Abstract
BACKGROUND: Anti-PD-1-based immunotherapy has improved outcomes in stage IIB to IV resected melanoma patients in clinical trials. However, little is known about real-world outcomes, prognostic factors and patterns of relapse. METHODS: This is a retrospective multicenter observational study including patients with resected melanoma treated with subsequent anti-PD-1-based adjuvant immunotherapy. Data on clinical and demographic characteristics, delivered treatment, prognostic factors, time and pattern of relapse were collected. RESULTS: We included 245 patients from eight centers; 4% of patients were at stage IIB-C, 80% at stage IIIA-D and 16% at stage IV. Recurrence-free survival (RFS) rates at 18 and 36 months were 60% and 48%, respectively, with a median RFS of 33.7 months. Prognostic factors associated with recurrence were melanoma primary site (HR 2.64, 95% CI 1.15-6.01) and starting adjuvant therapy more than 12 weeks after the last resection (HR 1.68, 95% CI 1.13-2.5); presence of serious immune-related adverse events was associated with better RFS (HR 0.4, 95% CI 0.19-0.87). Early relapses accounted for 63% of the total recurrences, with a higher number of metastatic sites (18%); in contrast, late relapses presented more frequently with brain metastases (20%). CONCLUSIONS: In our patients with resected melanoma who underwent anti-PD-1-based adjuvant immunotherapy, survival outcomes were worse than those reported in clinical trials. Primary melanoma site and time interval between the last resection and the start of adjuvant therapy were associated with survival.