COMBINED PSMA-PET/CT AND MRI STAGING IN INTERMEDIATE AND HIGH-RISK PATIENTS PROSTATA-CANCER (COMBINE-P)

Overview

This retrospective, multicentre comparative analysis aims to evaluate a new staging method for i) improved detection of intraprostatic index lesions, ii) local T-staging and iii) lymph node (LN) staging in men with clinically significant prostate cancer (csPCa) at intermediate/high risk by combining prostate-specific membrane antigen (PSMA) positron emission tomography (PET) imaging using different tracers ((18)F) DCFPyL, Gallium-68, Fluor-18) and multiparametric magnetic resonance imaging (MRI) in patients with prostate cancer (PCa) who subsequently underwent radical prostatectomy (RP). Another secondary endpoint will be the additional value of PSMA-PET/CT in men with unremarkable MRI. Men at intermediate risk (PSA \> 10 ng/ml to 20 ng/ml or Gleason score 7 or cT category 2b) or high risk (PSA \> 20 ng/ml or Gleason score ≥ 8 or cT category 2c) who underwent PSMA-PET/CT and mpMRI followed by RP will be analysed in three different subgroups corresponding to the modalities i) PSMA-PET/CT with 18-F-DCFPyL (subgroup/arm A), ii) Gallium-68 PSMA-PET/CT (subgroup/arm B) and Fluorine-18 PSMA-PET/CT (subgroup/arm C). The validation of the accuracy of the detection of intraprostatic index lesions, local and lymph node staging by MRI and PSMA-PET-CT with different tracers is carried out using the histological radical prostatectomy specimens. In addition, the prediction of the International Society of Urolgenital Pathology (ISUP) graduation group (GG) within intraprostatic index lesions will be determined using the SUV (standardised uptake value) in PSMA-PET-CT and using ADC values (Apparent Diffusion Coefficient of the diffusion-weighted MRI sequence) in MRI (7,8). The ability of PSMA-PET-CT to predict extraprostatic, i.e. capsule-transcending, tumour growth is also analysed in comparison with MRI. In addition, the correlation of tumour localisation (right vs. left) in relation to positive lymph nodes (right vs. left) is analysed. Finally, the added value of PSMAPET-CT in the case of negative, unsuspicious MRI is determined. Overall, our analysis aims to improve patient care by analysing the potential of non-invasive "digital biopsy" in terms of lesion detection and prediction of the histological grading group. In addition, a proof-of-concept for personalised lymph node dissection based on prediction of lymph node metastasis and patient-tailored nerve sparing with accurate prediction of extracapsular extension will be tested based on combined preoperative PSMA-PET and MRI imaging. The results of these two analyses will have a direct impact on clinical practice and the further use of highly specialised imaging. In addition, this multi-centre data analysis will provide the European Prostate Cancer Center of Excellence (EPCCE) group with a proof-of-concept for future projects.

Study protocol - COMBINE-P 1. Please enter a meaningful study title that describes the project at hand COMBINED PSMA-PET/CT AND MRI STAGING IN INTERMEDIATE AND HIGH-RISK PATIENTS PROSTATA-CANCER (COMBINE-P) - A multicentre retrospective analysis in the European Prostate Cancer Centres of Excellence for Prostate Cancer (EPCCE) 2. Name and title of the study coordinators, degree/ profession, institute/ clinic Prof. Dr Jan Philipp Radtke, Deputy Director and Senior Consultant, Department of Urology, Düsseldorf University Hospital Prof. Dr Peter Albers, Clinic Director, Clinic for Urology, Düsseldorf University Hospital Dr Isabelle Busshoff, Assistant Physician, Department of Urology, Düsseldorf University Hospital 2.1 Study Site Coordinators, degree/ profession, institute/ clinic 2.1.1 Germany - University Hospital Munich - Ludwigs-Maximilian-University Munich Prof. Dr Christian Stief, Full Professor of Urology and Chairman, Department of Urology Dr Thilo Westhofen, Consultant, Department of Urology 2.1.2 Germany - University Hospital Tübingen Prof. Dr Arnulf Stenzl, Professor of Urology and Chairman, Department of Urology Prof. Dr Steffen Rausch, Professor of Urology, Department of Urology, University Hospital Tübingen 2.1.3 Germany - University Hospital Bochum Prof. Dr Joachim Nodus, Full Professor of Urology, Marien Hospital Herne University of Ruhr-University Bochum Prof. Dr Florian Roghmann, Professor of Urology, Marien Hospital Herne University of Ruhr-University Bochum 2.1.4 Germany - University Hospital Düsseldorf Prof. Dr Lars Schimmöller, Institute of Diagnostic and Interventional Radiology Prof. Dr Frederik L. Giesel, Department of Nuclear Medicine 2.1.5 United Kingdom - Christie Clinic Foundation Trust Manchaster Prof. Dr Vijay Sangar, Full Professor of Urology, Department of Urology 2.1.6 Belgium - University Hospital Leuven Prof. Dr Steven Joniau, Full Professor of Urology, Department of Urology Prof. Dr Karolien Goffin, Full Professor of Nuclear Medicine , Department of Nuclear Medicine 2.1.7 France - Hospital Civils de Lyon Prof. Dr Alain Ruffion, Full Professor of Urology, Department of Urology 2.1.8 Switzerland - University