The Impact of Delayed Radical Prostatectomy on Recurrence Outcomes After Initial Active Surveillance: Results from a Large Institutional Cohort

Abstract

Background and objectiveActive surveillance (AS) of prostate cancer (PCa) involves regular monitoring for disease progression. The aim is to avoid unnecessary treatment while ensuring appropriate and timely treatment for those whose disease progresses. AS has emerged as the standard of care for low-grade (Gleason grade 1, GG1) PCa. Opponents are concerned that initial undersampling and delay of definitive management for patients with GG2 disease may lead to adverse outcomes. We sought to determine whether the timing for definitive management of GG2 PCa, either upfront or after initial AS, affects recurrence outcomes after radical prostatectomy (RP). MethodsParticipants were diagnosed with cT1–2N0/xM0/x, prostate-specific antigen (PSA) <20 ng/ml, and GG1–2 PCa between 2000 and 2020 and underwent immediate RP for GG2 or AS followed by delayed RP on upgrading to GG2. The outcome was recurrence-free survival (RFS) after surgery, with recurrence defined as either biochemical failure (2 PSA measurements ≥0.2 ng/ml) or a second treatment. Multivariable Cox proportional-hazards regression models were used to calculate associations between the timing for definitive RP and the risk of recurrence, adjusted for age at diagnosis, percentage of positive biopsy cores (PPC), PSA density, PSA before RP, year of diagnosis, surgical margins, genomic risk score, and prostate MRI findings.Key findings: Of the 1259 men who met the inclusion criteria, 979 underwent immediate RP after diagnosis of GG2, 190 underwent RP within 12 mo of upgrading to GG2 on AS, and 90 men underwent RP >12 mo after upgrading to GG2. The 5-yr RFS rates were 81% for the immediate RP group, 80% for the delayed RP ≤12 mo, and 70% for the delayed RP >12 mo group (univariate log-rank p = 0.03). Cox multivariable regression demonstrated no difference in RFS outcomes between immediate RP for GG2 disease and delayed RP after upgrading on AS. PPC (hazard ratio [HR] per 10% increment 1.08, 95% confidence interval [CI] 1.02–1.15; p = 0.01) and PSA before RP (HR 1.06, 95% CI 1.03–1.09; p < 0.01) were significantly associated with the risk of recurrence. Conclusionsand clinical implications: PPC and PSA before RP, but not the timing of definitive surgery after upgrade to GG2, were associated with the risk of PCa recurrence after RP on multivariable analysis. These findings support the safety of AS and delayed definitive therapy for a subset of patients with GG2 disease. Patient summaryIn a large group of 1259 patients with low-grade prostate cancer, we found that delaying surgical treatment after an initial period of active surveillance resulted in no differences in prostate cancer recurrence. Our results support the safety of active surveillance for low-grade prostate cancer.