Predictive Value Of Circulating B Cells And T Cell Subsets In Melanoma Patients Treated With Neoadjuvant Ipilimumab And Interferon.

e15036 Background: Patients with locally/regionally advanced melanoma were treated on a clinical trial with a neoadjuvant combination of ipilimumab (ipi) and high dose IFNα2b (HDI) (Tarhini et al, JITC 2018). In this study, immune cell composition in peripheral blood samples collected at various time points was measured to determine any correlation with clinical outcomes and investigate the immune modulating effect of the combination therapy. Methods: Patients were randomized to neoadjuvant ipi at 3 mg/kg or 10 mg/kg, both given in combination with HDI. Tumor radiologic responses were designated as complete (CR), partial (PR), stable disease (SD) or disease progression (PD). Pathologic complete response (pCR) was defined as absence of viable tumor on histologic assessment. Peripheral blood mononuclear cells (PBMC) from treated patients (N = 28) were tested at baseline (before initiating ipi-HDI), then at 6-weeks, 3-months and 12-months (following neoadjuvant ipi-HDI). High complexity (14-color) flow cytometry analysis was performed to detect key immunological biomarkers including myeloid derived suppressor cells (MDSCs), B cells, regulatory T cells (Tregs), PD-1 and TIM3 expression on T-cells, and differentiation of T-cells into Th1, Th2 or Th17 phenotype at different time points during systemic immunotherapy. Statistical significance was determined using R-package employing Kruskal’s test. Results: Lower levels of peripheral Tregs (p = 0.02), MDSCs (p = 0.05), and CD4 effector memory cells (p = 0.04) at 3-months post treatment correlated with radiologic response. In addition, lower change from baseline at 3 months in CD4/CD8 ratio (p = 0.04), levels of Tregs (p = 0.01) and CD4 effector memory cells (p = 0.02) was associated with radiologic response. Patients exhibiting pCR had significantly lower Tregs (p = 0.04) at 6-months post treatment and significantly higher CD8 central memory cells at both 3 months (p = 0.04) and 12 month time-points (p = 0.01) as compared to patients without pCR. Finally, patients without pCR had significantly lower change from baseline in CD19 B cells at 6 months (p = 0.01) and 12 months (p = 0.04) as compared to patients with pCR. Conclusions: Our data demonstrates that the levels of immunosuppressive cells including Tregs and MDSCs in periphery are negatively associated with response. Higher levels of CD8 memory cells and B cells on-treatment are associated with clinical benefit.