Prognostic implications of glycemia in non-diabetic patients with metastatic melanoma undergoing immunotherapy.

Abstract

9530 Background: Immune checkpoint inhibitors (ICIs) have revolutionized the cancer therapeutic landscape, significantly improving the survival of patients (pts) with advanced malignancies. Previous studies have shown that diabetic pts have a higher risk of cancer-related mortality compared to those without diabetes. This retrospective study aimed to investigate the prognostic impact of glycemia on ICI treatment outcomes in non-diabetic metastatic melanoma pts. Methods: Glycemic levels were assessed at three distinct time points within a two-week period prior to the initiation of ICI therapy in 1079 non-diabetic metastatic melanoma pts treated with anti-PD1 and anti-CTLA4 either as monotherapy or in combination. Blood glucose concentrations were determined enzymatically using the cobas c-501 system (Roche, normal range 70-110 mg/dL). Interleukin-6 (IL-6) levels were assessed in 378 pts using Electrochemiluminescence Immunoassays and gene profiling analysis was performed on 95 baseline RNA using NanoString IO360 panel. Pts characteristics are listed in Table 1. Spearman's correlation was used to assess the association between variables. Survival rates were analyzed using the Kaplan-Meier method. Results: ROC curve analysis identified a blood glucose cut-point of 93.33 mg/dL. Pts with low glycemia had a better overall survival (median: 27.7 vs 14.5 months, HR=0.68, p < 0.0001) and progression free survival (median: 7.4 vs 4.3 months, HR=0.74; p < 0.001) compared to pts with elevated glycemia. This trend was confirmed in subgroups analysis (anti-PD1; anti-CTLA4), except for pts treated with the combination of anti-PD1 plus anti-CTLA4 as well as in line of treatment stratification (first, second and ≥3). Glycemia was found to be positively associated with elevated IL-6 levels (rho 0.16, p<0.01). Transcriptomic analysis showed an association between glycemia and genes related to inflammatory activity (S100A12; CD40) and cell cycle regulation (CNTFR; PTEN). Glycemia predicts prognosis independently of LDH and line of treatment in multivariate analysis. Conclusions: Elevated glycemia is associated with poor prognosis in pts with metastatic melanoma treated with ICIs. Biomarker analysis revealed an association between glycemia levels with pro-inflammatory cytokine IL-6 and genes linked to inflammation and cell cycle progression. Further investigations are needed in order to endorse the data and validate the glycemia cut-point. Patient characteristics. Patient characteristics N = 1079 Median age 59 (range 19-91) Gender: female/male, n (%) 448 (41)/631 (59) CNS metastases at baseline, n (%) 201 (20) BRAF Status, n (%) Wild type, n (%) 596 (55) Mutation, n (%) 404 (38) NA, n (%) 80 (7) T2DM, n (%) 47 (4) ORR, n (%) 261 (24) Anti-PD1, n (%) 646 (60) Anti-CTLA4, n (%) 272 (25) Anti-PD1+ Anti-CTLA4, n (%) 161 (15) First line, n (%) 623 (58) Second line, n (%) 325 (30) Third line ≥, n (%) 131 (12)