Observation With or Without Subsequent Salvage Therapy for Pathologically Node-positive Prostate Cancer With Negative Conventional Imaging: Results From a Large Multicenter Cohort

Background and objective: More than 10% of patients with negative clinical metastatic status (cN0M0) on conventional imaging for prostate cancer (PCa) harbor lymph node involvement (pN+) at final pathology following radical prostatectomy (RP) and lymphadenectomy. Our aim was to assess outcomes of initial observation for cN0M0 pN+ PCa and identify prognostic factors that may help in clinical decision-making. Methods: We performed a retrospective multicenter study of patients with cN0M0 PCa on conventional imaging (computed tomography and/or magnetic resonance imaging, and a bone scan) who were found to have pN+ disease at RP between 2000 and 2021. Biochemical recurrence (BCR) and systemic progression/recurrence were the primary outcomes. Kaplan-Meier curves and Cox proportional hazards model were used for survival and multivariate analysis. Key findings and limitations: A total of 469 men were included in this retrospective multicenter trial. Median prostate-specific antigen (PSA) was 10.1 ng/ml (interquartile range [IQR] 6.6–18.0). Among these patients, 56% had grade group ≥4, 53.7% had stage ≥pT3b, 42.6% had positive margins, and 19.6% had PSA persistence. The median number of positive nodes and of nodes removed were 1 (IQR 1–3) and 20 (14–28), respectively. At median follow-up of 41 mo, 48.5% experienced BCR. The 5-yr BCR-free survival rate was 31.7% (95% confidence interval [CI] 26.33–37.1%). Salvage treatments were needed in 211 patients and included radiotherapy (RT; n = 53), RT + androgen deprivation therapy (ADT; n = 88), ADT alone (n = 68), and salvage lymphadenectomy (n = 2). The 5-yr estimated survival rates were 66.3% (95% CI 60.4–72.1) for metastasis-free survival, 97.7% (95% CI 95.5–99.8%) for cancer-specific survival, and 95.3% (95% CI 92.4–98.1%) for overall survival. On multivariable analysis, PSA persistence was an independent predictor of BCR (odds ratio [OR] 51.8, 95% CI 12.2–219.2), exit from observation (OR 8.5, 95% CI 4.4–16.5), and systemic progression (OR 3.0, 95% CI 1.771–4.971). Conclusions: Initial observation in the management of pN+ cN0M0 PCa is feasible and has excellent survival rates in the intermediate term. Patients with worse disease features, especially PSA persistence, have a higher likelihood of recurrence and progression and may be candidates for more aggressive upfront management. Patient summary: We investigated the value of initial observation for men with prostate cancer with negative scan findings for metastasis who were then found to have positive lymph nodes after surgery to remove the prostate. Our results show that initial observation is a good option for patients with less aggressive prostate cancer features.