstraw P. The seventh edition of the American Joint Committee on Cancer/International Union Against Cancer Staging Manuals: the new era of data-driven revisions. J Thorac Cardiovasc Surg. 2010;139:819–821

7. Rice TW, Rusch VW, Apperson-Hansen C, et al. Worldwide esophageal cancer collaboration. Dis Esophagus. 2009;22:1–8

8. Hsu PK, Wu YC, Chou TY, Huang CS, Hsu WH. Comparison of the 6th and 7th editions of the American Joint Committee on Cancer tumor–node–metastasis staging system in patients with resected esophageal carcinoma. Ann Thorac Surg. 2010;89:1024–1031

9. Gaur P, Hofstetter WL, Bekele BN, et al. Comparison between established and the Worldwide Esophageal Cancer Collaboration staging systems. Ann Thorac Surg. 2010;89:1797–803, 1804 e1791–3

10. Armitage P, Colton T. Encyclopedia of biostatistics. New York: Wiley; 1998. 11. Kee KM, Wang JH, Lee CM, et al. Validation of clinical AJCC/UICC TNM staging system for hepatocellular

carcinoma: analysis of 5,613 cases from a medical center in southern Taiwan. Int J Cancer. 2007;120:2650–2655

12. Heijl M, Lanschot JJ, Koppert LB, et al. Neoadjuvant chemoradiation followed by surgery versus surgery alone for patients with adenocarcinoma or squamous cell carcinoma of the esophagus (CROSS) BMC Surg. 2008;8:21

13. Eloubeidi MA, Wallace MB, Hoffman BJ, et al. Predictors of survival for esophageal cancer patients with and without celiac axis lymphadenopathy: impact of staging endosonography. Ann Thorac Surg. 2001;72:212–219

14. Hulscher JB, Sandick JW, Boer AG, et al. Extended transthoracic resection compared with limited transhiatal resection for adenocarcinoma of the esophagus. N Engl J Med. 2002;347:1662–1669

15. Siewert JR, Stein HJ. Classification of adenocarcinoma of the esophagogastric junction. Br J Surg. 1998;85:1457–1459

16. Lerut T, Nafteux P, Moons J, et al. Three-field lymphadenectomy for carcinoma of the esophagus and gastresophageal junction in 174 R0 resections: impact on staging, disease-free survival, and outcome: a plea for adaptation of TNM classification in upper-half esophageal carcinoma. Ann Surg. 2004;240:962–972

17. Peyre CG, Hagen JA, DeMeester SR, et al. The number of lymph nodes removed predicts survival in esophageal cancer: an international study on the impact of extent of surgical resection. Ann Surg. 2008;248:549–556

18. Westerterp M, Koppert LB, Buskens CJ, et al. Outcome of surgical treatment for early adenocarcinoma of the esophagus or gastro-esophageal junction. Virchows Arch. 2005;446:497–504

19. Feinstein AR, Sosin DM, Wells CK. The Will Rogers phenomenon. Stage migration and new diagnostic techniques as a source of misleading statistics for survival in cancer. N Engl J Med. 1985;312:1604–1608


171

Cha

pter

11

20. Rizk NP, Ishwaran H, Rice TW, et al. Optimum lymphadenectomy for esophageal cancer. Ann Surg. 2010;251:46–50

21. Peters CJ, Hardwick RH, Vowler SL, Fitzgerald RC. Generation and validation of a revised classification for esophageal and junctional adenocarcinoma. Br J Surg. 2009;96:724–733

22. Grotenhuis BA, Poley JW, Wijnhoven BPL. Preoperative assessment of tumor location and station-specific lymph nodal status is inaccurate in patients with esophageal adenocarcinoma. World J Surg 2013 Jan;37(1):147-155

P. van Hagen

R. de Jonge

M. I. van Berge Henegouwen

G. J. Hötte

E. P. van der Stok

J. Lindemans

J. J. B. van Lanschot

B. P.L. Wijnhoven

Submitted

Chapter 12Vitamin B12 deficiency after esophagectomy with gastric

tube reconstruction for esophageal cancer

| Chapter 12

174

ABSTRACT

OBJECTIVE

To determine the prevalence and incidence of vitamin B12

deficiency after esophagectomy for cancer.

BACKGROUND

It is unknown if patients after esophagectomy with gastric tube reconstruction are at an increased

risk for vitamin B12

deficiency.

METHODS

A cross sectional cohort (group A) and a prospective cohort (group B) of patients who underwent

esophagectomy for cancer in two tertiary referral centers in the Netherlands were included. Serum

levels of holo-transcobalamin (Holo-TC) and methyl malonic acid (MMA) were determined. Vitamin

B12

deficiency was defined as Holo-TC <21pmol/l and/or MMA >0.45 µmol/l. Vitamin B12

status was

assessed in group A at a single time point between one and three years postoperatively and before

and every three months after resection in group B.

RESULTS

Ninety-nine patients were analyzed in group A. The median time between surgery and analysis

of vitamin B12

deficiency was 19.3 months. In 11 of 99 (11%) patients, vitamin B12

deficiency was

detected. In group B, 5 of 88 (5.6%) patients had vitamin B12

deficiency preoperatively, and another 9

(10.2%) patients developed vitamin B12

deficiency after the operation at a median time of 6 months

postoperatively. The estimated one-year incidence of vitamin B12

deficiency was 18.2%. None of the

patients with vitamin B12

deficiency had a megaloblastic anemia.

CONCLUSION

Vitamin B12

deficiency can be anticipated in 18% of patients after esophagectomy with gastric tube

reconstruction for cancer. During follow-up, Holo-TC and MMA levels should be measured to detect

vitamin B12

deficiency and commence treatment timely.

Vitamin B12

deficiency after esophagectomy with gastric tube reconstruction for esophageal cancer |

175

Cha

pter

12

INTRODUCTION

Vitamin B12

, or cobalamin, is a water-soluble co-enzyme that plays a crucial role in the synthesis of DNA

and the myelinization of neural cells. Vitamin B12

must be obtained through the diet (present mainly

in meat, fish and dairy products), for it cannot be synthesized by the human body.1 After ingestion,

it is bound to intrinsic factor (IF), a glycoprotein which is produced by the parietal cells of the gastric

mucosa. The vitamin B12

-IF complex is then absorbed in the terminal ileum, after which vitamin B12

is

bound to Transcobalamin-II, to form holo-transcobalamin (Holo-TC, active vitamin B12

) and released

into the circulation.2 Methyl malonic acid (MMA) is a precursor of succinyl-CoA, an essential metabolite

in the Krebs cycle. Vitamin B12

is an essential cofactor in the conversion of MMA to succinyl-coA. In case

of vitamin B12 deficiency, MMA will accumulate.

A deficiency of vitamin B12

can lead to a wide variety of signs and symptoms. Megaloblastic anemia

and neurological disorders are the most common and severe complications resulting from vitamin

B12

deficiency.1 When diagnosed early, supplementing vitamin B12

can reverse the deficiency and cure

symptoms.

The incidence of esophageal cancer has risen over the last decades, making it the eighth most

common cancer worldwide.3 About half of the patients are potentially curable at presentation4 The

standard treatment for locally advanced esophageal cancer in the Netherlands is surgical resection,

preceded by neoadjuvant chemoradiotherapy.5 During the operation, the thoracic esophagus is

removed including a substantial part of the proximal stomach. For reconstruction of the digestive

tract, the stomach is tubularized and anastomosed to the cervical esophageal remnant, thereby

performing a so-called gastric tube reconstruction.6 Esophagectomy is accompanied by profound

anatomical and functional changes including a bilateral vagotomy.7-9

In bariatric surgery, there is evidence that partially resecting or functionally bypassing the stomach

can lead to nutritional deficiencies.7 Whether the altered anatomy after gastric tube reconstruction

following esophagectomy induces vitamin B12

deficiency is unknown. The aim of this study was to

determine the prevalence and incidence of vitamin B12

deficiency following esophagectomy with

gastric tube reconstruction in patients treated for potentially curable esophageal cancer.


Inclusion criteria

Two cohorts of patients were defined: a cross-sectional cohort (group A) and a prospective longitudinal

cohort (group B).

Group A: Patients from group A were recruited from two high volume centers (Erasmus University

Medical Center, Rotterdam and Academic Medical Center, Amsterdam, The Netherlands). Both centers

had an institutional database where all patients were registered who underwent an esophagectomy

in the past 20 years. Patients were eligible to participate if they had undergone esophagectomy with

| Chapter 12

176

gastric tube reconstruction for esophageal cancer between one and three years prior to inclusion

in the study. Excluded were all patients that had clinical signs or symptoms of recurrent disease as

determined by physical examination and/or additional diagnostic examinations.

Group B: All patients who were scheduled for esophagectomy with gastric tube reconstruction for

potentially curable esophageal cancer at the Erasmus University Medical Center Rotterdam were

eligible for inclusion. Patients who were already on enteral or parenteral vitamin B12

were excluded.

All patients had to be eighteen years or older and provide written informed consent before entering

the study. Patients were included one day prior to surgery after informed consent was obtained and

followed for at least one year after the operation. The study protocol was approved by the medical

ethical committees of both participating centers.

