Efficacy of adjuvant anti-PD-1 antibodies and interferon in patients with nail apparatus melanoma: A retrospective, multicenter study (ADJ-NAIL study).

Abstract

9572 Background: The clinical efficacy of anti-PD-1 antibodies (PD-1) for nail apparatus melanoma (NAM) is less effective than that for advanced cutaneous melanoma. Despite the significant need for effective adjuvant (adj) therapies to improve survival in NAM, the efficacy of adj PD-1 and interferon (IFN) is unknown because of the rarity of NAM. Thus, this study aimed to investigate the efficacy of adj PD-1 and adj IFN therapies for NAM. Methods: Thisretrospective study reviewed data of patients with stage IIB, IIC, and III NAM without adj therapies (observation: OBS), who received adj PD-1 therapy (adj-PD-1) and adj IFN therapy (adj-IFN) after complete resection across 42 Japanese institutions. The Kaplan–Meier analysis and multivariable Cox proportional hazard models were used to estimate survival probabilities. Propensity-score matching (PSM) was employed to adjust for differences in the baseline characteristics between each group. Results: A total of 397 patients with NAM (OBS, n = 219; adj-PD-1, n = 99; adj-IFN, n = 79) were included. The baseline characteristics were significantly different among the three groups in terms of age ( P < 0.001) and stage ( P < 0.001). The other baseline characteristics were comparable. A significant difference was noted in the recurrence-free survival (RFS) among the OBS, adj-PD-1, and adj-IFN groups (5-year RFS 42% vs. 21% vs. 48%; P = 0.02). However, distant metastasis-free survival (DMFS) and overall survival (OS) were not significantly different among the three groups (5-year DMFS 50% vs. 44% vs. 54%; P = 0.15, 5-year OS 59% vs. 46% vs. 58%; P = 0.28). Multivariable Cox proportional hazard models revealed that adj-PD-1 and adj-IFN did not positively affect survival outcomes compared with OBS (adj-PD-1: RFS hazard ratio [HR] 1.14; P = 0.47, DMFS HR 1.04; P = 0.84, OS HR 1.19; P = 0.50, adj-IFN: RFS HR 0.73; P = 0.11, DMFS HR 0.77; P = 0.21, OS HR 0.96; P = 0.84). After PSM, the patient backgrounds were balanced at a 1:1 ratio between the OBS and adj-PD-1 groups (n = 71 each) and adj-IFN groups (n = 75 each). No significant differences were found in the survival outcomes between the OBS and adj-PD-1 groups (5-year RFS 30% vs. 20%; P = 0.96, 5-year DMFS 39% vs. 39%; P = 0.95, 5-year OS 49% vs. 39%; P = 0.53) and between the OBS and adj-IFN groups (5-year RFS 35% vs. 48%; P = 0.06, 5-year DMFS 41% vs. 54%; P = 0.09, 5-year OS 56% vs. 57%; P = 0.73). In patients who experienced a relapse and were treated with nivolumab/ipilimumab combination therapy, the progression-free survival from the initiation of combination was significantly shorter in the adj-PD-1 group than in the OBS group (median PFS 1.8 months vs. 4.0 months; P = 0.03). Conclusions: In NAM, adj-PD-1 and adj-IFN did not improve survival. Furthermore, the use of adj-PD-1 may attenuate the efficacy of nivolumab/ipilimumab combination therapy after relapse.