EARL FDG-PET/CT Accreditation Programme: Feasibility, Overview and Results of First 45 55 Successfully

Accredited Sites

_

Overview of presentation

•

Why

do you

need

accreditation?

‐

background on

SUV uncertainties

•

What

has been realized

so

far

by

EARL?

•

How

to get

accreditation?

•

Results

for

the first

55 sites

Standardised Uptake Value

][/][]/[

kgweightMBqDosemlkBqcSUV t

TBW

SUV ‘is’

activity concentration ratio

Weight is sometimes replaced by BSA, LBM, BMI…

Use of SUV in response assessment  studies 

Absolute SUV:-Patient

eligibility

-Patient

stratification-Lesion

selection

(PERCIST)

-Residual

SUV

Relative

of % SUV changes-% change

of the same

lesions

(EORTC)

-% change

of the (5) hottest lesions

per scan + ΔSUV=0.9 (PERCIST)

For all applications

absolute SUV and SUV changes

are used

PET imaging / SUV uncertaintiesTechnical factors–

Relative calibration between PET scanner and dose calibrator (10%)–

Residual activity in syringe (5%)–

Incorrect synchronisation

of clocks (10%)–

Injection vs

calibration time (10%)–

Quality of administration (50%)

Physics related factors–

Scan acquisition parameters (15%)–

Image reconstruction parameters (30%)–

Use of contrast agents (15%)–

ROI (50%)

Biological factors–

Uptake period (15%)–

Patient motion and breathing (30%)–

Blood glucose levels (15%)

R. Boellaard

2009, J Nucl

Med Supplement Issue 50: 11S

Impact of blood

glucose level

Glu

200 mg% Glu

79 mg%Karoline Spaepen-Sigrid Stroobants

Department of Nuclear MedicineUniversity Hospital Gasthuisberg

Leuven, Belgium

Factors affecting SUV biological factors –

uptake period

Lowe VJ et al. Optimum scanning protocol?for

FDG-PET evaluation of pulmonary malignancy. J Nucl

Med. 1995, image taken from Shankar et al. JNM 2006

Outcome of quantitative FDG-PET studies using standardised

uptake values depend on many biological and technical (physics) factors

*Example of one of the many small

factors: effects of different number of OSEM iterations, as seen in the Netherlands, on SUV

SUVmax

= 4.0

5.9

6.4

8.6SUV 50%= 3.0

4.1

4.6

5.9

Types of standards

/ recommendations

•

Minimal performance standards:–

“Focus”

on

accuracy

–

Lower

threshold

•

Harmonizing

performance standards–

“Focus”

on

reducing

inter‐institute, ‐scanner, ‐

patient

variability

–’precision’–

Lower

and

upper limits

Scanner validation

program

Need

for

harmonisation

of resolution-dependent quantitation, rather

than

minimal performance standards0 1 2 3

Hospital

4.00000

8.00000

12.00000

16.00000

SUV

51-6

1 m

in n

orm

alis

ed to

wei

ght s

can

1

0

2

4

6

8

10

12

14

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Site

Ave

rage

SU

V

FDG PET standardisation

0

2

4

6

8

10

12

14

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Site

Ave

rage

SU

V

W/o standardisation

and

with standardisation

(today)

Primary

aim

of EARL harmonising

QC

program: Reducing

inter

and intra institute

variability

in SUV results

0

2

4

6

8

10

12

14

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Site

Ave

rage

SU

V

FDG PET standardisationWith standardisation

today and……………………….tomorrow

0

2

4

6

8

10

12

14

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Site

Ave

rage

SU

VPrimary

aim

of EARL QC program (NOW): Reducing

inter

and intra institute

variability

in SUV results

Secondary

aim

(NEXT)Enhance

a common

harmonized

standard

Overview of presentation

•

Why

do you

need

accreditation?

‐

background on

SUV uncertainties

•

What

has been realized

so

far

by

EARL?

•

How

to get

accreditation?