Treatment

A transhiatal or transthoracic resection with gastric tube reconstruction was performed as previously

described.4,8 When appropriate, patients were treated with neoadjuvant chemoradiotherapy according

to a previously described regimen of Paclitaxel/Carboplatin with concurrent radiotherapy of 41.4 Gy in

23 fractions.5 The standard treatment protocol for esophageal cancer in both participating centers did

not include routine testing or (prophylactic) substitution of vitamin B12

.

Blood samples

Group A: Blood samples were collected once in the outpatient clinic during a routine visit. If a vitamin

B12

deficiency was diagnosed, substitution treatment was provided.

Group B: Blood samples were collected at baseline (one day before surgery) and during each regular

visit to the outpatient clinic every three months postoperatively. If a patient developed a vitamin B12

deficiency during follow-up, substitution treatment was provided and the patient was excluded from

further follow-up. If progression of disease (i.e. locoregional recurrence or distant dissemination) was

discovered during follow-up, the patient was also excluded from the study.

Laboratory testing

Patients’ blood samples underwent a standard set of laboratory tests. Routine laboratory procedures

(Roche Modular and Cobas 8000; Roche, Almere, the Netherlands) were used to measure serum

creatinin, pre-albumin, c-reactive protein (CRP) and folate (E170/c602). Holo-transcobalamin (Holo-TC,

or active VB12

) (AxSYM analyser; Abbott, Hoofddorp, The Netherlands) and methyl malonic acid (MMA)

(LC-MS/MS) were measured as previously described.9 Hemoglobin (Hb), mean corpuscular volume

(MCV), platelets, and white blood count (WBC) were measured on a Sysmex HST-302 or XN-9000

(Sysmex, Etten-Leur, The Netherlands).

Vitamin B12


177

Cha

pter

12

Definition of B12 deficiency

Vitamin B12

deficiency was defined as a level of Holo-TC <21 pmol/l and/or an MMA level of >0.45

µmol/l.9


Baseline characteristics, incidence and prevalence of vitamin B12

deficiencies were calculated

using descriptive statistical methods. For group B, the estimated one-year incidence of vitamin B12

deficiency was calculated using the Kaplan-Meier method. Laboratory results that were obtained after

the 12 month limit were excluded from analysis. Post-hoc subgroup analyses were not performed. A

multilevel regression analysis was performed to estimate the mean Holo-TC and MMA levels and to

test for significant increase or decrease of these levels during the follow-up period. A p-value of <0.05

(two-sided) was considered as statistically significant.

RESULTS

Between August 2010 and July 2012, a total of 187 patients were included in the study. Ninety-nine

patients were included in group A (cross-sectional cohort) and 88 patients in group B (longitudinal

cohort). Baseline characteristics of these patients are shown in Table 1.

Group A: The median time after surgery in group A was 19.3 months (IQR 13.5 - 30.0 months). In 11

of 99 (11%) patients, a vitamin B12

deficiency was detected at a median time of 20.8 months (IQR 14.3

- 29.7) after surgery. Five of these 11 patients had a Holo-TC below 21 pmol/l, five of 11 had an MMA

level >0.45 µmol/l and in one patient both Holo-TC and MMA were abnormal (Table 2).

Six of 11 patients in group A with a vitamin B12

deficiency suffered from mild anemia (minimal

measured Hb 8.0 mmol/l). No patients had a megaloblastic anemia (maximum MCV 99 fl).

Group B: Eighteen patients (20%) were followed for the maximum duration of follow-up (12 months),

while 37 (42%) patients could be followed for at least nine months. The median follow-up was 6.4

months for all patients, and 8.9 months for patients who were not excluded from follow-up due to

vitamin B12

deficiency. Other reasons for an incomplete follow-up were progression of disease or death

during the first postoperative year (n=40), incorrect planning of outpatient clinic visits (n=21), patient

withdrawal (n=7) and prescription of vitamin B12

supplements outside the study protocol (n=2).

During the study period, in 14 of 88 (15.9%) patients a vitamin B12

deficiency was detected. Eight of

14 patients had a Holo-TC below 21 pmol/l, five patients had a MMA level >0.45 µmol/l and in one

patient, both Holo-TC and MMA were abnormal (Table 3).

Laboratory results, including the time at which a deficiency was diagnosed, for the 14 patients with

vitamin B12

deficiency in group B are shown in Table 2. Five patients (5.6%) had vitamin B12

deficiency at

| Chapter 12

178

baseline. For the remaining nine patients, the median time to diagnosis of deficiency was 6.0 months

(range 3.5-9.9 months). One patient had severe anemia (Hb 4.8), whilst 6 patients suffered from mild

anemia (Hb from 7 to 8.4). There was no patient with megaloblastic anemia.

The estimated cumulative 1-year incidence of a vitamin B12

deficiency in group B was 22.9% (Figure

1). If patients who were already vitamin B12

deficient at baseline were excluded from analysis, the

estimated 1-year incidence was 18.2%.

TABLE 1. Baseline characteristics of 187 study patients, according to group (cross-sectional analysis (A) or longitudinal analysis (B)).

Group A (n=99) Group B (n=88)

Sex (male, %) 74% 72%

Age (mean, range) 62 (38-79) 63 (19-79)

No of observations12345

99 (100%)N/AN/AN/AN/A

22 (25.0%)18 (20.5%)14 (15.9%)17 (19.3%)17 (19.3%)

Length of follow-up (in days,median, IQR)

589 (412-916) 195 (48-329)

Baseline laboratory results(reference values)

Hemoglobin (mmol/l)(M:8.6-10.5, F: 7.5-9.5)

MCV (fl)(80-100)

Creatinin (umol/l)M: 65-115, F: 55-90)

Folate (nmol/l)(8.0-28.0)

Holo-Transcobalamin* (pmol/l)(>21.0)

MMA (µmol/l)(≤0.45)

Platelets (109/l)(150-370)

WBC (109/l)(3.5-10.0)

(median, range)

8.8 (5.8-10.4)

91 (70-103)

48 (77-148)

19.3 (3.9-45.4)

54 (10-128)

0.21 (0.07-2.87)

228 (123-391)

6.5 (2.8-12.4)

(median, range)

8.1 (5.5-9.8)

92 (70-120)

75 (47-214)

24.4 (9.0-54.4)

61 (20-128)

0.22 (0.02-0.63)

227.5 (16-449)

6.0 (3.8-14.4)

MCV: mean corpuscular volume, WBC: white blood count, MMA: methylmalonic acid N/A: not applicable*maximum possibly measured value of Holo-Transcobalamin was 128 pmol/l (upper linearity limit)

Vitamin B12


179

Cha

pter

12

TABLE 2. Laboratory results, timing of vitamin B12

deficiency, hemoglobin and MCV in individual patients who were diagnosed with vitamin B

12 deficiency during follow-up after esophageal resection with gastric tube

reconstruction for potentially curable esophageal carcinoma.

Number, M/FTime from operation to deficiency in months

Holo-TC (pmol/l) MMA (µmol/l) Hemoglobin

(mmol/l) MCV (fl)

Group A

1, M 12.0 24 0.56 8.0 89

2, M 13.9 19 0.31 10.4 95

3, M 14.4 35 0.59 8.2 88

4, M 14.5 39 0.49 9.1 98

5, M 15.6 15 0.2 10.2 91

6, F 19.1 14 0.27 8.2 99

7, M 19.5 20 0.21 10.2 90

8, M 22.5 22 2.87 9.9 99

9, M 29.7 10 2.28 8.4 92

10, M 33.1 13 0.35 8.4 90

11, M 35.1 26 0.67 8.4 88

Group B

1, M 0 33 0.59 7.0 85

2, M 0 66 0.63 8.0 90

3, M 0 55 0.59 9.0 98

4, M 0 20 0.24 8.7 84

5, M 0 20 0.28 7.7 83

6, M 3.5 20 0.25 4.8 87

7, M 3.7 16 0.15 9.2 89

8, M 4.6 25 0.66 8.4 87

9, M 4.9 17 0.34 8.5 92

10, M 6.0 18 0.36 8.3 96

11, F 6.3 34 0.50 7.9 88

12, M 6.5 18 0.19 8.3 86

13, F 8.9 10 0.63 7.5 96

14, M 9.9 16 0.12 9.3 94

M: Male, F: Female, Holo-TC: tolo-transcobalamin, MMA: methyl malonic acid, MCV: mean corpuscular volume

| Chapter 12

180

TABLE 3. Summary of patients in group B, tested at each measuring point during follow-up, including number of vitamin B

12 deficient patients at each point, according to definition of deficiency.

