Marieke van Son

46 CHAPTER 3 indicate the clinical target volume (CTV), allowing coverage of microscopic spread(9). Using a treatment planning system (Oncentra Prostate; Elekta Nucletron,Veenendaal, the Netherlands), a pre-treatment plan was constructed with the following goals and constraints: 1. CTV D95%≥19Gy, or minimal CTV D90%≥17Gy, 2. Bladder and rectum D1cc<12Gy and 3. urethra D10%<21Gy(10). Intra-operatively, live transrectal ultrasound (TRUS)-images were rigidly fused with the pre-treatment MR-delineations. MR-com- patible self-anchoring catheters were inserted in and around the CTV transperineally. After insertion, an intra-operative MRI was performed for catheter reconstruction and adaptation of delineations to account for anatomical changes, after which the treat- ment plan was updated. A final MRI was performed for position verification, after which irradiation was given. Outcome assessment The Common Terminology Criteria for Adverse Events (CTCAE) version 4.0 was used for GU, GI and ED toxicity assessment. QoL was measured using the RAND-36, EO- RTC-QLQ-C30 and PR-25 questionnaires(11-14). In addition, International Prostate Symptoms Score (IPSS) and the International Index of Erectile Function (IIEF) were assessed. Serial PSA measurements were performed at 1, 3, 6, 9 and 12 months, sub- sequently every 6 months until 24 months and yearly afterwards until 10 years. The nadir+2 ng/ml (Phoenix) definition was used to assess biochemical failure (BF), which was an indication for a PET/CT-scan. If localised intraprostatic recurrence was found without metastatic disease, an mp-MRI was performed. Tumour control and toxicity risk were carefully weighed to decide whether focal or whole-gland salvage treatment should be performed. If focal salvage treatment was contemplated, MRI-guided biopsy confirmation was performed. Statistical analyses Continuous variables are described as medians with interquartile ranges (IQR) and categorical variables as absolute numbers with percentages. Differences in continuous variables were tested with the Wilcoxon signed rank test. To correct for multiple testing, statistical significance was pragmatically set at p<0.001. Median QoL-score differences of ≥10 points were deemed clinically relevant, apart from statistical considerations(14). BDFS, metastases-free survival (MFS) and overall survival (OS) were assessed using the Kaplan-Meier method. An explorative univariable Cox-regression analysis was performed for the first BF event including the variables age (before treatment), Ameri- can Joint Committee on Cancer (AJCC) grade group, T-stage, PSA, pre-treatment PSA doubling time (PSADT; as calculated with Memorial Sloan Kettering Cancer Center online tool), AJCC prognostic stage group and PSA nadir post-treatment. IBM SPSS v23.0 was used for descriptives and R v3.5.1 for graphs and survival analyses (https:// www.R-project.org/ , ‘survminer’, ‘rms’, ‘ggplot2’ packages).