•

Results

for

the first

55 sites

FDG PET and PET/CT: EANM Procedure  Guidelines for Tumour

PET Imaging: 

version 1.0

Eur.J.Nucl.Med.Mol.Imag. 2010

The EANM guideline for FDG PET and PET/CT 

provides recommendations for: 

•

Minimising

physiological or biological effects

by patient preparation guidelines

•

Procedures to ensure accurate FDG administration 

•

Matching of PET study statistics (‘image quality’)

by prescribing FDG dosage as 

function of patient weight, type of scanner, acquisition mode and scan duration

•

Matching of image resolution

by specifying image reconstruction settings and 

providing activity concentration recovery coefficients specifications (QC 

experiment)

•

Standardisation

of data analysis

by prescribing region of interest strategies and 

SUV measures

•

Multi‐center QC/QA

procedures for PET and PET/CT scanners

Multi‐center QC and calibration

•

Daily QC conform standard procedure of system / 

manufacturer

•

Calibration QC using (cylindrical) phantom (15‐30cm 

diameter)

• “Adjusted”

NEMA NU 2‐2001 Image Quality 

procedure/measurement to measure recovery coefficients 

as function of sphere size (= ‘effective image resolution’)

•

CT‐QC cf

recommendations of ESR/national law

•

Misc. QC (e.g. for scales, alignment etc)

Based on the QC experiments as described in the EANM guideline 

published in EJNMMI 2010

Manuals, SOPs, online questionnaire completed in August 2010

Training of EARL coordinator (S. Ettinger) September 2010,phyiscs

and scientific

support provided

by

T.Sera

& R.Boellaard

Pilot program (in collaboration with EORTC, trail EORTC 22071‐

24071) began in October 2010

European accreditation program EARL, EANM, EORTC

Accreditation: current

status..

•

Web portal for

uploading

QC data established

•

SW tools for

semi‐automated

analysis

of results

•

Semi‐automated

storage

of results

in dbase

•

First pilot

(n=12) succesfully

completed

in March

2011

•

At present more than

50 sites accredited

•

Accredited

sites will

be

listed

on

EARL site

•

Signet:

•

Program formally

endorsed

by

EORTC

Overview of presentation

•

Why

do you

need

accreditation?

‐

background on

SUV uncertainties

•

What

has been realized

so

far

by

EARL?

•

How

to get

accreditation?

•

Results

for

the first

55 sites

Accreditation:  website & submission

portal

How to get accredited (1) ? 

• Express interest through

EARL website

• Sign

written

agreement

to comply

with

guideline

(GL)

• Fill

out online questionnaireinformation

on

PET/CT systems, DC, wellcounters, SW etc

• Perform

QC experiments

as described

in GL• Manual• SOP for

calibration

QC

• SOP for

image quality/SUV recovery

QC

How to get accredited (3) ? 

• Calibration

QC specification:•

maximum allowable

calibration

deviation

= + or

–

10%

(global)

• SUV recovery

specifications:• for

SUVmax

(focus –as SUVmax

is used

clinically!)

• for

SUVmean

0.0

5000.0

10000.0

15000.0

20000.0

[BQ

ML]

VOI A50%, new limits

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.1 1 10 100

Sphere volume (mL)

SUV

reco

very

coe

ffici

ent

VOI A50%, new limits

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.1 1 10 100

Sphere volume (mL)

SUV

reco

very

coe

ffici

ent

Standardised

software tools for analysis and interpretation of 

QC experiments were developed:

European accreditation program EARL, EANM, EORTC

Calibration QC:-

Automatic VOI placement- Verification of calibration- Verification of inter-&intra-plane uniformity

IQ QC- Recovery coefficients (volume & act.conc.)- Cold spot recovery using central insert (scatter)- Verification of calibration using back ground VOI

Multicenter

FDG PET/CT quantification

0 1 2 3

Hospital

4.00000

8.00000

12.00000

16.00000

SUV

51-6

1 m

in n

orm

alis

ed to

wei

ght s

can

1

Need

for

harmonization

of resolution-dependent quantitation, rather

than

minimal performance standards

Overview of presentation

•

Why

do you

need

accreditation?

‐

background on

SUV uncertainties

•

What

has been realized

so

far

by

EARL?

•

How

to get

accreditation?

•

Results

for

the first

55 sites

Multi‐center

harmonization

of quantification

• Occasionally

it

takes

1-4 iterations

before

IQ-RC is within

specs•

RC harmonizing

specs

‘centered’

around

RC=1 for

SUVmax

for

spheres

larger

than

2.5 mL

0.0

5000.0

10000.0

15000.0

20000.0

[BQ

ML]

Results pooled over all experiments

0

0.25

0.5

0.75

1

1.25

1.5

1.75

2

0.1 1 10 100

Sphere volume (mL)

Rec

over

y co

effic

ient

GE=blue, Philips=red, Siemens=green

0

0.25

0.5

0.75

1

1.25

1.5

1.75

2

0.1 1 10 100Sphere volume (mL)

SUV

RC

Some

‘typical’

image artefacts

Edge

or

ring (Gibbs) artefacts:

•

Nearly

always

seen

with

PSF based

reconstructions

•

Frequently

on

specific

TF systems

(>50% of cases)

