T145: Comprehensive flow cytometry tracking of regulatory T cells and other lymphocyte subsets during HD IL-2 therapy for melanoma

High dose IL-2 (HD IL-2) has been extensively used as an immunotherapy against metastatic melanoma However, why HD IL-2 is effective only in a subset of patients and whether predictive biomarkers, before or early during the course of therapy, can be used to improve response rates remain unresolved. In addition, it has been found that IL-2 therapy potently expands CD4+CD25+Foxp3+ T-regulatory cells (Tregs) but how Treg cell levels, phenotype, and function change and whether specific subsets of Tregs are activated and expanded during HD IL-2 therapy is remain unclear. In this study, we performed comprehensive multi-parameter FACS analysis of patient blood before and two days after the last bolus of IL-2 infusion during cycle 1 of HD IL-2 therapy. Two lymphocyte subsets were found to expand the most during the first cycle of IL-2 therapy: CD4+CD25+Foxp3+ Tregs expressing an activation marker, inducible costimulator (ICOS), and CD3-CD56hiCD16loPerforin+ NK cells. ICOS+ Tregs expressed significantly higher levels of CD25, Foxp3 and had a more activated phenotype than ICOS− Tregs as indicated by lower levels of CD45RA and CD127 expression. Further phenotypic characterization revealed a more suppressive phenotype on ICOS+ Treg with higher expression levels of CD39, CD73, and TGF-β/LAP than ICOS− Treg. ICOS+ Tregs were also the predominant Treg cells that secreted IL-10 and have potent T-cell suppressor function. Majority of ICOS+ Tregs from HD IL-2-treated patients were Ki67+ and exhibited an enhanced proliferative response to IL-2 ex vivo relative to ICOS− Tregs. Functional analysis revealed that ICOS+ Tregs secreted little IFN- and IL-2 in comparison to CD4+Foxp3 – cells. Furthermore, analysis on 38 IL-2-treated patients at MD Anderson, we found that non-responders had a significantly higher degree of ICOS+ Treg expansion than responders during the first cycle of IL-2 therapy, while no significant changes in the ICOS− or bulk Treg population. In conclusion, our data suggests that tracking changes in ICOS+ Tregs early during the course of HD IL-2 therapy may be a new predictive biomarker of clinical outcome.