Checkpoint inhibitors (CPIs) have demonstrated a heterogenous spectrum of response and disease progression that may not be fully captured by conventional response criteria, such as a limited degree of progression, known as oligoprogression, which could benefit from local treatment. We retrospectively analyzed data from all patients diagnosed with metastatic melanoma, who received CPI between January 2006 and March 2018 at Royal Marsden. We enrolled 36 patients who experienced progression in a maximum of 3 metastatic sites, after achieving disease control from therapy with CPI, and were radically treated with the locoregional approach. We carried out Kaplan-Meier analysis to obtain progression free-survival post-first oligoprogression (PFS-PO1), overall survival (OS) post-first oligoprogression, and OS estimates. The median time to oligoprogression from the start of CPI was 12 months. At a median follow-up of 34 months, the median PFS-PO1 was 32 months, with 50% of patients not progressed at the time of the data cutoff. The median OS-post-first oligoprogression was not reached. At a median follow-up of 52 months (from the first cycle of CPI), the median OS was not reached, with 75% of patients alive at the time of analysis. Univariate and multivariate analyses demonstrated that baseline American Joint Committee on Cancer stage IV M1a or M1b is associated with a longer PFS-PO1 compared with stage M1c or M1d. We observed that local therapy for oligoprogression after CPI can result in durable disease control, suggesting that locoregional treatment should be considered in patients being treated with immunotherapy. However, prospective evaluation, perhaps in randomized trials, is needed.