SALVAGE LYMPH NODE DISSECTION FOR PATIENTS TREATED WITH RADICAL PROSTATECTOMY WITH BIOCHEMICAL RECURRENCE AND IMAGING-DETECTED NODAL METASTASES.

Manuela Tutolo, NazarenoSuardi, Milan, Italy, Jeffrey Karnes, Rochester, MN, Steven Joniau, Leuven, Belgium, KarimTouijer, New York, NY, DaniarOsmonov, Kiel, Germany, Alberto Briganti, Marco Bianchi, Milan, Italy, Hein Van Poppel, Leuven, Belgium, PatrizioRigatti, Klaus Peter Junemann, Kiel, GermanyFrancesco Montorsi, Milan, Italy, Alberto Briganti

INTRODUCTION AND OBJECTIVES: The aim of the study was to determine the outcome of patients with biochemical (BCR) and clinically recurrent nodal disease submitted to salvage lymph node dissection (sLND).

METHODS: The study included 162 patients from 5 tertiary referral centers affected by BCR after RP associated with nodal recurrence detected at either (11C) choline PET/TC scan or conventional imaging. All patients received extended pelvic +/- retroperitoneal sLND between 2002 and 2011. Pre-op variables were used to predict complete response to sLND (defined as a the first post-op PSA <0.2 ng/ml ), subsequent BCR (defined as PSA ≥0.2 ng/ml and rising) and clinical recurrence (CR). The KaplanMeier method was used to assess post-op outcomes. Univariable (UVA) and multivariable (MVA) Cox regression analyses were used to identify outcome predictors.

RESULTS: Mean and median PSA at sLND was 3.6 and 1.9 ng/ml, respectively. Mean time to BCR after RP was 29.2 months, respectively. Patients were submitted to sLND with a mean number of 24.6 nodes removed(median:20; range 3-87). The extent of LND was: pelvic only, retroperitoneal only and pelvic + retroperitoneal in 76 (46.9%), 2 (1.2%) and 84 (51.9%) patients. Positive nodes were found in 132 patients (81.4%). The mean number of positive nodes was 6.1 (median:2; range 0-66). Overall, 66 (40.7%) patients showed complete response at the first PSA test after surgery at a median follow-up of 40 months after sLND. Inthese patients, the 3- and 5-year BCR-free survival was 59 and 40%, respectively. This results in a number of 11 patients (out of 162) who are BCR-free without ADT at 5-year follow-up. At MVA Cox regression analyses, only time from RP to BCR was an independent predictor of complete PSA response (OR:1.01;p=0.04). In the overall population, the 5-year CR- and CSM-free survival were 47 and 86%, respectively. At MVA analyses, only the number of positive nodes (HR:1.05; p<0.001) and the complete PSA response (HR:0.5; p=0.04) represented independent predictors of CR after sLND. Patients without complete PSA response and/or with >2 nodes at sLND had a 2-fold increased probability of CR at 5 years.

DISCUSSION: sLND is associated with complete PSA response in 40% of highly selected patients with nodal recurrence after RP. Of these, only 40% are BCR-free at 5 years, therefore allowing for at least a delay in the use of further treatments.

CONCLUSIONS:Men with low volume disease and complete response to surgery seem to benefit the most from this surgical approach in terms of clinical progression.