Measurement of variation in imaging report findings for 68Ga-PSMA-11 PET/CT in patients with prostate cancer: an international study with participation of multiple centers

ISRCTN ISRCTN13499475
DOI https://doi.org/10.1186/ISRCTN13499475
Secondary identifying numbers 13102016.1.0
Submission date
13/10/2016
Registration date
27/10/2016
Last edited
07/01/2022
Recruitment status
No longer recruiting
Overall study status
Completed
Condition category
Cancer
Prospectively registered
Protocol
Statistical analysis plan
Results
Individual participant data

Plain English summary of protocol

Background and study aims
68Ga-PSMA-11 PET/CT is a new imaging technique developed to use for investigating how advanced a cancer is and where it is located. It can be used, for example, to monitor the response of a cancer to radiotherapy. The overall value of a imaging technique such as this depends on the degree of reader agreement – that is, if a number of people agree on what the image is showing. Knowing how different people will interpret the same image (interobserver variability) and how many people will agree that it shows the same thing (reproducibility) is essential for interpreting the results of studies and for designing new studies. The aim here is to determine interobserver agreement for 68Ga-PSMA-11 PET/CT interpretations for prostate cancer and to compare the findings among highly experienced and less experienced readers.

Who can participate?
Nuclear medicine physicians or radiologists with PET/CT experience.

What does the study involve?
Participants are given 68Ga-PSMA-11 PET/CT images from anonymous patients with prostate cancer and, upon a visual inspection, are asked to report on their findings using a standard template. They are asked to interpret what they see from a number of images of lymph nodes affected by localized prostate cancer and divided into different regions or sections. The participants are asked to judge whether each image they see shows prostate cancer. The results are compared to that of histopathology and follow-up imaging taken for all the patients more than six months before the 68Ga-PSMA-11 PET/CT imaging. Interobserver agreement between participants is assessed.

What are the possible benefits and risks of participating?
Not provided at time of registration

Where is the study run from?
Department of Nuclear Medicine, Ludwig-Maximilians-University Munich (Germany)

When is the study starting and how long is it expected to run for?
October 2016 to March 2017

Who is funding the study?
Department of Nuclear Medicine, Ludwig-Maximilians-University Munich (Germany)

Who is the main contact?
Dr Wolfgang Fendler

Contact information

Dr Wolfgang Fendler
Scientific

Marchioninistraße 15
81377 München
Munich
81377
Germany

Study information

Study designObservational
Primary study designObservational
Secondary study designCross sectional study
Study setting(s)Hospital
Study typeDiagnostic
Scientific title68Ga-PSMA-11 PET/CT interobserver agreement for prostate cancer assessments: an international multicenter prospective study
Study acronymPSMAGREE
Study objectivesThe overall value of an imaging method is associated with the degree of reader agreement. Knowledge of interobserver variability and reproducibility is therefore essential for interpreting study results and design of future trials. The aim of this study is to determine interobserver agreement for interpretations of 68Ga-PSMA-11 PET/CT and to compare findings among readers with low and high levels of experience.
Ethics approval(s)Ethics Committee of the Ludwig-Maximilians-University Munich, 10/06/2016, ref: 594-16UE
Health condition(s) or problem(s) studiedProstate cancer
InterventionAnonymized PET/CT images of 50 patients included in the study (one scan per patient) will be electronically submitted to more than ten nuclear medicine physicians/radiologists from Europe, Asia, Australia or North America. Data include standard DICOM files of CT and attenuation-corrected PET images.

All reader report a visual interpretation of the 68Ga-PSMA-11 PET/CT image datasets using a standard template. Visual interpretation will be performed for locoragional lymph nodes - divided into pelvic regions, extra-pelvic lymph nodes, bone and visceral organs. Each region will be judged positive or negative for prostate cancer involvement.

