Abstract PO-048: Combined single-T cell clonotyping and phenotyping defines a predictive response signature that identifies patients more likely to respond to therapy

Immunomodulatory antibodies such as checkpoint inhibitors are showing very promising results in the clinic, but only a minority of patients respond to treatment and it is still challenging to identify these subpopulations. Prospective identification of likely responders would greatly facilitate the clinical application of immunotherapies and offers the promise of guiding combination treatments in a principled manner. Various markers, such as microsatellite Instability, tumor mutational burden, general T cell infiltration, and PD-L1 expression have been suggested to be applied for patient selection; however, their ability to enrich for responder patients remains limited. Recently, research on PD-1 therapies suggested the importance of the presence of specific T cell clonotypes and their expansion for predicting responses using single-cell immune cell clonotyping and phenotyping. We hypothesized that patients with T cells that expand after PD-1 blockade are more likely to have a positive clinical outcome. Using published single-cell RNA sequencing data, we identified T cells that have expanded after PD-1 blockade and we then extracted their gene signatures prior to treatment, both for CD4+ and CD8+ T cells. These signatures therefore predict the ability of T cells to expand post-treatment. Using additional data from three independent clinical trials with a PD-L1 antibody, we assigned expansion scores to each patient that indicated how strongly they expressed the signatures for either CD4+ or CD8+ T cell expansion. We found that patients with a higher CD8 expansion score exhibited greater progression-free survival. Surprisingly, we also observed that patients with a higher CD4 expansion score showed decreased progression-free survival which may hint at the involvement of regulatory T cells. Further, we present a novel general concept for identifying predictive response signatures to immunomodulatory antibodies using our single-cell Drug Intelligent Science (DIS™) platform. It combines immune cell clonotyping and phenotyping with functional profiling of the response to drug treatment. By investigating correlations between T cell clones responding to the ex vivo treatment conditions and their phenotypes in a non-treated condition, single cell analysis could provide a predictive response signature for the treatment. This new approach generates biomarker hypotheses that can then be tested in patient populations to determine which patients are more likely to respond to a treatment. In summary, we have demonstrated that examining T cell clonotypes can help to define predictive response signatures pre-treatment that identify patients that are more likely to respond to immunotherapy. Our single-cell DIS technology offers a novel platform to mine these relationships and generate patient-specific clinical response predictions for effective patient stratification. Citation Format: Dean Lee, Julianna Crivello, Ross Fulton, Zachary Duda, Alexandra Staskus, Charina Ortega, Monika Manne, Roshan Kumar, Francisco Adrian, Liang Schweizer, Matt Delince, Andreas Raue. Combined single-T cell clonotyping and phenotyping defines a predictive response signature that identifies patients more likely to respond to therapy [abstract]. In: Proceedings of the AACR Virtual Special Conference on Tumor Heterogeneity: From Single Cells to Clinical Impact; 2020 Sep 17-18. Philadelphia (PA): AACR; Cancer Res 2020;80(21 Suppl):Abstract nr PO-048.