Abstract LB006: Cell marker-based classification of TNBC subtypes and patient outcomes

Abstract Background: Triple-negative breast cancer (TNBC) is an aggressive BC subtype characterized by significant molecular and clinical heterogeneity, influenced by genetic, phenotypic, and environmental factors, as well as interactions of the tumor with its surrounding milieu. In the recent past, gene expression-based subtyping methods have provided a functional understanding for the observed molecular diversity/heterogeneity. In this study, we propose an approach to reclassifying TNBC while considering the cellular heterogeneity of the tumor as well as the tumor milieu. Methods: RNA-Sequencing data for TNBC cohort of 250 diverse women [European (EA, self-reported race, SRR): 114 and African (AA, SRR): 136)], was obtained from our collaborators at LSUHSC. 38/250 were labeled ERhi via IHC but histologically confirmed to be TNBC patients. We utilized 261 breast specific cell type markers to classify our cohort via hierarchical (HC) and consensus (CC) clustering. Differentially expressed genes (DEGs) were identified at p-adj ≤ 0.05. Endotype specific networks were generated using protein-protein interactions, DEGs, and genes regulated by transcription factor. Survival was assessed via log-rank tests. Genetic admixture was performed on 227/250 patients using method described by Pakstis, A. J. et al. Results: We identified three distinct mechanistic subtypes (endotype, CC1-3) containing nine distinct groups (HC1-9), revealing substantial heterogeneity in cell-type and tissue-state within TNBC. Each subtype exhibited unique biomarkers, reflecting diverse immune, extracellular matrix, adipogenesis, and inflammatory signatures. Endotype specific networks revealed patients in CC1 exhibited upregulation of CD8 T cell activation, interferon and interleukin markers, while CC2 showed upregulation of extracellular matrix genes, and CC3, the subtype containing ERhi patients, showed upregulation of fatty acid regulation, hormone sensitivity, and pro-inflammatory genes. Survival analysis demonstrated favorable outcomes for CC1 and less favorable outcomes for CC2. Additionally, SRR-inferred survival differences highlighted poorer outcomes for CC2 groups. We found SRR-inferred prognostic biomarkers such as TTC6, ANKRD30A, and KRT86 associated with distinct prognostic profiles in CC2. Genetic admixture showed a comparable distribution of SRR and genetic ancestry across subtypes. We developed a user-friendly Shiny App for our subtyping method and validated it using TCGA TNBC data. Conclusion: TNBC subtyping using cellular heterogeneity of biopsied samples provides a mechanistic approach to patient classification. Subtype-specific mechanisms suggest tailored therapeutic strategies and provide survival outcomes specific to each subtype. Our method can be applied to other cohorts using the application developed in this work. Citation Format: Zahra Mesrizadeh, Kavitha Mukund, Fokrul Hossain, Jovanny Zabaleta, Denise Danos, Luis Del Valle, Xiao-Cheng Wu, Chindo Hicks, Yuan Chun Ding, Augusto Ochoa, Lucio Miele, Victoria L. Seewaldt, Shankar Subramaniam. Cell marker-based classification of TNBC subtypes and patient outcomes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr LB006.