Background: Immune checkpoint inhibitor (ICI) therapy for advanced melanoma in the first-line setting often results in durable responses and prolonged overall survival (OS) after achieving a complete response (CR). While generally well-tolerated, there are financial toxicities and treatment related adverse events (TRAEs) associated with ICI use. There are limited data to inform the optimal duration of therapy after CR, and how specific patient factors may lead to variable OS in patients (pts) with CR. Methods: Pts with locally advanced stage III or stage IV melanoma who started first-line ICI therapy with pembrolizumab, nivolumab, or ipilimumab/nivolumab (ipi/nivo) and achieved CR were identified at a single institution under an Institutional Review Board approved protocol. Pts were only included in the sample if they had initiated treatment by 2013 and discontinued therapy by February of 2021 and had documented follow-up at least 1 year after discontinuing ICI therapy. Recurrence rates were determined by the presence of any recurrence after CR. Pts were analyzed by baseline stage (III vs IV), BRAF status, use of steroids for TRAEs, time from treatment start to CR, and total duration of therapy. OS was determined by time from treatment start to death or date of last follow-up. Results: 78 Pts with CRs from ICI therapy who discontinued treatment were identified at the University of Pennsylvania. Median follow-up time was 63 months from ICI treatment start. Median age at treatment start was 65 years (range 33-94). 51 Pts (68%) had an ECOG performance status (PS) of 0 vs 24 (32%) with ECOG PS of 1-2. 21 Pts (27%) had locally advanced stage III and 57 (73%) had stage IV disease. Of stage IV pts, 8 were M1a, 16 M1b, 22 M1c, and 11 M1d. 61 (78%) pts were treated with pembrolizumab, 5 (6%) nivolumab, 12 (15%) ipi/nivo. 34 pts (44%) were treated with steroids for TRAEs. Of the 78 total pts, 13 (16.7%) developed a recurrence after stopping treatment. The median time to recurrence was 57 months (range 9-110). Melanoma specific survival was 96%. Of the pts with recurrence, 3 died of melanoma, 9 are disease free after additional local and/or systemic treatment and one patient is alive with disease. There was no significant difference (p< 0.05) in OS based on baseline stage, BRAF status, or use of steroids for TRAEs. Pts were treated for a median of 6 months (range 1 day to 24 months). Conclusions: With over 5 years median follow-up time, the majority of pts with advanced melanoma who attain a CR from first-line ICI therapy remain disease free after treatment discontinuation. Patient factors including stage at treatment initiation, BRAF status, and steroid usage during treatment were not associated with significant variation in OS in this cohort. Most pts with recurrences remain disease free after additional local and/or systemic treatment. Few melanoma specific deaths occurred in pts with a CR to ICI who discontinued treatment.