Observation moment 1 (baseline, n=88)

2(n=66)

3(n=49)

4(n=33)

5(n=18)

Total(n=88)

Holo-TC <21 pmol/l 2 3 1 2 0 8

MMA >0.45 µmol/l 3 0 2 0 0 5

Holo-TC <21 pmol/l and MMA >0.45 µmol/l

0 1 0 0 0 1

Total number of vitamin B12

deficiencies

5 4 3 2 0 14

Cumulative number of vitamin B12

deficiencies

5 9 12 14 14

Holo-TC: holo-transcobalamin MMA: methylmalonic acid

Follow-up in months129630

1 - c

umul

ative

dise

ase-

free

surv

ival

1,0

0,8

0,6

0,4

0,2

0,0

Page 1

No at risk 88 60 45 37 17

FIGURE 1. Kaplan Meyer curve (1-year disease-free survival function) for vitamin B12

deficiencies according to clinical algorithm in 88 patients that were followed during one year after esophageal resection with gastric tube reconstruction for potentially curable esophageal cancer.Estimated 1-year incidence: 22.9%

Multilevel regression analysis showed that the predicted mean Holo-TC level significantly decreased

during follow-up at 3, 6, 9 and 12 months after surgery when compared to baseline levels (p<0.001

for all time points) (Figure 2). The predicted lowest mean level of Holo-TC was at 11.4 months (50.0

pmol/l). For MMA levels, no statistically significant increase was seen when comparing the predicted

mean levels with baseline levels during the follow-up period (Figure 2).

Vitamin B12


181

Cha

pter

12

Follow-up 12 months9 months6 months3 monthsBaseline

Mea

n pr

edic

ted

valu

es o

f Hol

o-TC

70

65

60

55

50

Page 1

Follow-up 12 months9 months6 months3 monthsBaseline

Mea

n pr

edic

ted

valu

es o

f Hol

o-TC

70

65

60

55

50

Page 1

Follow-up12 months9 months6 months3 monthsBaseline

Mea

n pr

edic

ted

valu

es o

f MM

A

,27

,26

,26

,25

,25

,24

p=0.065

p=0.085

p=0.109

p=0.062

P<0.001

P<0.001 P<0.001

P<0.001

FIGURE 2. Mean predicted levels of Holo-TC (upper graph, in pmol/l) and MMA (lower graph, in µmol/l) during 1-year follow-up of 88 patients who underwent esophageal resection with gastric tube reconstruction for potentially curable esophageal cancer. P-values per time point are for change from baseline value.

When combining group A and B, a total of 25 of 187 (13.4%) patients were diagnosed with vitamin B12

deficiency. Fourteen of 88 (15.9%) patients were diagnosed in the first year (group B), 8 of 61 (13.1%)

during the second year (group A), and 3 of 38 (7.9%) during the third year after surgery (group A).

| Chapter 12

182

DISCUSSION

The present study shows that about 18% of patients subjected to esophagectomy for cancer (group

B) develops vitamin B12

deficiency during the first postoperative year. The prevalence of preoperative

vitamin B12

deficiency in this population was 5.6%. The prevalence within the second and third

postoperative year (group A) was 11.1%.

The percentage of patients with a vitamin B12

deficiency preoperatively, i.e. the estimated prevalence

in the non-operated population, is comparable with data from the general population.10-12 Although a

decreased preoperative intake of vitamin B12

might have occurred in these patients with esophageal

cancer (caused by e.g. dysphagia or horror carnis), this does not seem to affect the prevalence of

vitamin B12

deficiencies.

The absorption of vitamin B12

is a multifactorial process.2 After esophagectomy with gastric tube

reconstruction, absorption can be perturbed due to 1) reduced dietary intake of vitamin B12

as a result

of the change in anatomy leading to early satiety, 2) reduced digestion of protein-bound vitamin B12

from food due to reduced acid secretion and reduced length of time that the food is exposed to this

acid in the gastric conduit, and 3) reduced production of intrinsic factor due to a/ reduction of the

gastric mucosal surface, b/ gastritis due to biliary reflux or c/ bilateral truncal vagotomy.8-13 It is unclear

whether all three mechanisms play a role in the development of vitamin B12

deficiencies in patients

following esophagectomy, or if there is a single factor leading to vitamin B12

deficiency.

After bariatric surgery, where the anatomy of the upper gastrointestinal tract is deliberately altered to

reduce both intake and absorption, the prevalence of vitamin B12

deficiency two years after surgery

is between 4 and 36%. This is higher than in the general population.14 Vitamin B12

deficiency is more

frequently seen after a Roux-en-Y gastric bypass compared to a gastric sleeve resection.15 Vitamin

B12

deficiency seems to occur at a later time point than in patients after esophagectomy in the

current study population. In a large Dutch cohort of patients who underwent gastric bypass surgery,

vitamin B12

deficiencies were almost all diagnosed in the second postoperative year (mean time to

deficiency 14.1 months , CI 12.4-15.8 months).16 In the present study, the mean time to deficiency in

the longitudinally observed group was 6 months. This phenomenon is more clearly visible by the

early decrease in mean Holo-TC level during the first postoperative year that was detected in the

multilevel regression and the rise of MMA as an early metabolic marker for vitamin B12

deficiency

(Figure 2). The serum levels of both biochemical markers indicate that there is no stable vitamin B12

turnover after esophagectomy. It remains however unclear why deficiencies develop relatively early

after esophagectomy. Possibly, preoperative consumption of vitamin B12

-containing foods might

be decreased due to an obstructed food passage and/or chemoradiotherapy-induced nausea. This

might result in subnormal levels of vitamin B12

already preoperatively.

There is no worldwide consensus on the definition of vitamin B12

deficiency. Although many clinicians

still use total serum vitamin B12

to detect possible deficiencies, it has become more and more clear

that metabolically relevant deficiencies are better diagnosed by measuring Holo-TC and MMA

levels. Especially an increase in MMA should be considered a consequence of intracellular vitamin

Vitamin B12


183

Cha

pter

12

B12

shortage, and is therefore a more specific marker for metabolic deficiency. This is why MMA is

generally considered the golden standard for detecting effective vitamin B12

deficiency.17 Holo-TC is

considered an early marker of a negative vitamin B12 balance. When detected early, decrease in Holo-

TC and increased MMA may precede clinical manifestations of vitamin B12

deficiency, facilitating early

preventive measures. This might also be a reason why in the current study, no megaloblastic anemias

were found.18

In the present study, the focus was on the detection of metabolic vitamin B12

deficiencies using

laboratory testing. Symptoms that patients might have as a result of possible a vitamin B12

deficiency

were not an endpoint. Signs and symptoms of B12

deficiency are rather aspecific and may also develop

following neo-adjuvant and/or surgical treatment. Therefore, postoperative symptoms were not

recorded in this study.5,19

Because patients who underwent esophagectomy might not suffer from malabsorption, but

rather develop deficiencies due to insufficient dietary intake or reduced gastric acid secretion, oral

supplements might also be effective in the prevention and treatment of vitamin B12

deficiency in

this group.20 Studies have shown that high-dosed oral supplements can be as effective as parenteral

hydroxycobalamin in the treatment of vitamin B12

deficiencies.20,21 This may be better tolerated by the

patient and reduce costs.

The Dutch guidelines for the diagnosis and treatment of esophageal cancer do not give specific

recommendations (e.g. frequency and timing) for testing patients for vitamin B12

deficiency after

esophagectomy. The present study indicates that preoperatively and at least once postoperatively,

but not before one year after esophagectomy, assessment of vitamin B12

status is of use. Due to the

design of the study, it cannot be determined at what point in time patients in the cross-sectional

group have developed a deficiency. Therefore, repeated testing of patients with marginal Holo-TC and

MMA levels is recommended.

| Chapter 12

184

REFERENCES1. Stabler SP. Clinical practice. Vitamin B12 deficiency. N Engl J Med. 2013 Jan 10;368(2):149-60.

2. Nielsen MJ, Rasmussen MR, Andersen CB, Nexo E, Moestrup SK. Vitamin B12 transport from food to the body’s cells--a sophisticated, multistep pathway. Nat Rev Gastroenterol Hepatol. 2012 Jun;9(6):345-54.

3. Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: Sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2014 Sep 13.

4. Arnold M, Soerjomataram I, Ferlay J, Forman D. Global incidence of oesophageal cancer by histological subtype in 2012. Gut. 2014 Oct 15.

5. van Hagen P, Hulshof MC, van Lanschot JJ, et al. Preoperative chemoradiotherapy for esophageal or junctional cancer. N Engl J Med. 2012 May 31;366(22):2074-84.

6. Orringer MB, Sloan H. Substernal gastric bypass of the excluded thoracic esophagus for palliation of esophageal carcinoma. J Thorac Cardiovasc Surg. 1975 Nov;70(5):836-51.

7. Majumder S, Soriano J, Louie Cruz A, Dasanu CA. Vitamin B12 deficiency in patients undergoing bariatric surgery: preventive strategies and key recommendations. Surg Obes Relat Dis. 2013 Nov-Dec;9(6):1013-9.

8. Donington JS. Functional conduit disorders after esophagectomy. Thorac Surg Clin. 2006 Feb;16(1):53-62.

9. Heil SG, de Jonge R, de Rotte MC, et al. Screening for metabolic vitamin B12 deficiency by holotranscobalamin in patients suspected of vitamin B12 deficiency: a multicentre study. Ann Clin Biochem. 2012 Mar;49(Pt 2):184-9.

