Abstract LB402: From cellular communication networks to rational selection of reagents for combination trials

Abstract Novel therapeutic agents targeting solid cancers entering Phase I have merely a 3-4% probability of achieving clinical approval. Given the existence of over 300 oncology-related therapeutic-agent families, exhaustive clinical evaluation of individual agents—and particularly, combination of agents—across all cancer types is unfeasible. There is an urgent need for better methodologies to identify viable targets both for monotherapies and for combination therapies. We developed a unique empirical toolset and a novel analytical methodology to investigate the cellular communication network (CCN) that protects tumors from an immune-mediated assault. Our approach uses high-dimensional flow cytometry (FC - up to 18-colors) to precisely quantify all immune, tumor and stromal cells found in human tumors. FC-sorted cell subsets are then RNA-sequenced, providing highly reproducible deep transcriptomic profiles (∼15,000 genes) for both abundant and rare cellular constituents. Integration of such data enables construction of CCN models which quantitatively evaluate the activity of all communication layers (CLs, e.g., PD1-PDL1, VEGF-VEGFR) within individual patient tumors. While immune checkpoint inhibitors (ICIs) target specific receptors or ligands, their therapeutic mechanism is the disruption of specific cellular CLs within tumors. Currently, no methodology directly quantifies this communication. Our analyses across five different cancers (N=42 samples) demonstrated robust correlations (R2>0.92) between the mean activity of specific communication layers (CLs) and observed clinical responses to therapeutic agents targeting those CLs. These strong correlations were consistently observed for both immunotherapeutic agents (as ICIs) and non-immunologic agents, as anti-angiogenic agents targeting VEGF-VEGFR signaling. Utilizing a similar approach, we could also precisely predict the clinical benefit of individual patients (N=13) from diverse cancers, treated with various treatments, which all had a PD1-backbone (p=0.0012). This paradigm shift from conventional biomarkers (as PDL1 staining) to complex biomarkers quantifying intratumoral cellular communication networks may enhance predictive accuracy for therapeutic responses, both at the single cancer level and at the single patient level. Stratification of potentially efficacious therapeutic agents, prior to clinical testing, could facilitate rational design of clinical trials and patient-tailored therapeutic interventions with better chances to succeed, saving time, money and patients' lives. Citation Format: Ilan Volovitz, Reut Yosef, Avia Keidar, Yael Silberberg, Kelly Lipczyc, Efrat Elis, Ravit Geva. From cellular communication networks to rational selection of reagents for combination trials [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr LB402.