Introduction & Objectives. Salvage lymph node dissection (SLND) represents a possible treatment option for prostate cancer (PCa) patients experiencing nodal recurrence after local treatment. However, SLND may be associated with intra- and post-operative complications, whereas the oncological benefit may be limited for specific groups of patients. Therefore, we aimed at identifying the optimal candidates for SLND based on pre-operative characteristics.
Materials & Methods. The study included 538 patients who experienced PSA rise and nodal recurrence after RP who underwent SLND at eight tertiary referral centres. Lymph node recurrence was documented by positron emission tomography / computed tomography (PET/CT) scan using either 11C-choline or 68Ga-prostate-specific-membrane-antigen (PSMA) ligand. The study outcome was systemic progression (skeletal and / or visceral metastasis) developed during follow-up after SLND. Multivariable Cox regression analysis was used to develop a predictive model for the study outcome. Predictors consisted of patient age, PSA level at SLND, PSA doubling time (PSADT), diagnostic tracer (11C-choline vs. PSMA), site of nodal positive imaging (pelvic vs. retroperitoneal vs. both), and number of positive spots at PET/CT. Multivariable-derived coefficients were used to develop a novel risk-calculator.
Results. Median patient age, PSA level at SLND, and PSADT were 66 years, 2.3 ng/ml, and 0.4 months, respectively. The PET/CT scan was performed using 11C-choline and PSMA in 401 (75%) and 137 (25%) cases, respectively. Pre-operative imaging was positive in pelvic, retroperitoneal, and pelvic + retroperitoneal regions in 400 (15%), 58 (11%), and 80 (15%) patients. The number of positive spots was 1, 2, and ≥3 in 277 (52%), 120 (22%), and 141 (26%) patients. At a median follow-up of 44 months, 88 (16%) patients experienced systemic progression. At multivariable analysis, age (HR: 0.96; p=0.046), PSA at SLND (HR: 1.02; p=0.006), PSADT (HR: 0.99; p=0.001), PSMA tracer (HR: 0.11; p=0.003), positive imaging in both pelvic and retroperitoneal regions (HR: 1.69; p=0.02), and ≥3 positive spots (HR: 1.71; p=0.01) were significantly associated with M1b-c stage. The multivariable model had a predictive accuracy of 75%. Three pre-operative groups were defined based on the risk estimated by the risk calculator: low-risk (<33%), intermediate-risk (33-66%), and high-risk (>66%). Distant metastasis-free survival at 3 years was significantly different among the three groups (3% vs. 10% vs. 39%, p<0.0001).
Conclusions. We reported the largest series available treated with SLND. At mid-term follow-up, roughly 15% of men developed systemic progression after surgery. We developed a risk calculator based on pre-operative characteristics to discern patients who would benefit the most from SLND from other patients who should be spared from the side effects of SLND.