10. Allen LH. How common is vitamin B-12 deficiency? Am J Clin Nutr. 2009 Feb;89(2):693S-6S.

11. Clarke R, Grimley Evans J, Schneede J, et al. Vitamin B12 and folate deficiency in later life. Age Ageing. 2004 Jan;33(1):34-41.

12. Herrmann W, Schorr H, Obeid R, Geisel J. Vitamin B-12 status, particularly holotranscobalamin II and methylmalonic acid concentrations, and hyperhomocysteinemia in vegetarians. Am J Clin Nutr. 2003 Jul;78(1):131-6.

13. Okada N, Nishimura O, Sakurai T, Tsuchihashi S, Juhri M. Gastric functions in patients with the intrathoracic stomach after esophageal surgery. Ann Surg. 1986 Aug;204(2):114-21.

14. Stein J, Stier C, Raab H, Weiner R. Review article: The nutritional and pharmacological consequences of obesity surgery. Aliment Pharmacol Ther. 2014 Sep;40(6):582-609.

15. Kwon Y, Kim HJ, Lo Menzo E, Park S, Szomstein S, Rosenthal RJ. Anemia, iron and vitamin B12 deficiencies after sleeve gastrectomy compared to Roux-en-Y gastric bypass: a meta-analysis. Surg Obes Relat Dis. 2014 Jul-Aug;10(4):589-97.

16. van der Beek ES, Monpellier VM, Eland I, Tromp E, van Ramshorst B. Nutritional deficiencies in gastric bypass patients; incidence, time of occurrence and implications for post-operative surveillance. Obes Surg. 2015 May; 25(5):818-823.

17. Allen RH, Stabler SP, Savage DG, Lindenbaum J. Diagnosis of cobalamin deficiency I: usefulness of serum methylmalonic acid and total homocysteine concentrations. Am J Hematol. 1990 Jun;34(2):90-8.

18. Carmel R. Current concepts in cobalamin deficiency. Annu Rev Med. 2000;51:357-75.

19. Vidal-Alaball J, Butler CC, Cannings-John R, et al. Oral vitamin B12 versus intramuscular vitamin B12 for vitamin B12 deficiency. Cochrane Database Syst Rev. 2005(3):CD004655.

20. Hjelms E, Thirup P, Schou L. Gastric intrinsic factor production and vitamin B12 absorption after oesophageal resection using stomach as substitute. Eur J Cardiothorac Surg. 1999 Sep;16(3):273-5.

21. Kuzminski AM, Del Giacco EJ, Allen RH, Stabler SP, Lindenbaum J. Effective treatment of cobalamin deficiency with oral cobalamin. Blood. 1998 Aug 15;92(4):1191-8.

PART IIISummary and appendices

Summary of the thesis

Samenvatting in het Nederlands

Future perspectives

PhD Portfolio

Acknowledgements

Curriculum Vitae

Summary of the thesis |

189

SUMMARY OF THE THESIS

Esophageal cancer is a highly aggressive disease with a rising incidence, especially in the Western

World. Almost half of the patients who present with esophageal cancer are beyond curative therapy,

and for patients with potentially curable disease, 5-year survival rates rarely exceed 30%. Although

surgical resection is the most important part of the treatment, evidence in favor of neoadjuvant

treatment (i.e. chemo(radio)therapy prior to surgery) is accumulating. These multimodality treatment

regimens can induce shrinkage of the tumor and thereby facilitate a microscopically radical resection,

which is an important prognostic factor for survival. However, this potential benefit should outweigh

the toxicity that neoadjuvant treatment adds to the already high morbidity and mortality that are

associated with the curative resection of esophageal cancer.

This thesis consists of two parts. In part one: “Preoperative analysis and treatment” the benefits of

neoadjuvant chemoradiotherapy (nCRT) are described and the impact of several (pre-)clinical delays

in the treatment are evaluated. Part two: “Postoperative outcome and prognostication” focusses on

(disease-free) survival and prognostication after potentially curative treatment.

Part 1: Preoperative analysis and treatment

Over the past decades, evidence for the positive effect of nCRT in patients with esophageal cancer

has accumulated, especially in meta-analyses. However, due to various reasons, in most randomized

trials no survival benefit could be demonstrated. In chapter 2, the results of a multicenter randomized

trial which compared nCRT followed by surgery with surgery alone in patients with potentially curable

esophageal or esophagogastric junction cancer are described after a minimal follow-up of two years.

This trial, in which 178 patients were allocated to the nCRT arm and 188 to the surgery alone arm,

showed a statistically significant estimated survival benefit at five years of 13%. This might be due to

the high rate of microscopically radical resections, which was 92% in the nCRT arm versus 79% in the

surgery alone arm. The toxicity of the nCRT was acceptable, and postoperative morbidity and mortality

were comparable between groups. The overall rate of pathologically complete response (pCR) was

29%, but with 49% especially high in patients with squamous cell carcinoma. This large randomized

trial confirmed the growing support for incorporating nCRT in the treatment of esophageal cancer.

To affirm the results that were found in chapter 1, a second analysis was performed after a longer

period of follow-up. In chapter 3, these results are shown. After a minimal follow-up of five years for

each patient, the 5-year survival rate differed 14% in favor of nCRT, which was statistically significant.

Also, there was a statistically significant difference in 5-year progression-free survival of 17% in favor

of the nCRT arm. These results show that the effect of nCRT on overall and progression-free survival is

sustained during a longer period of time and nCRT followed by surgery should be accepted as a new

standard of care.

The interval between nCRT and surgery (time to surgery, TTS) can be prolonged by various factors.

One of the reasons of such delay might be, that the nCRT itself can cause a prolonged TTS due

| Summary of the thesis

190

to toxic side-effects, having an effect on patients’ nutritional status or causing hematological or

infectious complications. A prolonged TTS might theoretically cause a more difficult resection with

more complications, or even outgrowth of residual tumor. On the other hand, it might also result in

more pathologically complete responders, as was shown earlier in patients treated witch nCRT for

rectal cancer. In chapter 4, the impact of a prolonged TTS was investigated retrospectively in 325

patients who were treated with nCRT followed by surgery within a ten year period. Prolonged TTS

was associated with an increased probability of pCR. No effect on overall or progression-free survival

could be demonstrated, however, there was a slightly increased risk for postoperative complications

in patients who had a TTS of more than 45 days. Prolonging TTS up to at least twelve weeks could

safely be done. This allows for an active surveillance strategy to be tested in patients who show no

clinical signs of residual tumor after nCRT.

An additional relevant time interval can be recognized between the diagnosis and actual treatment

of esophageal cancer. In chapter 5, the impact of this so called hospital delay was evaluated. It

was found that a prolonged hospital delay had a negative impact on postoperative morbidity and

mortality, resulting in mortality rates up to 8.5% and morbidity rates up to 67.2% in patients with a

delay of more than 8 weeks. However, between all the evaluated time intervals that were tested, no

difference in overall or progression-free survival could be observed. The interval between onset of

complaints and diagnosis (the pre-hospital delay) had no influence on short- or long-term outcome

of patients.

Before the start of any treatment, the patients should be presented and discussed in a multidisciplinary

tumor board (MDT). During this MDT, a personalized definitive treatment plan is established, in which

the appropriate medical, radiotherapeutic and surgical treatment is determined. In chapter 6, the

value of the MDT in a tertiary referral center was evaluated. Over a period of eight months, 171

patients were discussed in 34 MDTs. In more than one third of discussed treatment cases, a substantial

alteration was made in the proposed treatment plan. These results demonstrate the value of a regular

MDT for esophageal cancer and support the Dutch guidelines, in which MDTs are mandatory.

In chapter 7, the HER2/neu status of esophageal adenocarcinoma biopsies of patients who underwent

primary surgery was compared with the HER2/neu status in the primary tumor as determined by three

dedicated pathologists in a multicenter retrospective study. HER2/neu overexpression was found in

22% of the primary tumors. In 93% of these cases, the reported HER2/neu status was concordant with

that in the preoperatively obtained biopsy. Inter-observer agreement amongst all three pathologists

was moderate, especially in the interpretation of the in-situ hybridization results.

Part 2: Postoperative outcome and prognostication 18F-Fluorodeoxyglucose positron emission tomography (FDG-PET) has been investigated as a tool

for monitoring response to neoadjuvant therapy and as a predictor for survival in patients with

esophageal cancer. If non-responders could be identified early during therapy, possible toxicity could

be avoided by proceeding to surgery earlier. In chapter 8, no correlation could be demonstrated

Summary of the thesis |

191

between the change in FDG-PET signal that was measured early during (i.e. two weeks after the

start of ) nCRT and long-term survival. This implies that FDG-PET cannot be used for identifying non-

responders early during nCRT.

Although long-term survival is improving, due to advances in both medical, radiotherapeutic and

surgical treatment, esophageal cancer is still lethal in more than half of the patients initially considered

potentially curable. This implies that multimodality treatment is failing in a large proportion of patients.

To better understand why nCRT followed by surgery is not a guarantee for definitive cure, the pattern

of recurrence in 213 patients who underwent this treatment are described in chapter 9. After nCRT,

significantly less locoregional recurrences were observed when compared to patients treated with

surgery only. A lesser, although still statistically significant, effect was seen for peritoneal metastases

and hematogenous dissemination. Moreover, locoregional recurrences within the radiation field

occurred in only 5% of cases, emphasizing the high locoregional control that is achieved with nCRT.

