Timo Soeterik

142 CHAPTER 8 METHODS Study population After receiving institutional review board approval, patients diagnosed with prostate cancer undergoing ePLND between January 2015 and September 2019 at three Dutch teaching hospitals (St. Antonius Hospital Nieuwegein/Utrecht, Hospital Group Twente Almelo/Hengelo, and Canisius Wilhelmina Hospital Nijmegen), were included. Patients underwent ePLND combined with radical prostatectomy or prior to radiation therapy. In general, patients with a risk of LNI >5% (based on DRE T-stage), calculated using the MSKCC web calculator, were considered as candidates for ePLND. 9 However, deviations were allowed at the discretion of the treating urologist. DRE T-stage, mpMRI T-stage, preoperative prostate-specific antigen (PSA), highest International Society of Urological Pathology (ISUP) grade observed on most recent preoperative biopsy, total number of biopsy cores taken on systematic biopsy and the relative number of cores containing prostate cancer on systematic biopsy were collected. Patients were included if they underwent systematic biopsies with or without MRI-guided target biopsy and mpMRI for local staging prior to ePLND. Patients undergoing salvage ePLND or those who received androgen deprivation therapy prior to ePLND were excluded. Covariates and endpoints Prostate-specific antigen, DRE T-stage, mpMRI T-stage, total number and relative number of positive biopsy cores as well as pathological lymph node status were collected during standard clinical practice. Biopsy grading was performed according to the new Gleason Grade group classification. 13 Digital rectal examination was performed during the primary diagnostic evaluation by urologists with >5 years of experience in diagnosing and staging prostate cancer. DRE consisted of systematic palpation of all prostate regions including both lateral sides, the posterior region and the sulcus. DRE was performed in either the dorsal lithotomy or lateral position. Findings were reported according to the clinical classification of the American Joint Committee on Cancer. 14 During the study period, 3-Tesla MRI scanners were used at the three institutions. Radiological reporting was performed by dedicated uro-radiologists. Reporting was done according to the Prostate Imaging – Reporting and Data System (PI-RADS) v2 guidelines. 15 MpMRI T-stages were defined as T1c (non-visible lesion), T2a (unilateral suspicious lesion, involving <50% of the prostatic lobe), T2b (unilateral suspicious lesion, involving >50% of the prostatic lobe) T2c (bilateral suspicious lesion), T3a (definite or high degree of suspicion for extraprostatic extension, T3b (definite or high-degree of suspicion of seminal vesicle invasion) and T4 (invades adjacent structures). The MRI protocols used at the three institutions are presented in the Supplemental Section (Table S1).