Abstract
The efficacy of anti-programmed death-1 (PD-1) monotherapy for advanced melanoma has been established, but it is unknown to what extent patients benefit in the real world. In this observational study with nationwide population-based data from the Dutch Melanoma Treatment Registry, we analyzed real-world outcomes of first-line anti-PD-1 monotherapy in advanced melanoma patients diagnosed in 2015 to 2016. Overall survival (OS) was estimated with the Kaplan-Meier method. Competing risks analysis was used to estimate probabilities for second-line treatment, with death as competing risk. With a Cox model, the association of factors with OS was estimated. Patients who received anti-PD-1 monotherapy (n=550) had a median age of 65 years and 502 (95%) patients had an Eastern Cooperative Oncology Group performance status (ECOG PS) of 0-1, 383 (70%) had normal lactate dehydrogenase (LDH), 370 (67%) had stage IV-M1c disease, and in 441 (81%), brain metastases were absent. The median OS was 24 months [95% confidence interval (CI): 20-30 mo]. The median OS of patients normally eligible for phase III trial participation was 31 months (95% CI: 23-not estimable). The BRAF mutation was associated with superior OS. ECOG PS of ≥1, symptomatic brain metastases, and liver metastases were associated with inferior OS and, together with elevated LDH, with death before second-line treatment. Patients with a complete response had a 2-year OS probability from first reported complete response of 92% (95% CI: 86%-99%). Real-world advanced melanoma patients in the Netherlands have benefitted from anti-PD-1 monotherapy. ECOG PS ≥1, symptomatic brain metastasis, liver metastasis, and elevated LDH are important prognostic factors for survival. The additional information that this study provides could help to improve more effective use in the real world.