Hospital Bern Prof. Dr George Thalmann, Professor of Urology, Department of Urology 2.1.9 Italy - Hospital IRCCS San Raffaele Milan Prof. Dr Francesco Montorsi, Full Professor of Urology, Department of Urology Prof. Dr Alberto Briganti, Professor of Urology, Department of Urology Dr Armando Stabile, Consultant, Department of Urology 2.1.10 Austria - University Hospital Vienna Prof. Dr Shahrokh Shariat, Full Professor of Urology, Department of Urology Dr Pawel Rajwa, Consultant, Department of Urology 2.1.11 Sweden - Lund University, Skane University Hospital Prof. Dr Anders Bjartell, Professor of Urology, Department of Urology 3\. a. Is this an initial application or is there already an ethics vote from another EC? First vote 4\. Please provide a brief summary of your study. Please explain the aim of the study and what new findings you are expecting This retrospective, multicentre comparative analysis aims to evaluate a new staging method for i) improved detection of intraprostatic index lesions, ii) local T-staging and iii) lymph node (LN) staging in men with clinically significant prostate cancer (csPCa) at intermediate/high risk by combining prostate-specific membrane antigen (PSMA) positron emission tomography (PET) imaging using different tracers ((18)F) DCFPyL, Gallium-68, Fluor-18) and multiparametric magnetic resonance imaging (MRI) in patients with prostate cancer (PCa) who subsequently underwent radical prostatectomy (RP). Another secondary endpoint will be the additional value of PSMA-PET/CT in men with unremarkable MRI. Men at intermediate risk (PSA \> 10 ng/ml to 20 ng/ml or Gleason score 7 or cT category 2b) or high risk (PSA \> 20 ng/ml or Gleason score ≥ 8 or cT category 2c) who underwent PSMA-PET/CT and mpMRI followed by RP will be analysed in three different subgroups corresponding to the modalities i) PSMA-PET/CT with 18-F-DCFPyL (subgroup/arm A), ii) Gallium-68 PSMA-PET/CT (subgroup/arm B) and Fluorine-18 PSMA-PET/CT (subgroup/arm C) (1-6). The validation of the accuracy of the detection of intraprostatic index lesions, local and lymph node staging by MRI and PSMA-PET-CT with different tracers is carried out using the histological radical prostatectomy specimens. In addition, the prediction of the International Society of Urolgenital Pathology (ISUP) graduation group (GG) within intraprostatic index lesions will be determined using the SUV (standardised uptake value) in PSMA-PET-CT and using ADC values (Apparent Diffusion Coefficient of the diffusion-weighted MRI sequence) in MRI (7,8). The ability of PSMA-PET-CT to predict extraprostatic, i.e. capsule-transcending, tumour growth is also analysed in comparison with MRI. In addition, the correlation of tumour localisation (right vs. left) in relation to positive lymph nodes (right vs. left) is analysed. Finally, the added value of PSMAPET- CT in the case of negative, unsuspicious MRI is determined. Overall, our analysis aims to improve patient care by analysing the potential of non-invasive "digital biopsy" in terms of lesion detection and prediction of the histological grading group. In addition, a proof-of-concept for personalised lymph node dissection based on prediction of lymph node metastasis and patient-tailored nerve sparing with accurate prediction of extracapsular extension will be tested based on combined preoperative PSMA-PET and MRI imaging. The results of these two analyses will have a direct impact on clinical practice and the further use of highly specialised imaging. In addition, this multi-centre data analysis will provide the European Prostate Cancer Center of Excellence (EPCCE) group with a proof-of-concept for future projects. 5\. Please indicate which of the provisions and principles in the applicable version are relevant to the project and will therefore be taken into account and observed during implementation. a. Professional code of conduct for doctors b. Declaration of Helsinki c. GDPR d. Federal Data Protection Act e. NRW State Data Protection Act f. Health Data Protection Act NRW 6\. Please state the exact period to which the planned data collection relates (e.g. "01/2005 - 12/2014" or "01/2007 to the date of application"). Multicentre, retrospective data analysis. At the University Hospital Düsseldorf of the Heinrich-Heine University Düsseldorf, the retrospective data analysis will cover the period 04/2021-04/2023. At the other centres, the retrospective data analysis will cover the period 01/2016-04/2023. 7\. Case number planning Please enter the planned number of cases (e.g. n = x patients). Multicentre n = 600 patients. At the University Hospital Düsseldorf approx. n=60 patients. 