MP22-16\u2003LONG-TERM OUTCOMES AND PATTERNS OF RECURRENCE IN PATIENTS WITH CLINICAL LYMPHADENOPATHIES UNDERGOING RADICAL PROSTATECTOMY AS PART OF A MULTIMODAL TREATMENT

Abstract

INTRODUCTION AND OBJECTIVES: Although radical prostatectomy (RP) has been proposed as a feasible and effective therapeutic approach in prostate cancer (PCa) patients with enlarged lymph nodes at imaging (cN1), no data are available on the long-term outcomes and patterns of recurrence of surgically managed patients with cN1 disease. METHODS: Overall, 149 patients with lymphadenopathies in the pelvis and/or retroperitoneum detected by CT scan (n=92; 61.7%) or MRI (n=57; 38.3%) treated with RP and nodal dissection between 2005 and 2013 with at least 5 years of follow-up were identified. Clinical recurrence (CR) was defined as the onset of metastases. The site of the first CR was stratified as: local (i.e., prostatic fossa and regional nodes), retroperitoneal and distant (i.e., bone and visceral metastases). Cox regression analyses assessed predictors of distant metastases after adjusting for adjuvant treatments. Poisson smoothed cumulative incidence plots assessed 10-year cancer-specific mortality (CSM) and other cause mortality (OCM) after stratifying patients according to predicted risk of 8-year distant metastases. RESULTS: Median number and size of lymphadenopathies were 1 and 14 mm and 25 (20%) patients had retroperitoneal involvement. The median number of positive nodes was 7 and 117 (78%) patients had lymph node invasion. Median follow-up was 86 months. Overall, 91 (61%) patients received adjuvant hormonal therapy. Overall, 53, 30 and 11 patients had CR, CSM and OCM. The 10-year CR-free rates were 56 and 78%. Overall, 11 vs. 6 vs. 21% patients had local vs. retroperitoneal vs. distant CR. The 10-year CSM-free rates were 66 vs. 62 vs. 44% for local vs. retroperitoneal vs. distant CR. Grade group 4-5 and retroperitoneal involvement were associated with distant CR (P⩽0.01). At 10-year, 21 and 7% patients experienced CSM and OCM. The risk of dying from OCM was the same as CSM in patients with a 8-year distant metastases probability <15%. Conversely, 34 vs. 5% patients died from PCa vs. OCM among men with a risk ≥15%. CONCLUSIONS: Although the risk of CSM was overall higher than OCM, in men with an estimated risk of distant metastases <15% the risk of CSM did not exceed the risk of OCM. These findings have important implication for patient selection and for the administration of timely salvage therapies at recurrence. Figure. No caption available. Source of Funding: none