However, it is known that even in patients who showed pCR (i.e. the best objective pathological result

of nCRT), recurrent disease can develop. In chapter 10, the recurrence pattern of 62 patients with pCR

was compared with that of 126 patients with non-pCR. As expected, significantly more recurrences

occurred in the non-pCR group, however, still 32% of patients with pCR developed a distant metastasis

and 13% suffered from a locoregional recurrence. A reason for this unexpectedly high number

of recurrences might be a sampling error at pathological examination of the resection specimen,

resulting in a falsely negative pathological outcome. This might explain why in multivariable analysis

only lymph node involvement was found to be an independent prognosticator for survival.

In chapter 11, the prognostic power of the 7th edition of the UICC-AJCC TNM classification for solid

tumors was compared to that of the 6th edition in patients with esophageal adenocarcinoma. This

study confirmed that application of the 7th staging system results in a better prognostic stratification

of overall survival compared to the 6th edition. Even though the 7th edition of the TNM classification

was derived from a large cohort of patients with predominantly squamous cell carcinoma who

underwent a transthoracic esophagectomy, it also performs better in patients with adenocarcinoma

who underwent a transhiatal surgical approach, supporting its generalizability for different esophageal

cancer practices.

During esophageal resection, the anatomy of the upper gastrointestinal tract is changed profoundly,

which may lead to problems in the intake and absorption of vitamin B12, a vitamin essential for DNA

synthesis and neural function. In Chapter 12, the incidence, prevalence and timing of vitamin B12

deficiencies was investigated in patients who underwent esophageal resection for potentially curable

esophageal cancer. This analysis showed that during the first postoperative year, the estimated

incidence of vitamin B12

deficiencies is 18.2%. This is remarkably higher than in the general population,

and moreover, these deficiencies developed earlier than expected. Early and routine testing for

vitamin B12

deficiencies is therefore advised during follow-up after esophagectomy.

Samenvatting in het Nederlands |

193

SAMENVATTING

Het slokdarmcarcinoom is een zeer agressieve vorm van kanker waarvan de incidentie nog altijd

toeneemt, met name in de Westerse wereld. Van alle patiënten die zich presenteren met slokdarmkanker

komt ongeveer de helft niet meer in aanmerking voor een in opzet curatieve behandeling, en voor

patiënten met een in potentie curabel carcinoom ligt de 5-jaars overleving zelden boven de 30%.

Hoewel chirurgische resectie de belangrijkste vorm van behandeling betreft, is er groeiend bewijs

voor de effectiviteit van neoadjuvante behandeling, d.w.z. behandeling met chemo(radio)therapie

voorafgaand aan resectie. Deze neoadjuvante behandelingsstrategieën kunnen een reductie

van tumorvolume tot gevolg hebben waardoor microscopisch radicale resectie, een belangrijke

prognostische factor, waarschijnlijker wordt. Echter, om deze neoadjuvante therapieën te kunnen

toepassen moet dit potentiele voordeel opwegen tegen de toxiciteit die zij toevoegen aan de al

hoge morbiditeit en mortaliteit die waarmee de curatieve resectie van slokdarmkanker gepaard gaat.

Dit proefschrift bestaat uit twee delen. In deel één: ‘Preoperatieve analyse en behandeling’ worden

de voordelen van neoadjuvante chemoradiotherapie (nCRT) beschreven, alsmede de invloed die

verschillende (pre)klinische vertragingen in de behandeling hebben op de uitkomst van patiënten.

Deel twee: ‘Postoperatieve uitkomst en prognosticering’ richt zich op de (ziektevrije-) overleving en

de prognose na in opzet curatieve behandeling van slokdarmkanker.

Deel 1: Preoperatieve analyse en behandeling

Gedurende de laatste decennia heeft het bewijs voor de effectiviteit van nCRT zich opgestapeld,

met name door de positieve resultaten die volgden uit meta-analyses. Om verschillende redenen

kon er in de meeste gerandomiseerde trials echter geen overlevingsvoordeel worden bewezen.

In hoofdstuk 2 worden de resultaten van een multicentrisch gerandomiseerd onderzoek

beschreven waarin nCRT gevolgd door chirurgische resectie werd vergeleken met chirurgie alleen

als behandeling voor potentieel curabel slokdarmcarcinoom of carcinoom van de slokdarm-maag

overgang na een minimale follow-up van 2 jaar. In dit onderzoek, waarin 178 patiënten werden

gerandomiseerd voor nCRT + chirurgie en 188 voor chirurgie alleen, werd een statistisch significant

geschat overlevingsvoordeel na vijf jaar van 13% gevonden. Deze winst in overleving zou kunnen

worden veroorzaakt door het grote verschil in microscopisch radicale resecties, dat 92% was in de

nCRT groep tegenover 79% in de chirurgie alleen groep. De toxiciteit van de nCRT bleek acceptabel,

en de postoperatieve morbiditeit en mortaliteit waren vergelijkbaar tussen de groepen. Het

percentage patiënten met een pathologisch complete respons (geen vitale tumorcellen in het gehele

resectiepreparaat, pCR) was 29%, maar was met name hoog (49%) in de groep patiënten met een

plaveiselcelcarcinoom. Dit grote gerandomiseerde onderzoek ondersteunt het gebruik van nCRT bij

de behandeling van slokdarmcarcinoom.

Ter bevestiging van de resultaten die werden gevonden in hoofdstuk 1 werd een tweede analyse

uitgevoerd na een lagere periode van follow-up. De resultaten van deze analyse worden beschreven

| Samenvatting in het Nederlands

194

in hoofdstuk 3. Na een minimale follow-up van 5 jaar van alle geïncludeerde patiënten was het

verschil in 5-jaars overleving 14% ten faveure van de nCRT groep. Ook werd er een significant verschil

gevonden in de ziektevrije 5-jaars overleving van 17% in het voordeel van de nCRT groep. Deze

resultaten laten zien dat het effect op de totale en ziektevrije overleving gedurende een langere

periode aanhoudt en dat nCRT gevolgd door chirurgie als een nieuwe standaardtherapie zou moeten

gelden voor het potentieel curabele slokdarmcarcinoom.

Het tijdsinterval tussen nCRT en chirurgie (tijd tot chirurgie, TTC) kan door verschillende factoren

worden verlengd. Eén van de oorzaken van een dergelijke vertraging zou kunnen zijn, dat de nCRT zelf

hieraan debet is, vanwege het feit dat door de toxiciteit patiënten in een slechte voedingstoestand

komen, of moeten herstellen van hematologische of infectieuze complicaties. Een langere TTC kan

theoretisch een moeilijkere chirurgische resectie tot gevolg hebben met een bijkomende toename

van het aantal complicaties, of zelfs tot progressie van de tumor. Anderzijds kan het ook een hoger

percentage patiënten met pCR tot gevolg hebben, zoals eerder werd aangetoond bij patiënten met

een rectumcarcinoom. In hoofdstuk 4 werd de invloed van een langere TTC retrospectief onderzocht

bij 325 patiënten die werden behandeld met nCRT gevolgd door chirurgie over een periode van

tien jaar. Een langere TTC was geassocieerd met een hogere kans op pCR, er kon echter geen effect

worden aangetoond op de ziektevrije of totale overleving . Wel was er een licht verhoogd risico

op postoperatieve complicaties bij patiënten die een TTC van meer dan 45 dagen hadden. De TTC

kan veilig worden verlengd tot ten minste twaalf weken. Hierdoor kan een strategie van ‘waakzame

surveillance’ worden onderzocht bij patiënten die na nCRT geen aanwijzingen hebben voor residu

van vitale tumor.

Nog een ander tijdsinterval kan kan een rol spelen, namelijk het tijdsinterval tussen de diagnose

en de behandeling van slokdarmkanker. De invloed van dit zogenaamde ‘ziekenhuis-delay’ werd

geëvalueerd in hoofdstuk 5. Hierbij bleek dat een langer ziekenhuis-delay een negatieve invloed

heeft op de postoperatieve uitkomst, resulterend in 8,5% postoperatieve mortaliteit en 67,2%

postoperatieve morbiditeit bij patiënten met een ziekenhuis-delay van meer dan 8 weken. Een

significant verschil in (ziektevrije-)overleving kon niet worden aangetoond. Het tijdsinterval tussen

het begin van symptomen en het stellen van de diagnose slokdarmcarcinoom bleek niet aantoonbaar

van invloed te zijn op de korte- of lange-termijn uitkomst van patiënten.

In de tijd tussen diagnose en behandeling dient iedere patiënt met slokdarmcarcinoom besproken

te worden in een multidisciplinaire tumorbespreking (MDT). Tijdens dit MDT wordt voor de patiënt

een definitief individueel behandelplan vastgesteld met een heldere beschrijving van geschikte

behandeling. De waarde van dit MDT werd onderzocht in hoofdstuk 6. Hiertoe werd de uitkomst

van het MDT bijgehouden gedurende een periode van 8 maanden, waarin 171 patiënten werden

besproken tijdens 34 bijeenkomsten. In meer dan een derde van de besproken ziektegevallen werd

een substantiële wijziging aangebracht in het voorgestelde behandelplan. Dit resultaat bewijst de

waarde van een wekelijks MDT en ondersteunt de Nederlandse richtlijn, waarin een MDT verplicht

wordt gesteld.

