Threedimensional Multiparametric Ultrasound for Prostate Cancer Detection

Improving the AI-based Image Analysis Algorithm for 3D Multiparametric Ultrasound for the Detection, Grading and Localization of Prostate Cancer

Rationale: Current imaging techniques for the detection and grading of prostate cancer are imperfect, leading to unnecessary biopsies, suboptimal treatment decisions and missed clinically significant cancers. The hypothesis of this study is that computer assisted analysis of 3D multiparametric ultrasound (mpUS) images can accurately detect, grade and localize prostate cancer. 3D mpUS may then become a more cost-effective and more streamlined imaging strategy than the current standard: mpMRI.

Objective: The primary objective is to collect high-quality 3D mpUS and histology data, to train and improve the classifier algorithm with the goal of achieving an accurate ultrasound imaging tool for the detection of clinically significant prostate cancer.

Secondary objectives are related to the preliminary assessment of the performance of 3D mpUS with computer assisted analysis.

Study design: This is a prospective, multi-center study in men with a suspicion of prostate cancer who are scheduled for prostate biopsies, and men with confirmed prostate cancer who are scheduled to undergo a radical prostatectomy. Prior to prostate biopsies or the radical prostatectomy, 3D mpUS imaging will be performed. The ultrasound images will be analyzed and used for algorithm training using the biopsies and/or prostatecomy specimens as gold standard. Additional research coupes of pathology material (both biopsies and radical prostatectomy specimens) from study subjects will be anonymized and separately analyzed and stored in a central, independent institution. The outcome of the 3D mpUS analysis and the additional pathology evaluation are for research purposes only and will not interfere with standard patient care.

Study population: 1) Male patients of age ≥18 suspected for prostate cancer who are scheduled for systematic and/or targeted biopsy after mpMRI examination.

2) patients of age ≥18 with confirmed prostate cancer who are scheduled for radical prostatectomy.

Main study parameters/endpoints:

• Gleason/Grade group scoring based on histology. Using histology as the reference standard the accuracy of the algorithm will be optimized to be differentiating between benign tissue and various grades of malignancy.
• Localization and size of lesions at full-gland histology in the subset of patients undergoing radical prostatectomy. Correlation in tumour size and location will be optimized between 3D mpUS findings and histology of the full gland.

For the secondary objective, preliminary assessment of the performance of 3D mpUS, the following endpoints are evaluated

• Among all clinically significant detected cancers confirmed by histology, the proportion of these cancers that would have been detected by 3D mpUS will be calculated. The number of false positive findings by 3D mpUS both as an absolute count and expressed as a mean rate per patient.
• The concordance in the detection and grading of abnormalities between mpMRI and 3D mpUS by examining the frequency and type of disagreements and calculating the kappa statistic.

Inclusion criteria

In order to be eligible to participate in this study, a subject must meet all of the following criteria:

• Men ≥18 years with a clinical suspicion of prostate cancer or confirmed prostate cancer.
• Scheduled for either systematic and/or targeted biopsy after mpMRI examination or radical prostatectomy
• Signed informed consent

Exclusion criteria

A potential subject who meets any of the following criteria will be excluded from participation in this study:

• No mpMRI performed prior to prostate biopsy or radical prostatectomy

• A history of chemotherapy for PCa or currently being treated with chemotherapy for PCa.

• A patient history that includes any of the following prostate related interventions:

  • Brachytherapy or external radiotherapy for PCa;
  • Focal therapy for prostate cancer;
  • Prostate biopsy within the last 30 days.

• Hormonal therapy for prostate cancer within the last six months

• A patient history with a cardiac right to left shunt.

• Current treatment with dobutamine

• Known severe pulmonary hypertension (pulmonary artery pressure >90 mmHg), uncontrolled systemic hypertension or respiratory distress syndrome

• Incapable of understanding the language in which the patient information is given.