Timo Soeterik

32 CHAPTER 2 tumours (T1a/b), patients with treatment strategies other than AS, and those initially treated at a hospital outside the Santeon group. Active surveillance follow-up AS was performed according to the PRIAS follow-up regimen. The protocol includes prostate-specific antigen (PSA) testing and digital rectal examination every 3-6 months, and a confirmatory biopsy one year after diagnosis and every three years thereafter. 5 AS could be discontinued due to disease progression (e.g. Gleason upgrading), or for non- oncological reasons (e.g. patient’s preference, or onset of other diseases with a worse prognosis). Outcome measures The primary outcome measure was the proportion of patients who did not meet one or more of the PRIAS inclusion criteria: clinical stage T1c–T2, Gleason sum score >6, two or fewer positive biopsy cores, PSA <10 ng/ml, and PSA density (PSAD) <0.2 ng/ml/ml. Secondary outcome measures evaluated in the overall population included the percentage of patients with metastasis (bone and/or local or distant lymph node) and rates of AS discontinuation because of tumour progression. In addition, outcomes for patients who underwent deferred radical prostatectomy were analysed separately. In this subgroup we evaluated the number of patients who underwent a non-nerve sparing procedure, had positive surgical margins and had unfavourable pathological findings (defined as a pathological T-stage ≥T3 and/or Gleason sum score ≥8). The incidence of biochemical recurrence was also assessed for these patients, defined as PSA ≥0.2 ng/ml two consecutive times after radical prostatectomy. 6 Statistical analysis Univariate and multivariate Cox regression analysis with backward elimination were used for comparisons between PRIAS-eligible and PRIAS-ineligible subgroups. Analysis of time to tumour progression (progression-free survival) was performed with the Kaplan-Meier method. To replace missing values, the multiple imputations package for SPSS V.24.0 (IBM Corp, Armonk, NY, USA) was used.