Samenvatting in het Nederlands |

195

In hoofdstuk 7 wordt de HER2/neu status van preoperatief afgenomen biopten van het slokdarm

adenocarcinoom van niet voorbehandelde patiënten vergeleken met die van de primaire tumor in

het resectiepreparaat. Dit vond plaats in het kader van een multicentrisch retrospectief onderzoek

waarbij alle preparaten door drie gespecialiseerde pathologen werden beoordeeld. Overexpressie van

HER2/neu was aanwezig in 22% van de tumoren. In 93% van de gevallen kwam de HER2/neu status in

het biopt overeen met die in de primaire tumor. De overeenstemming tussen de drie beoordelende

pathologen was matig, met name waar het de beoordeling van de in-situ-hybridisatie betrof.

Deel 2: Postoperatieve uitkomst en prognosticering.18F-Fluorodeoxyglucose positron emissie tomografie (FDG-PET) werd eerder onderzocht als techniek

om respons op neoadjuvante therapie te monitoren en overleving te voorspellen bij patiënten met

een slokdarmcarcinoom. Als hiermee niet-responderende patiënten vroeg tijdens de therapie zouden

kunnen worden geïdentificeerd, zou de voorbehandeling voortijdig kunnen worden afgebroken,

waardoor eventuele toxiciteit van de therapie zou kunnen worden voorkomen en patiënten eerder

in aanmerking zouden kunnen komen voor chirurgie. In hoofdstuk 8 kon geen correlatie worden

aangetoond tussen de afname in intensiteit van het FDG-PET signaal dat snel na het starten van nCRT

werd gemeten en de lange-termijn overleving. Hieruit kan worden opgemaakt dat FDG-PET niet kan

worden gebruikt voor het identificeren van non-responders kort na het starten van nCRT.

Hoewel de lange-termijn overleving van patienten met een slokdarmcarcinoom verbetert, als

gevolg van ontwikkelingen in de medische, radiotherapeutische en chirurgische behandeling, is de

ziekte toch nog dodelijk in meer dan de helft van de patiënten die initieel als potentieel curabel

werden beoordeeld. De huidige behandelingen falen bij een groot deel van de gevallen. Om beter

te begrijpen waarom de behandeling van patiënten met nCRT gevolgd door chirurgie geen garantie

is voor curatie, werd het patroon van recidivering van de ziekte in kaart gebracht bij 213 patiënten

die een dergelijke behandeling ondergingen. De resultaten van dit onderzoek worden beschreven in

hoofdstuk 9. Na nCRT werden significant minder locoregionale recidieven gevonden dan na chirurgie

alleen. Een minder sterkeffect, hoewel toch statistisch significant, werd gezien voor het optreden

van peritoneale en hematogene metastasering. Bovendien traden slechts 5% van de locoregionale

recidieven op binnen de grenzen van het bestralingsveld, hetgeen de hoge mate van locoregionale

controle benadrukt die wordt bereikt met nCRT gevolgd door chirurgie.

Hoewel een pCR het beste objectieve pathologische resultaat van nCRT is, treden ook bij deze groep

patiënten recidieven op. In hoofdstuk 10 werd het patroon van recidivering van 62 patiënten met

een pCR vergeleken met dat van 126 patiënten zonder pCR die dezelfde nCRT ondergingen. Zoals

verwacht werden er significant minder recidieven gevonden in de pCR groep, echter, bij 32% werd

een metastase op afstand gevonden en bij 13% een locoregionaal recidief. Een verklaring voor dit

hoge aantal recidieven kan wellicht worden gevonden in het feit dat er tijdens de resectie door de

chirurg of tijdens de beoordeling door de patholoog niet-representatief materiaal wordt uitgenomen

respectievelijk beoordeeld hetgeen resulteert in een fout-negatieve pathologische uitkomst. Dit

| Samenvatting in het Nederlands

196

zou verklaren waarom tijdens multivariabele analyse alleen de lymfklierstatus als onafhankelijke

voorspeller voor overleving werd gevonden.

In hoofdstuk 11 werd de prognostische waarde van de zevende editie van de UICC-AJCC TNM

classificatie voor het slokdarmcarcinoom vergeleken met de zesde editie in een cohort patiënten met

slokdarm adenocarcinoom. Dit onderzoek toonde aan dat gebruik van de zevende editie een betere

prognostificatie van totale overleving tot gevolg heeft vergeleken met het gebruik van de zesde editie.

Hoewel de TNM classificatie werd ontwikkeld met gebruikmaking van een groot cohort patiënten

waarin de meeste patiënten een plaveiselcelcarcinoom hadden en werden behandeld met een

transthoracale resectie, presteerde de zevende editie ook beter als het werd toegepast op patiënten

met een adenocarcinoom die een transhiatale resectie ondergingen, wat de bruikbaarheid van het

systeem bewijst wanneer het wordt toegepast bij andere behandelingen van het slokdarmcarcinoom.

Na een oesofagusresectie is de anatomie van het bovenste deel van de tractus digestivus dusdanig

veranderd dat functionele problemen te verwachten zijn. Eén van deze problemen kan zich voordoen

in de inname en absorptie van vitamine B12, een vitamine die essentieel is voor de opbouw van DNA

en het functioneren van neuronen. In hoofdstuk 12 werd de incidentie, prevalentie en timing van

het voorkomen van vitamine B12

deficiënties bij patienten die een oesofagusresectie ondergingen

onderzocht. Uit deze analyse volgde dat de geschatte 1-jaars incidentie van vitamine B12

deficiënties

na oesofagusresectie 18,2% is. Dit is hoger dan in de algemene populatie, en bovendien ontwikkelen

deze deficiënties zich relatief vroeg na de operatie. Het verdient dus aanbeveling om bij patiënten die

een oesofagusresectie hebben ondergaan routinematig en in een vroeg stadium onderzoek te doen

naar een mogelijke vitamine B12

deficiëntie.

Future perspectives |

197

FUTURE PERSPECTIVES

Historically, esophageal cancer has always been a malignancy that is accompanied by high morbidity

and a poor prognosis. Medical research that was performed over the last decades has improved

survival and reduced morbidity significantly by improving patient selection, surgical techniques and

optimizing postoperative care. Despite these improvements, esophageal cancer is still highly lethal,

especially in advanced cases. Recent developments, especially in multimodality modality treatment,

have shown that there is still room for more improvement. This improvement can be expected in

the fields of screening for esophageal cancer, patient selection for more individualized therapy, and

further development of (neo)adjuvant and targeted therapies. In this paragraph, possible options

for diagnosis and treatment are described on which future experimental and clinical research should

focus.

Screening for esophageal cancer

Early stages of esophageal cancer have a better prognosis than advanced stages. It would therefore

be favorable to detect tumors in an early or premalignant (Barrett’s esophagus) stage, also making less

invasive therapies (e.g. endoscopic resection) possible which are accompanied with less morbidity.

Unfortunately, the current modality to detect esophageal cancer (endoscopy with biopsies)

is not sensitive enough, diminishing the diagnostic yield of a screening program. Also, it can be

accompanied by morbidity. This makes a screening program too hazardous and not cost-effective.

However, if other means of detecting esophageal cancer (preferably by using highly sensitive and

specific biomarkers that can be collected non-invasively) could be found, a screening program in

selected groups of patients might be opportune.1

Patient selection for neoadjuvant and targeted therapy

Neoadjuvant chemoradiotherapy has a well-established survival benefit in potentially curable

esophageal cancer patients. Although the toxicity of the treatment is acceptable in large groups of

patients, it is certainly not a therapy that one should be subjected to without an evident indication.