8\. Informed consent 1. Was written consent obtained from the patient? No 2. If "No" In addition to the time and cost involved in following up the treated patients to obtain their "informed consent", the probability of non-consent can be considered lowoverall. The retrospective analysis of the data as part of the research project is based exclusively on data acquired in the course of medically necessary treatment at the Department of Urology. A retrospective collection of additional data is not planned. The data used is extracted from archived data of the Clinic for Urology by employees of the Clinic for Urology who are bound to confidentiality, with immediate anonymisation, so that it is not possible to trace the data back to the individual patient. The use of the data has no further effect on the treatment of patients. The evaluation does not result in any health or ongoing care risk for the patient. The analysis does not result in any relevant therapy outcomes for the patient. The patients analysed in the retrospective analysis have already consented to the use of their collected data for scientific purposes as part of the MRProRoutine study (study number: 5910R; study registration ID: 201/0341/1). c. Are there any indications that patients (including individual patients) have objected to their data being used for research purposes? There is no evidence for this exception. 9\. Data management 1. What is the source of the data? Data is collected from archived patient files of patients from the Department of Urology at the University Hospital Düsseldorf. As part of the multicentre, retrospective data analysis, the anonymised data of the HHU patients are compared with those of the Inselspital Bern, the University Hospital Vienna, the Hospital IRCCS San Raffaele Milan, Anser/ERASMUS University Hospital Rotterdam, the Hospital Civils Lyon, University Hospital Leuven/Belgium, University Hospital Lund/Sweden, University Hospital Tübingen, University Hospital Ludwigs- Maximilian University Munich, from the Medico patient information system are stored in a database. In addition, data from the PACS image processing and information systems of the University Hospital Düsseldorf, the Inselspital Bern, the University Hospital Vienna, the Hospital IRCCS San Raffaele Milan, Anser/ERASMUS University Hospital Rotterdam, the Hospital Civils Lyon, the University Hospital Leuven, the University Hospital Lund/Sweden, the University Hospital Tübingen, the Hospital Herne of the University Hospital of the Ludwig Maximilian University Munich are stored retrospectively and anonymised. 2. For data from patient files: Were the patients treated in your clinic? Yes. 3. How is the data collected? Anonymised at source. 4. How is the data coded? By means of an ascending series of numbers. 5. Who collects the data? Dr Isabelle Bußhoff, Assistant Physician (Clinic for Urology, UKD) Prof Dr Jan Philipp Radtke (Clinic for Urology, UKD) Prof Roghmann Florian (Department of Urology, Marien Hospital Herne) Prof Dr Georg Thalmann (Department of Urology, Inselspital Bern) Prof Dr Shahrokh Shariat (Department of Urology, University Hospital Vienna) Prof Dr Francesco Montorsi (Department of Urology, Hospital IRCCS San Raggaele Milan) Prof Florian Roghmann (Clinic for Urology, Marien Hospital Herne) Prof Dr Chris Bangma (Anser/ERASMUS University Medical Center Rotterdam) Prof Dr Alain Ruffion (Clinic for Urology, Hospital Civils Lyon) Prof Dr Steven Joniau (Department of Urology, University Hospital Leuven) Prof Dr Karolien Goffien (Department of Nuclear Medicine, University Hospital Leuven) Prof Dr Christian Stief (Department of Urology, University Hospital of the Ludwig Maximilian University of Munich) Prof Dr Steffen Rausch (Department of Urology, University Hospital Tübingen) Prof Andrers Bjartell (Department of Urology, Lund University Hospital) 6. Who anonymises the data? Dr Isabelle Bußhoff, assistant doctor (Clinic for Urology) Prof Dr Georg Thalmann (Department of Urology, Inselspital Bern) Prof Dr Shahrokh Shariat (Department of Urology, University Hospital Vienna) Prof Roghmann Florian (Department of Urology, Marien Hospital Herne) Prof Dr Francesco Montorsi (Department of Urology, Hospital IRCCS San Raggaele Milan) Prof Dr Chris Bangma (Anser/ERASMUS University Medical Center Rotterdam) Prof Dr Alain Ruffion (Clinic for Urology, Hospital Civils Lyon) Prof Dr Steven Joniau (Department of Urology, University Hospital Leuven) Prof Dr Karolien Goffien (Department of Nuclear Medicine, University Hospital Leuven) Prof Dr Christian Stief (Department of Urology, University Hospital Ludwigs- Maximilian University Munich) Prof Dr Steffen Rausch (Department of Urology, University Hospital Tübingen) Prof Andrers Bjartell (Department of Urology, Lund University Hospital) 7. Who analyses the data? Dr Isabelle, Bußhoff, Assistant Physician (Clinic for Urology) Prof Dr Jan Philipp Radtke Prof Dr Frederik L. Giesel Prof Dr Lars Schimmöller 8. Who has access to the assignment key? There is no allocation key as the data is anonymised 9. Where is the allocation key kept? There is no allocation key as the data is anonymised 10. How is the assignment key saved? There is no allocation key as the data is anonymised 11. Where is the research data stored? Please note that the allocation key must not be kept/saved in the same folder/directory as the pseudonymised research data. The data is stored in a password-protected Excel database on the server system of the University Hospital Düsseldorf. The