Some

‘typical’

image artefacts

75 kg patient, 120 MBq, 3 min/bed(BG = ~1kBq/cc)

0.25

0.5

0.75

1

1.25

1.5

0.1 1 10 100

Sphere volume (mL)

Rec

over

y co

effic

ient

VOI & EANM VOI & PSF+TOFMAX & EANM MAX & PSF+TOF

Ring artefacts

give

upward

bias in SUVmaxR. Boellaard, JNM 2011 +Many

pres @eanm2012

Use

of PSF-recon

requires

change

of data analysis

(i.e. SUVmean)-

VOI based-

No std

VOI method

yet

(CT, PET)

- SUVmax

is usually

used

clinically

Use

of SUVmax

–

EANM -STD

Why

do we use

SUVmax

SUVmax

suffers fromupward

bias due

to noise

(Boellaard

et al, JNM 1996, 2011, Lodge

et al, JNM 201poor

reproducibility

and accuracy

for

PSF (HD) reconstructions

(Boellaard

et al., JNM 2011, Tong, IEEE TNS 2011)

Despite

these limitions:• May represent

metabolically

most active

part of tumor

• VOIs

are not

standardized• CT and PET based

manual

segmentation

suffer

from

observer

variability• PET based

automated

delineation

methods:• variability

of methods• variability

in implementation

of same

method•

performance depend

strongly

on

underlying

image characteristics

(Cheebsumon

et al. JNM2011, EJNMMI 2011)• cannot

deal well

with

tracer

uptake

heterogeneity•Therefore, need

to optimize

image quality

for

use

of SUVmax

RC specifications for SUVmax

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0.1 1 10 100

Sphere volume (mL)

SUV

RC

Multi‐center

harmonization

of quantification

0.0

5000.0

10000.0

15000.0

20000.0

[BQ

ML]

-15

-10

-5

0

5

10

15

0 1 2 3 4

1=GE, 2=Philips, 3=Siemens

% c

alib

ratio

n er

ror i

n N

EMA

BG

Comparable

calibration

accuracy

and SUV recovery

among

sites and vendors is feasible

(n=55)

Multi‐center

harmonization

of quantification

Scanner performance harmonization

is feasible

on

a large

scale, but

long term sustainability

requires

support and service from

vendors –goal of SNM-CTN & EANM/EARL

VENDORS could:- Include

EARL accreditation

in acceptance

testing

OR

-

Setup

EARL compliant

protocol during

application

training- HD-TOF reconstruction

for

visual

inspection

+ 2nd online EARL recon- HD-TOF reconstruction

for

visual

inspection

+ EARL compliant

postprocessing- either

of these 2 are saved

as EARL STD protocol

Multi‐center

harmonization

of quantification

Main

principles

of EANM GL and EARL accreditation• Standardisation

of PET examinination

–procedure• Quantification

is combination

of:• image resolution• image noise• data analysis

methods

(SUVmax

de facto

the standard

in practice)

• EARL QC’s

s based

on

exploration

to find

highest

common

denominator

in • performance of scanner calibration• SUV-RCs

–

SUVmax

!!

•

Scanner performance harmonization

is feasible

on

a large

scale, but

long term sustainability

requires

support and service from

vendors –goal of SNM-CTN & EANM/EARL

Why do we need a guideline for  quantitative FDG PET ? 

0 1 2 3

Hospital

4.00000

8.00000

12.00000

16.00000

SUV

51-6

1 m

in n

orm

alis

ed to

wei

ght s

can

1

Recent (2009) observation on site differences in SUV-Site 1 & 2 closely followed NL standardised

protocol-Site 3 did not –

almost a factor of 2 lower SUV on average

FDG PET and PET/CT: EANM Procedure Guidelines for Tumour

PET Imaging: version 1.0

Ronald Boellaard, Mike O’Doherty, Wolfgang A. Weber, Felix M. Mottaghy, Markus N. Lonsdale, Sigrid G. Stroobants, Wim

J.G. Oyen, Joerg

Kotzerke, Otto S. Hoekstra, Jan Pruim, Paul K. Marsden, Klaus Tatsch, Corneline

J. Hoekstra, Eric.P. Visser, Bertjan

Arends, Fred J. Verzijlbergen, Josee

M. Zijlstra, Emile FI Comans, Adriaan

A. Lammertsma, Anne M. Paans, Antoon

T. Willemsen, Thomas Beyer, Andreas Bockisch, Cornelia Schaefer-Prokop, Dominique Delbeke, Richard P. Baum, Arturo Chiti, Bernd J. Krause.

Eur.J.Nucl.Med.Mol.Imag. 2010

Thank

you

for

your

attention

!