Short term toxicity (e.g. hematologic toxicity of chemotherapy) can delay further treatment and

potentially negatively affect survival, while long term toxicity (e.g. late radiotherapy effects) might

cause serious illness later in life. Therefore, it is of great importance to identify patients who will

potentially benefit from the treatment, but it maybe even more important to identify those who

will suffer harm from the treatment without gaining any profit. Pretreatment pathological findings

can already identify patients who are less likely to benefit from multimodality treatment (e.g. signet

cells, poorly differentiated adenocarcinomas), but these factors are not strong enough predictors to

withhold patients a potentially curative treatment. Genetic profiling (of patient, tumor or both) could

be used to distinguish between responders and non-responders, prior to the start of therapy. In breast

cancer, such a method is already in use, which identifies patients who will benefit from adjuvant

chemotherapy.2

| Future perspectives

198

Response evaluation during neoadjuvant therapy using nuclear imaging techniques have not

been proven to successfully predict response when used early during chemoradiotherapy. The

inflammatory response that is caused by radiotherapy might be interfering with the signal of the

tumor, thereby masking a metabolic response. New nuclear tracers, with more specific affinity for the

tumor might help to overcome this problem, making it possible to stop neoadjuvant therapy and

proceed to surgery in patients who show no response or even progression. However, such tracers are

still under investigation, and more clinical evidence should be gathered before such a strategy can be

confidently followed.3

After neoadjuvant therapy, a substantial number of patients have no detectable vital tumor cells in

the resection specimen, a so called pathologically complete response (pCR). It is questionable if these

patients should have undergone surgical resection, the tumor might well be fully sterilized by the

neoadjuvant therapy and esophagectomy might only induce morbidity and even mortality, without

adding to survival. These patients could therefore potentially benefit from an active surveillance

strategy, in which patients with a clinically complete response are closely monitored and only operated

upon when there is evidence of residual or recurrent disease. Unfortunately, it is not yet clear if a pCR

can be accurately predicted by performing clinical investigations. Ongoing and future studies will

provide the answer to this question, but until then, surgery will remain the cornerstone of treatment

in potentially curable esophageal cancer. (referentie pre-SANO protocol, submitted BMC cancer)

Improving multimodality treatment

Although neoadjuvant chemo(radio)therapy has been proven to improve radicality of resection

and long-term survival, still more than half of the patients with esophageal cancer that are treated

with curative intent develop recurrent disease. To date, there is no cure for these patients. Therefore,

prevention of recurrences seems to be the key in further improving survival in patients with potentially

curable esophageal cancer. To do so, several strategies could be investigated.

First, intensifying neoadjuvant therapy by either expanding radiotherapy fields, increasing radiotherapy

dosage (e.g. adding a so-called booster), or improving neoadjuvant chemotherapy component of

the neoadjuvant treatment might decrease the number of outfield locoregional recurrences, infield

locoregional recurrences, and hematogenous dissemination, respectively. Especially the shift from

locoregional recurrence to hematogenous dissemination that is clinically observed after neoadjuvant

chemoradiotherapy followed by esophagectomy is a target for further improving multimodality

treatment.

Secondly, adjuvant chemotherapy should be considered to be implemented in the treatment of

esophageal cancer, like it already is standard of care in many other malignancies (e.g. breast and colon

cancer). Especially in esophageal adenocarcinoma, which is less responsive to chemoradiotherapy

than squamous cell carcinoma, adjuvant chemotherapy might be a strategy to reduce hematogenous

dissemination. It goes without saying that further investigations on this subject have to be performed

in phase III trials before implementation can be recommended.

Future perspectives |

199

A third option for further investigation is the potential of so called targeted therapies, such as immuno-,

hormonal-, and molecular therapies. Over the past decades, the molecular characteristics of tumors

and their relation with tumor behavior have become more and more clear. Several agents have already

been proven to be effective in other malignancies (e.g. trastuzumab in HER2 positive breast cancer

and cetuximab in colorectal and head-and-neck cancer) , and some have even been investigated

in esophageal cancer.4 However, most of these therapies are only potentially beneficial for (small)

subgroups of patients, which makes further experimental and translational research so important,

hopefully leading to the discovery of more potent targets for therapy in esophageal cancers.

However, perhaps the key to surviving esophageal cancer not only lies in preventing recurrences, but

in treating them as well. For instance, surgical resection or radio frequent ablation of liver metastases

was already found to be a curative therapy in patients with disseminated colorectal cancer, and might

be an option in esophageal cancer as well.5

Although the recent developments in the multimodality treatment of esophageal carcinoma are

substantial, they are only just another step forward, and many more steps have to be taken on the

road ahead to fight this truly malignant disease.

1. Kumar S, Huang J, Abbassi-Ghadi N, Mackenzie HA, Veselkov KA, Hoare JM, et al. Mass Spectrometric Analysis of Exhaled Breath for the Identification of Volatile Organic Compound Biomarkers in Esophageal and Gastric Adenocarcinoma. Ann Surg. 2015 Jan 8.

2. van ‘t Veer LJ, Dai H, van de Vijver MJ, He YD, Hart AA, Mao M, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature. 2002 Jan 31;415(6871):530-6.

3. Jiang M, Huang Q, Chen P, Ruan X, Luo Z, Zhao L, et al. Monitoring the early biologic response of esophageal carcinoma after irradiation with 18F-FLT: an in-vitro and in-vivo study. Nucl Med Commun. 2014 Dec;35(12):1212-9.

4. Kasper S, Schuler M. Targeted therapies in gastroesophageal cancer. Eur J Cancer. 2014 May;50(7):1247-58.

5. Tomlinson JS, Jarnagin WR, DeMatteo RP, Fong Y, Kornprat P, Gonen M, et al. Actual 10-year survival after resection of colorectal liver metastases defines cure. J Clin Oncol. 2007 Oct 10;25(29):4575-80

PhD Portfolio |

201

PHD PORTFOLIO

Summary of PhD training and teaching

Name PhD student: Pieter van Hagen PhD period: Jan 2009 – Jan 2016

Erasmus MC Department: Surgery Promotor: Prof. dr. J.J.B. van Lanschot

Supervisor: Dr. B.P.L. Wijnhoven

PHD TRAINING Year Workload (ECTS)

General Courses

BROK (Basiscursus Regelgeving Klinisch Onderzoek) 2010 1.0

Introduction to biomedical research 2012 0.9

Biostatistics for clinicians 2012 1.0

Seminars and Workshops

Course Biomedical Research Techniques 2009 0.3

Methodologie van patiëntgebonden onderzoek en 2009 0.3

voorbereiding van subsidieaanvragen.

Systematisch literatuuronderzoek in databases 2012 0.3

Journal Club 2009-2012 2.0

Presentations

National conferences 2009 1.0


International conferences 2010 5.0






2. TEACHING Year Workload (ECTS)

Lecturing for students and nurses at Department of Surgery 2010-2012 1.0

Examination of Basis Life Support of medical students 2009-2012 1.0

Acknowledgements |

203

ACKNOWLEDGEMENTS

Een proefschrift behoeft een dankwoord. Niet zelden is het namelijk het enige hoofdstuk dat écht

wordt gelezen. Dus om er zeker van te zijn dat ook dit proefschrift wordt gelezen spreek ik hieronder

mijn dank uit aan allen die bij naam genoemd zijn en aan allen die ik mogelijk ben vergeten te

noemen. Echter, om te beginnen wil ik mijn dank uitspreken aan alle patiënten die als proefpersoon

hebben meegewerkt aan de verschillende studies die in dit proefschrift beschreven staan. De cijfers

en lijnen in tabellen en grafieken staan voor mensen van wie ik de naam niet ken. Aan hen als eersten

veel dank.

Geachte promotor prof. dr. J.J.B. van Lanschot, beste professor, hartelijk dank voor de kansen die u

mij hebt geboden. Toen ik bij u solliciteerde als promovendus drukte u mij op het hart dat het toch

echt de bedoeling was om in ieder geval drie jaar vol te maken. U zal het met mij eens zijn dat ik

die afspraak ruimschoots ben nagekomen. Het zijn bewogen jaren geweest, met vele hoogte- en

dieptepunten. Ik heb er ongelooflijk veel van geleerd. Uw messcherpe kritieken hebben de artikelen

die dit proefschrift vormen altijd naar een hoger niveau getild. Hartelijk dank voor het vertrouwen en

uw begeleiding!

Geachte co-promotor dr. B.P.L. Wijnhoven, beste Bas, dankjewel voor het enthousiasmeren, corrigeren,

meedenken en sturen, met de VIDEO-studie als kroon op het werk. Maar bovenal dank voor de

gezelligheid. Onze trips naar Leuven zal ik niet gauw vergeten, net zoals de ski-reizen, congressen,

borrels en ‘pauzes’ in de buitenlucht. Ik wens je al het goede!

Geachte prof. dr. E.W. Steyerberg, beste Ewout, veel dank voor je inzet voor verschillende projecten,

bovenal voor de CROSS-studie. Hoewel ik nachtmerries over interactietermen heb gehad wist jij

met een simpele email me toch weer op het rechte statistische pad te krijgen. Dank voor het zitting

nemen in de leescommissie.

Geachte prof. dr. A.H.J. Matthijssen en prof. dr. R. van Hillegersberg, hartelijk dank voor het zitting

nemen in de kleine commissie en het beoordelen van het manuscript.

Dr. A. van der Gaast, beste Ate, jij hebt ‘jouw’ CROSS-studie laten opschrijven door de meest onervaren

arts-onderzoeker die men op dat moment kon vinden. Hoewel we vrijwel alle data cijfer voor cijfer

samen hebben doorgesproken, moet het je toch enigszins benauwd hebben. Gelukkig liepen de

curves gewoon uiteen! Heel veel dank voor al je vertrouwen, nuchterheid en rust.

Geachte prof. dr. F.J. van Kemenade en dr. M.C.C.M. Hulshof, hartelijk dank voor het zitting nemen in

de grote commissie.

| Acknowledgements

204

Lieve Brechtje Grotenhuis, wat een feest om weer met je te mogen werken in het IJsselland!

Chardonnay! Maar wat hebben we ook tijdens onze onderzoekerstijd een lol gehad. Als ik ooit twijfel

over een pathologische q dan weet ik je te vinden! Kip Siam!