For binary data, agreement among each observer and the reference reader will be evaluated using Cohen's κ. Overall agreement using pooled observer data will be evaluated using generalized estimation equations. For non-binary data, agreement among all observers was evaluated by intraclass correlation coefficient (ICC) using two-way mixed model for absolute agreement (single measures). Ninety-five percent confidence intervals (CIs) are reported for κ and ICC values. Interpretation of κ and ICC will be based on a classification provided by Landis and Koch: 0.0, poor; 0.0–0.20, slight; 0.21–0.40, fair; 0.41–0.60, moderate; 0.61–0.80, substantial; 0.81–1.00, almost-perfect reproducibility. Discrepancies in quantitative ratings among observers will be expressed as mean difference (Δ) ± standard deviation (SD). Sensitivity, specificity, positive predicive value, and negative predicive value will be calculated for each observer

Histopathology and follow-up (PSA and imaging) obtained for all patients more than six months before the index test serves as reference standard. Lymph nodes regions, organs and bone will be judged positive or negative for prostate cancer involvement based on the results from histopathology and follow-up imaging.
Intervention typeDevice
Pharmaceutical study type(s)
Phase
Drug / device / biological / vaccine name(s)
Primary outcome measureInterreader agreement, i.e. the variance between imaging findings of different readers, calculated by Cohen's κ.
Secondary outcome measures1. Accuracy for each observer compared to the reference standard, determined by sensitivity, specificity, positive predicive value, and negative predictive value
2. Comparison of accuracy between readers of low and high levels of experience, by difference in pooled sensitivity, specificity, positive predictive value, and negative predictive value
Overall study start date13/10/2016
Completion date01/03/2017

Eligibility

Participant type(s)Health professional
Age groupAdult
SexBoth
Target number of participants16
Total final enrolment16
Key inclusion criteriaNuclear medicine physician or radiologist with PET/CT experience.
Key exclusion criteriaPhysicians with prior knowledge of the included patient datasets.
Date of first enrolment01/11/2016
Date of final enrolment13/12/2016

Locations

Countries of recruitment

  • Australia
  • Austria
  • Denmark
  • France
  • Germany
  • Japan
  • United States of America

Study participating centres

Department of Nuclear Medicine, Ludwig-Maximilians-University Munich
Klinik und Poliklinik für Nuklearmedizin
Klinikum der Universität München
Marchioninistraße 15
81377 München
Munich
81377
Germany
University of California Los Angeles
Ahmanson Biological Imaging Clinic
10833 LeConte Ave.
Los Angeles
CA 90095-6948
United States of America
University Hospital Essen
Department of Nuclear Medicine
Essen
-
Germany
Hokkaido University Graduate School of Medicine
Department of Nuclear Medicine
Kita 15, Nishi 7, Kita-ku
Sapporo
060-8638
Japan
Technical University of Munich (TUM)
Department of Nuclear Medicine
Munich
-
Germany
University of Ulm
Department of Nuclear Medicine
Ulm
-
Germany
Bichat University Hospital
Department of Nuclear Medicine
Inserm 698
University Paris 7
46 rue Henri Huchard
Paris
75018
France
University of Rostock
Department of Nuclear Medicine
Rostock
18051
Germany
St Vincent's Public Hospital Sydney
Department of Nuclear Medicine
Sydney
-
Australia
Aalborg University Hospital
Department of Nuclear Medicine
Hobrovej 18-22
Aalborg
9000
Denmark
University of California San Francisco
Department of Nuclear Medicine
05 Parnassus Ave
San Francisco
CA 94143
United States of America
General Hospital of Vienna (AKH)
Department of Nuclear Medicine
Vienna
-
Austria
University Hospital of Würzburg
Department of Nuclear Medicine
Würzburg
-
Germany

Sponsor information

Department of Nuclear Medicine, Ludwig-Maximilians-University Munich
Hospital/treatment centre

Klinik und Poliklinik für Nuklearmedizin
Klinikum der Universität München
Marchioninistraße 15
81377 München
Munich
81377
Germany

ROR logo "ROR" https://ror.org/05591te55

Funders

Funder type

University/education

Department of Nuclear Medicine, Ludwig-Maximilians-University Munich

No information available

Results and Publications

Intention to publish date01/03/2017
Individual participant data (IPD) Intention to shareNo
IPD sharing plan summaryNot expected to be made available
Publication and dissemination planResults (Cohen's κ, intraclass correlation coefficient, sensitivity, specificity, positive predicitive value, negative predictive value, accuracy) will be published within three to six months of completion of the trial.
IPD sharing planThe datasets generated during and/or analysed during the current study are not expected to be made available.

Study outputs

Output type Details Date created Date added Peer reviewed? Patient-facing?
Results article 13/04/2017 07/01/2022 Yes No

Editorial Notes

07/01/2022: Publication reference and total final enrolment added.
01/11/2016: Added list of trial participating centres