Abstract C147: U54 Cancer Health Disparities Specialized Program of Research Excellence: The Lymphoma Outcomes SPORE
Abstract
Diffuse large B-cell lymphoma (DLBCL), the most common hematologic cancer in the US, is characterized by marked molecular and clinical heterogeneity. African American (AA) and Hispanic patients experience inferior outcomes due to biologic factors that remain poorly understood, in addition to well-described socioeconomic challenges. Further, these populations are underrepresented in research cohorts and clinical trials, and under-resourced community practices that disproportionately serve under-represented minority (URM) patients often lack in-depth hematopathology expertise and sufficient resources for rapid and precise diagnosis. Most disparities research has focused on improving access to care, overlooking opportunities for biologically informed precision interventions. Our group performed the first large-scale genomic analyses of DLBCL from patients with African ancestry (AFR) and found that, compared with European ancestry (EUR) DLBCL, AFR-DLBCLs are enriched for mutations in SETD2, MGA, and ATM – genes involved in DNA damage sensing. Using murine and human SETD2-mutant models, we demonstrated genomic instability linked to a senescence-associated secretory phenotype and a T-cell inflamed/exhausted microenvironment, identifying a biologically distinct DLBCL subtype enriched in AFR patients. Project 1 of our SPORE focuses on further defining this phenotype to identify targeted therapies that improve antitumor immunity and enhance existing treatments. Thus, we reason that DLBCLs from AFR and Hispanic patients are genetically and immunologically distinct from EUR-DLBCLs and are not fully captured by current molecular classification schemes. Project 2 addresses this knowledge gap by defining genetic alterations and the tumor microenvironment in URM cohorts; testing therapeutic vulnerabilities in ancestry-specific, patient-derived xenograft models; and deploying diagnostic and treatment strategies geared for community practices that serve URM patients. We also hypothesize that ancestry-associated disease subtypes can be identified from routine H&E slides using artificial intelligence (AI)-based deconvolution of complex tissue morphologies. Project 3 thus seeks to develop a diagnostic lymphoma classifier that provides accurate diagnosis from images of whole H&E slides that can be obtained in standard pathology laboratories, with dedicated bias mitigation strategies to ensure diagnostic accuracy in AFR and Hispanic populations. Spatial omics data will be integrated to develop H&E-based AI models that predict prognostic microenvironment features. Finally, we contend that these discoveries are clinically actionable through evidence-based community interventions, enabling precision therapy for AFR and Hispanic patients with lymphoma in clinical trials and routine care based on predictive biomarkers. Supported by Research Cores for Community Outreach & Engagement, Biospecimens & Pathology, and Biostatistics & Bioinformatics, our SPORE will establish a new paradigm for advanced lymphoma care in URM populations experiencing cancer disparities. Jean L. Koff, Jonathon B. Cohen, Lee A.D. Cooper, Michael R. Green, Giorgio Inghirami, David L. Jaye, Christopher E. Mason, Lorna H. McNeill, Yuri Fedoriw, Chijioke Nze, Ryan Sun, Francisco Vega, Roberta Zappasodi, Christopher R. Flowers. U54 Cancer Health Disparities Specialized Program of Research Excellence: The Lymphoma Outcomes SPORE [abstract]. In: Proceedings of the 19th AACR Conference on the Science of Cancer Health Disparities; 2026 Oct 18-21; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2026;35(10 Suppl):Abstract nr C147.