Joel Shapiro. Jij in het lab, ik in de Z-flat, en de mentor ergens tussenin. Je bent vervelend slim, maar

dat gaat gelukkig niet ten koste van je karakter. Dank voor je humor , je hulp en je vriendschap. We

maken er op 3 juni een groot feest van! Lieve Eelke Toxopeus, ik had m´n stethoscoop ervoor gegeven,

dat weet je. Gelukkig hoefde dat niet. Dank voor de gezelligheid! Koen Talsma, veel dank voor wat jij

met de ‘777’-database voor elkaar hebt gekregen. Veel geluk!

Mark van Heijl en Rachel Blom, Mrs Guide, wat is het leuk om jullie nog zo regelmatig te zien. Ik

beschouw de klim naar de congreslocatie in Kagoshima nog steeds als mijn grootste prestatie.

Hoewel het wegwerken van 24 dumplings een goede tweede is. Maarten en Vera (om het 020-clubje

compleet te maken), dank voor jullie hulp en samenwerking.

Alle leden van de Slokdarmwerkgroep, hartelijk dank voor jullie hulp. Ik heb in alle uren achter de

computer op vrijdag verschrikkelijk veel geleerd, met name dat oncologie niet in een richtlijn te

vatten is. Katharina Biermann, dankjewel voor het beoordelen, herbeoordelen en nóg eens bekijken

van al die coupes. Beste professor Tilanus, voor sommigen zou het emeritaat wat mij betreft niet

hoeven gelden.

Een speciale vermelding voor Renée Acosta. Lieve Renée, jij bent de beste die een patiënt kan

tegenkomen. Je inzet voor patiënten kent geen grenzen, je durft ruzie voor ze te maken. Heel veel

dank voor al je hulp bij de poliklinische controles van de VIDEO-patiënten, je enthousiasme en alle

gezellige momenten! Je bent een topwijf.

Publiceren doe je niet alleen. Heel veel dank aan alle co-auteurs voor het ontwerpen, includeren,

verzamelen van data, meedenken en meeschrijven.

Zonder het secure werk van alle datamanagers die meewerkten aan de verschillende studies was dit

proefschrift er nooit geweest. Conny Vollebregt spant echter de kroon. Haar werk staat aan de basis

van het grootste deel van de publicaties over slokdarmpathologie uit Rotterdam. Conny, dank voor je

tomeloze inzet. Tonny, Danielle en Aleida, veel dank voor jullie hulp op de poli!

Bewoners van de hunkerbunker!! Bedankt voor het lachen! Als je er zo lang hebt gezeten als ik (vanaf

Tamara Conjo!) zijn er teveel mensen langsgekomen om allemaal persoonlijk te bedanken. Ik beperk

me tot hen die mij letterlijk het meest na waren. Ruggentje, Aaltje Garnaaltje, Marcella Druppel, dank

jullie wel voor de onvergetelijke tijd.

Acknowledgements |

205

Chirurgen en arts-assistenten uit het IJsselland Ziekenhuis, ik had me geen betere plek kunnen

wensen om opgeleid te worden, en dat komt door jullie! Jullie vragen nooit hoe het nou met mijn

promotie gaat. Ook daarvoor veel dank.

Lieve Zarina Lalmahomed, wat hebben we lang naast elkaar gezeten en wat ben ik blij dat je tijdens

mijn verdediging naast mij zal staan. Vanaf het moment dat ik naast je neerplofte achter Martin’s jaren

’70 bureau waren we begonnen! Het nummer van de dag, later uitgebouwd naar thema van de dag,

de satésaus-ontploffing in mijn keuken, de BBQ’s, stafdag-cabarets (inclusief dollemansrit door Delft-

Den Haag-Rijswijk) en als klap op de vuurpijl het chirurgendagen cabaret, we waren er altijd samen

bij. Jouw werklust en doorzettingsvermogen zijn onnavolgbaar. Dankjewel voor de mooie tijd en voor

je vriendschap!

Karel Klop. Goede vrienden zijn goud waard! Samen hebben we onze studententijd nog eens

dunnetjes overgedaan, het was een feest. Als ik alle schitterende avonden zou moeten beschrijven

kwam er geen einde aan dit dankwoord en werd het net zo lang als de bestelling bij de Mac. Gelukkig

gaat het feest door in het IJsselland, ook al mogen we eigenlijk niet meer naast elkaar zitten tijdens de

overdracht. Laten we het erop houden dat ze onze humor gewoon niet zo goed begrijpen. Ik ben blij

dat ik je heb leren kennen en dat jij mijn paranimf wil zijn. Ik hoop dat je tijdens mijn verdediging mijn

huis niet probeert af te branden, dat zou niet zo origineel zijn.

Lieve schoonfamilie, lieve Jeanette, Egge, Irene, Serah, Jasper, Sebastiaan, Joop, Julius, Olivia, Jasmijn

en Rosa. Dank voor de gezelligheid, gastvrijheid, ondersteuning, interesse en warmte. Willen jullie nu

alsjeblieft nooit meer naar mijn onderzoek vragen?

Lieve Monique, Naud, Maxime, Vince, Quint , Isa en Martijn. Familie hoef je niet meteen te zijn, dat kun

je blijkbaar ook worden. Ik zou jullie voor geen goud willen missen!

Lieve Pien, lieve Roos, wij komen uit hetzelfde nest, dat merk je aan alles. Ik had me echt geen betere

zussen kunnen wensen. Dank voor jullie humor en zorgzaamheid. Lieve Bart en Mau, het is ontzettend

fijn als je zussen nieuwe vrienden voor je mee naar huis nemen! Lieve Anna, Gijs en Mies, wat een

verschrikkelijk leuke kinderen zijn jullie! Ik ben er trots op dat ik jullie oom ben.

Lieve mamma, jij weet als geen ander hoe het is om de data voor een grote studie in het gareel te

krijgen. Dank voor je wijze lessen in de GCP en je begrip voor mijn frustraties. Lieve pappa, doctor

zal jij denk ik nooit meer worden, maar ik hoop eigenlijk ooit net zo´n goede dokter als jij te worden.

Lieve mamma, lieve pappa, dank jullie wel voor alles wat jullie mij gegeven hebben, maar bovenal

voor jullie onvoorwaardelijke liefde.

| Acknowledgements

206

Lieve Dee, wat ben ik trots op jou. Jouw gedrevenheid en discipline zijn jaloersmakend. Je zorgt voor

de rustige basis in mijn leven. Wie had kunnen denken dat het liefste meisje ook nog eens de meest

geweldige moeder zou zijn? Samen kunnen we de wereld aan! Ik hou van je.

Lieve, lieve Sara, ik hou zo ongelóóflijk veel van jou. Voor jouw geluk zou ik dit hele proefschrift

inleveren. Ze mogen het houden. Ik kan niet wachten om je in je nieuwe rol als grote zus te zien. Kus!

Curriculum Vitae |

207

CURRICULUM VITAE

Pieter van Hagen werd op 6 juli 1981 geboren te Leiden. Op tweejarige leeftijd verhuisde hij met zijn

ouders naar Rotterdam, waar hij een bijzonder gelukkige jeugd beleefde en in 2000 zijn VWO-diploma

behaalde aan het Rotterdams Montessori Lyceum. In september van dat jaar begon hij aan de studie

geneeskunde aan de Universiteit Utrecht, waar hij in 2007 zijn artsenbul behaalde. In 2008 werkte hij

als ANIOS in het Erasmus Medisch Centrum te Rotterdam (prof. dr. J.N.M. IJzermans), waarna hij in

2009 begon aan een promotietraject onder begeleiding van prof. dr. J.J.B. van Lanschot. In januari

2013 mocht hij beginnen aan de opleiding tot chirurg in de opleidingsregio Rotterdam (dr. B.P.L.

Wijnhoven), waarvan hij het perifere deel van de opleiding mag volgen in het IJsselland Ziekenhuis te

Capelle aan den IJssel (Dr. I. Dawson). Pieter is al jaren gelukkig samen met Deodata. Op 12 maart 2014

werden zij de trotse ouders van Sara.

Esophageal Cancer preoperative management and postoperative outcome

Pieter van Hagen

Esophageal Cancer preoperative managem

ent and postoperative outcome

Pieter van Hagen

Uitnodiging

Voor het bijwonen van de openbare verdediging

van het proefschrift

Esophageal Cancer;preoperative management and

postoperative outcome

door

Pieter van Hagen

Op woensdag 18 mei 2016 om 13.30 uur.

Professor Queridozaal, Onderwijscentrum Erasmus MC.

Wytemaweg 80, Rotterdam

Receptie aansluitend aan de promotie

ParanimfenZarina Lalmahomed

Karel Klop

Pieter van HagenRolderdiephof 13521 DA Utrecht

[email protected]

13631_Hagen_OM.indd 1 15-04-16 16:43

Esophageal Cancer - EUR · 2016. 5. 20. · Chapter 10 Recurrence pattern in patients with a...

Documents

Transcript of Esophageal Cancer - EUR · 2016. 5. 20. · Chapter 10 Recurrence pattern in patients with a...