data from all centres is stored in the PIONEER data platform via the CASTOR Programme of the European Association of Urology (EAU) (https://prostatepioneer. eu/big-data-platform/pioneer-data-processing/). In the meantime, the applicants have successfully acquired the data storage in PIONEER and the funding for this from the EAU and the EAU Research Foundation. The data exchange models and data encryption of the PIONEER platform are described below: PIONEER works with two data access models - a centralised and a federated model. In the centralised data exchange model, a copy of the e-identified data is transferred to PIONEER, converted and stored in a central data warehouse for research purposes. In the federated model, data owners standardise their own data sets and set up analysis tools within their own data environment and make them available on request. The centralised model will incorporate data by converting population-based registries and epidemiological research data into a common OMOP (Observational Medical Outcomes Partnership) data model to enable the systematic analysis of different observational databases. Finally, the PIONEER Big Data Platform will not only provide access to data but also analytical tools (ATLAS, R) in a single innovative data platform that utilises two existing data platforms, tranSMART and OHDSI, developed in previous IMI projects. Within PIONEER, the data will not be personally identifiable, i.e. the data has been deidentified to ensure sufficient anonymity so that the person cannot be identified. As a result of this anonymisation process, the data within PIONEER's Big Data platform is not classified as personal data and as such the use of the data complies with all applicable data protection laws at EU level and does not fall within the scope of the General Data Protection Regulation (GDPR) without compromising the clinical relevance of the data. PIONEER achieves this by using two database models: A federated and a centralised database model. In the federated database model, the data does not leave its original location but is queried remotely, with PIONEER bringing the analysis to the data. This essentially means that the data remains anonymous to the researchers accessing the federated database. Data from a variety of sources are effectively 'linked' temporarily to answer specific queries. Individual data providers retain ownership and control of their data and only allow queries in response to authorised requests. This model has been used successfully by other IMI projects such as EMIF-AD. In the centralised database model, data is physically moved from the data provider to the centralised PIONEER server. To achieve this, PIONEER uses two anonymisation methods. The first is the complete removal of all direct identifiers (such as patient, name, number, photos or other images that could allow identification). The second is "generalisation", which addresses quasi-identifiers. "Generalisation" replaces the values of a particular attribute with less specific values or dilutes the attributes of affected individuals by modifying the respective scales or magnitudes. If the value is a categorical value, it can be changed to another categorical value that denotes a broader concept of the original categorical value. If the value is numeric, it can be changed to a range of values. For example, the granularity of individual birth rates can be reduced by generalising them to a date range or grouping them by month or year. Other numeric attributes (e.g. age, salary, weight, height or the dose of a drug) can be generalised by banding. By using the federated database model and applying the de facto anonymisation process to data entered into the centralised database, PIONEER does not process any personal data and therefore complies with data protection law and does not require any special authorisations, consents or approvals in relation to the project (https://prostate-pioneer.eu/big-data-platform/pioneer-data-processing/). 10\. Use of image and sound material a. Will image or sound material be analysed retrospectively as part of this study? Yes b. If "Yes": How is anonymisation carried out here? The data required for the study from the image material used is extracted directly from the image material from the PACS in anonymised form. The image material is therefore limited to the MRI and PET-CT images. Anonymisation is performed automatically within the respective PACS software used at the centre. 11\. Utilisation of genetic information a. Does the study use genetic information that allows the identification of a person? No b. If "Yes": Have the subjects/patients consented to the use of their genetic data for research purposes? Not applicable. 12\. Fees a. How is the study financed? The fees are financed by the EAU Research Foundation: EAU Research Foundation, Mr E.N. van Kleffensstraat 5, 6842 CV Arnhem, The Netherlands. Data storage in PIONEER is funded by the company Exini: EXINI Diagnostics AB, Scheelevägen 27, 223 70 Lund, Sweden. 13\. Other information on the study No further details