Background: Cancer-testis antigen NY-ESO-1 is a highly immunogenic melanoma antigen which has been adopted for adjuvant vaccine trials. Three such trials (NCT00124124, NCT00821652, NCT01079741) in high-risk melanoma patients were conducted at our center. Although exact treatment regimens differed between trials, all enrolled patients received an NY-ESO-1-based peptide vaccine. Methods: All vaccine trial patients had histologically-proven melanoma and were without evidence of disease at trial entry; every stage III patient for whom records were available was included for analysis. For comparison, a control cohort of stage III patients who received no adjuvant therapy was identified via our clinicopathological database; every such patient in the database was included for analysis. Recurrence patterns were compared via a chi-squared test. Overall survival (OS) was estimated by Kaplan-Meier analysis, and a Cox proportional hazard model was employed to control for thickness and ulceration, and calculate a hazard ratio (HR). Results: Information was available for 62 vaccinated patients and 116 control patients (median follow-up: 62.8 and 30.5 months, respectively). The vaccine and control groups had the following characteristics, respectively: male (52% vs 62%), median thickness (2.0 vs 2.7), ulceration (41% vs 47%), positive lymph nodes (81% vs 93%), and nodular histotype (52% vs 56%); only lymph node status was significantly different (p = 0.02). At last follow-up, 35 patients (56%) in the treatment group had recurred, compared to 84 patients (64%) in the control group. At first recurrence, vaccinated patients tended to recur with resectable disease more often than control patients (79% vs 60%, p = 0.10), though this result was not significant. OS, however, was significantly longer in the vaccine group compared to the control group (median not reached vs 57.5 months, HR 0.58, 95%CI 0.34 – 0.99, p = 0.047). Conclusions: In this small retrospective cohort of stage III melanoma patients, adjuvant NY-ESO-1 vaccination was associated with improved OS. NY-ESO-1 vaccination may exert an immunologic advantage reflected in a pattern of locally resectable disease upon recurrence compared to diffusely metastatic recurrences.