Timo Soeterik

38 CHAPTER 2 DISCUSSION In the present cohort of 1000 patients on AS, 49% did not meet one or more of the PRIAS criteria for (very) low-risk PCa. These PRIAS-ineligible patients experienced tumour progression significantly earlier compared with the PRIAS-eligible subgroup. In addition, PRIAS-ineligible patients had a higher risk of unfavourable histological and surgical outcomes following deferred radical prostatectomy. Baseline PSAD ≥0.2 ng/ml/ml was a strong individual predictor of (unfavourable) AS outcomes, and was associated with earlier tumour progression, a higher risk of unfavourable outcomes after deferred surgery, and a higher risk of metastasis. A major issue related to AS is the extent to which AS eligibility criteria may be expanded. 7,8 Although there is limited evidence on the safety of AS for patients with higher-risk PCa, the present results indicate that the selection criteria are already being expanded in daily practice. These findings suggest that patients (and their treating clinicians) might be willing to trade off a survival benefit to avoid treatment-related comorbidities such as urinary incontinence and erectile dysfunction. It is also possible that urologists may not be fully aware of the potential consequences of selecting higher- risk patients. Either way, this study provides additional information that is highly relevant for clinicians counselling these patients. Three previous studies evaluating AS outcomes for patients with lower- and higher- risk PCa reported a significantly higher rates of metastasis among intermediate-risk patients. For example, Bul et al reported a 10-year metastasis-free survival of 99.7% in low-risk patients and 96.4% in intermediate-risk patients (log-rank p = 0.03). 9 In the second study, Godtman et al reported that patients with low and intermediate-risk disease had a higher risk of AS failure (defined as death from PCa, PCa metastasis and/or biochemical progression and/or initiation of hormonal therapy after radical treatment) compared with very low-risk patients: HR 4.8 (95% CI 2.44 - 9.33). 10 A higher risk of metastasis in intermediate-risk patients was also observed in the Sunnybrook cohort. Musunuru et al reported an inferior 15-year metastasis-free survival in the intermediate- risk group compared with the low-risk group (HR 3.14, 95% CI 1.51 - 6.53). 11 Conflicting results were reported by Nyame et al, as they did not observe a significant difference in metastasis-free survival comparing men with intermediate/high-risk PCa and very low/ low-risk (HR 1.50, 95% CI 0.16 - 14.5). 12 The present study is unique in that we compared patients selected according to strict AS criteria based on the PRIAS protocol, with higher-risk off-protocol selected patients. In addition, we performed univariate analysis for all the separate criteria, which provided relevant information regarding the predictive value of each factor. Comparison of PRIAS-eligible and PRIAS-ineligible patients showed no significant difference regarding rates of metastasis. However, when evaluating all the criteria separately, PSAD ≥0.2 ng/ml/ml was a strong individual predictor for the risk of metastasis and was also strongly associated with higher rates of other unfavourable outcomes. This suggests that in this patient cohort, using the PRIAS criteria classification overall was