Abstract 18: Molecular dissection of chemotherapy response in triple negative breast cancer (TNBC) using microscaled proteogenomics

Introduction: There are no robust molecular predictors for response of TNBC to chemotherapy. Microscaled proteogenomics (PMC6985126) was therefore applied to biopsies from TNBC patients undergoing neoadjuvant carboplatin/docetaxel. Approach: Sufficient tumor-rich tumor core biopsies were accrued from 59 patients. A second biopsy was collected from 16 patients on day 3. Study endpoints were pathological complete response (pCR) and residual cancer burden (RCB). Analyses included exome-based mutational signatures, RNA-based TNBC subtyping, immune cell infiltrate scores, and multi-gene proliferation scores, protein and phosphoprotein-based stimulatory and inhibitory immunomodulators. Single-sample Gene Set Enrichment Analysis (ssGSEA) was used for gene set and pathway scoring. Non-parametric tests and outlier analyses were applied to identify differential genes, proteins and pathways. Results: Most cases were PAM50 basal-like and all Lehmann TNBC subtypes were represented. BRCA1/2 and PALB2 homologous recombination (HRD) mutations were observed in 7 cases. Neither subtype, HRD mutation nor HRD mutation signature associated with pCR, but higher mismatch repair defect signature associated with higher RCB. Comparison of baseline pCR vs non-pCR cases showed higher ssGSEA scores for metabolic pathways including oxidative phosphorylation, fatty-acid metabolism, and glycolysis in the non-pCR cases at the protein but not at mRNA level. Non-pCR cases were associated with chromosomal deletions in chemokine receptors, JAK2, and PD-L1, lower PD-L1 protein levels, and lower immune activation. Consistently, ssGSEA scores for interferon alpha and gamma response pathways were higher in pCR cases. Phospho-PD-L1 levels were anti-correlated with developmental pathways. Matched comparisons of baseline and on-treatment samples revealed increase in proteins involved in cell cycle, DNA replication, and mismatch repair following treatment. Metabolic proteins were also upregulated following treatment, while complement activation, immune, and cell adhesion-related proteins were downregulated. Conclusion: Microscaled proteogenomic analysis revealed a wealth of biological features associated with chemotherapy resistance beyond immune response markers, including metabolic features that are only present at the protein level. These data suggests that the development of a microscaled proteogenomic chemotherapy response predictor is a feasible objective for future studies. Citation Format: Meenakshi Anurag, Eric Jaehnig, Shankha Satpathy, Karsten krug, Jonathan T. Lei, Yongchao Dou, Beom-Jun Kim, Cathy M. Sullivan, D. R. Mani, Erik J. Bergstrom, Gloria V. Echeverria, Ian S. Hagemann, Kristen Otte, Henry Rodriguez, Ana I. Robles, Michael T. Lewis, Michael Gillette, Bing Zhang, Mothaffar F. Rimawi, Steven Carr, Foluso O. Ademuyiwa, Matthew J. Ellis. Molecular dissection of chemotherapy response in triple negative breast cancer (TNBC) using microscaled proteogenomics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 18.