Abstract A067: PD-1 blockade enhances OX40 expression on regulatory T-cells and decreases suppressive function through induction of phospho-STAT3 signaling

Abstract

Abstract Blockade of the co-inhibitory receptor PD-1 has had unprecedented success in the treatment of metastatic melanoma, with response rates of approximately 40% in previously untreated patients. However, since the majority of patients fail to respond, a better understanding of the mechanisms of action involved in response/resistance and novel approaches to overcome resistance are needed. We evaluated T-cells present in peripheral blood samples of resected stage III/IV melanoma patients treated with adjuvant nivolumab. Data from intra-patient, paired pre-treatment/post-treatment samples showed a significant (p<0.05) increase in phosphorylated STAT3 (pSTAT3) in patients without relapse, but not in patients relapsing. Increases in regulatory T-cells (Tregs) (p<0.05) in relapsers, but not in non-relapsers, and OX40 expression on Tregs (p<0.05) in both relapsers and non-relapsers were also observed. Based on these data we evaluated the ability of PD-1 blocking antibodies to induce pSTAT3 signaling in vitro. Culturing with anti-PD-1 significantly increased levels of pSTAT3 in T-cells, including Tregs, both at the basal state (p<0.05) and post-CD3/CD28 activation (p<0.01). Additionally, culturing T-cells with PD-1 blocking antibodies increased the production of the STAT3 regulated cytokine IL-10 (p<0.05), the number of Tregs (p<0.01) and Treg expression of OX40 (p<0.05). Increases in all these markers/populations were abrogated with the addition of a STAT3 inhibitor (p<0.01). Similar increases in the number of Tregs and the expression of OX40 were also found when T-cells were treated in vitro with recombinant IL-10. Addition of IL-10 neutralizing antibodies ameliorated upregulation of Treg numbers and OX40 expression resulting from PD-1 blockade. Additionally, serum levels of IL-10 were found to be significantly elevated in non-relapsing patients relative to relapsers (p<0.05). RNA-seq analysis comparing Tregs from relapsers to non-relapsers further supported the observed increase in Tregs resulting from nivolumab treatment, with pathway analyses demonstrating significantly increased cell cycle, DNA replication, mitosis and other proliferation related pathways in Tregs from non-relapsed patients. Next we evaluated changes in the function of patient Tregs in an allogenic mixed lymphocyte suppression assay. Tregs from non-relapsers had a significant reduction in suppressive capacity post-nivolumab treatment in paired analyses (p<0.05), while Tregs from relapsers had no significant change. Post-treatment samples from relapses were also found to be significantly more suppressive (p<0.01) than those of non-relapsers. Additionally, decreased percentages of CCR4+CD45RA+ Tregs, previously described as having enhanced suppressive capacity, were found in non-relapsers post-treatment, but not in relapsers. A similar decrease in CCR4+CD45RA+ Tregs was achieved in vitro using OX40 agonist antibodies. Therefore, we propose that PD-1 blockade triggers STAT3 signaling, a hereunto unknown effect. These data support a model in which STAT3 induction by PD-1 blockade results in IL-10 production, which induces proliferation of Tregs with reduced suppressive capacity and triggers OX40 expression. Based on these results and previous findings of others demonstrating that ligation of OX40 expressed on Tregs hinders suppressive capacity, combining OX40 agonist antibodies may enhance the effectiveness of PD-1 blockade by reducing Treg suppression. Citation Format: David M. Woods, Rupal Ramakrishnan, Andressa L. Sodré, Anders Berglund, Jeffrey Weber. PD-1 blockade enhances OX40 expression on regulatory T-cells and decreases suppressive function through induction of phospho-STAT3 signaling [abstract]. In: Proceedings of the Second CRI-CIMT-EATI-AACR International Cancer Immunotherapy Conference: Translating Science into Survival; 2016 Sept 25-28; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2016;4(11 Suppl):Abstract nr A067.