A Clinical Prognosis Model for Asian Advanced Melanoma Patients Receiving Immunotherapy.

Xuan Wang,Xiegiao Yan,Zhihong Chi,Lu Si,Xinan Sheng,Yan Kong,Lili Mao,Bin Lian,Bixia Tang,Li Zhou,Chuanliang Cui,Jun Guo

DOI: https://doi.org/10.1200/jco.2019.37.15_suppl.e14081

IF: 45.3

2019-01-01

Journal of Clinical Oncology

Abstract:e14081 Background: Immunotherapy PD-1 inhibitor therapy shows relative lower efficacy in Asia especially in acral and mucosal subtypes. The identification of predictive factors of immunotherapy remains a crucial but unmet clinical need. The objective of this study was to evaluate advanced melanoma patients treated with anti-PD-1 in order to identify risk factors for PFS, OS and develop a prognostic scoring system for immunotherapy. Methods: Patients with unresectable melanoma treated with PD-1 inhibitor enrolled on trials between 2015 and 2017 at Peking University Cancer Center were included. A nomogram to predict survival was developed based on a multivariable Cox model. The predictive performance of the model was assessed according to the C-statistic, Kaplan–Meier curve and calibration plots. Results: The discovery cohort comprised 133 patients with unresectable melanoma. 38.3% acral and 21.8% mucosal melanomas were included, 26.3% with liver metastases, 31.6% with an elevated serum LDH, 20.3% harbored BRAF mutations. The median PFS was 4.8 months, based on the Cox model, four factors were selected for the nomogram and assigned specific scores: tumor metastases number≥3, 1; tumor size≥80 mm,1; LDH level, 1 (higher limit of normal range); liver metastases,1. The model achieved relatively good discrimination and calibration, with a C-statistic of 0⋅804 (Table). BRAF mutations were associated with shorter PFS in univariate analysis but not in multivariate analysis. Subtypes were also not related to PFS in this cohort. The median OS was 22.8 months. Univariate analysis identified eight factors were associated with significantly worse OS among 21 factors ( BRAF and subtype included, which not found as risk factors). Based on the Cox model, five factors were selected for the nomogram and assigned specific scores: ECOG > 1, 0.6; Hemoglobin < 120g/ml,0.6; tumor size≥80 mm,1.4; LDH level, 0.5; liver metastases, 1. C-statistic was 0⋅85, with extremely strong predictability (Table). The overall survival rate for patients with a higher score was significantly worse than lower cohort (Score: < 0.6 vs. 1-2 vs. 2-4.1; OS: not reached vs. not reached vs. 8.3m; P < 0⋅001). Conclusions: The nomogram facilitates personalized assessment of prognosis for Asian patients with advanced melanoma with PD-1 inhibitor. [Table: see text]