Abstract 2570: Reprogramming T-cell networks to overcome immune evasion in ovarian cancer relapse

Abstract Ovarian cancer (OC) remains a major clinical challenge due to its high recurrence rates and resistance to standard therapies. High-grade serous ovarian cancer (HGSOC), the most common and aggressive OC subtype, often presents with tumor-infiltrating lymphocytes (TILs), yet immune checkpoint blockade (ICB) therapies have shown limited efficacy. Our studies aim to dissect the spatiotemporal evolution of the OC tumor microenvironment (TME), reprogram T cell networks, and develop next-generation immunotherapies to restore anti-tumor immunity. We integrated high-resolution spatial proteomics, single-cell transcriptomics, and computational modeling to map TIL dynamics across primary, recurrent, and treatment-resistant OC. Preclinical models, including syngeneic BRCA1-mutant and wild-type mouse models of OC recurrence, are employed to investigate tumor-intrinsic and TME-driven mechanisms of immune suppression during OC relapse. We utilized digital pathology multiplex immunofluorescence (mIF) to classify 697 ovarian cancer (OC) specimens from five independent cohorts, offering the most extensive CD8+ T cell-based immune profiling to date. Our predictive algorithm identified distinct tumor immune phenotypes in OC (i.e. purely inflamed, mixed inflamed, excluded, and desert) based on TIL infiltration and spatial organization, which correlate with HRD status and predict therapeutic outcomes. We observed significant immune and molecular heterogeneity between these tumor immune phenotypes and their dynamics during disease recurrence. Inflamed HRD tumors, maintained T cell-myeloid niches post-chemotherapy. In preclinical models, this was associated with restoration of genomic rearrangements. Recurrent murine HRD tumors upregulated the immunosuppressive PGE2-EP2/4 pathway and targeting of COX-driven PGE2 production during chemotherapy significantly prolonged relapse and survival in preclinical mouse models, identifying a key vulnerability for the recurrence of human HRD OCs. In contrast, homologous recombination repair proficient (HRP) OC tumors evolved into T-cell excluded or desert phenotypes, characterized by malignant cells overexpressing Nduf4l2/Galectins and Trem2/ApoE overactive tumor-associated macrophages (TAMs). Our data suggest that therapeutically targeting of TREM2 overexpressing TAMs may improve anti-tumor immune responses and delay recurrence after first-line chemotherapy in HRP OC. By unraveling the cellular and molecular networks driving immune evasion in recurrent OC, our findings provide a roadmap for precision immunotherapy. Our work highlights novel immune biomarkers and therapeutic targets that can be exploited to retune anti-tumor T cell responses, paving the way for clinically actionable strategies in recurrent OC and other immune-resistant malignancies. Citation Format: Denarda Dangaj Laniti. Reprogramming T-cell networks to overcome immune evasion in ovarian cancer relapse abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 2570.