Abstract PS2-10-06: Proteogenomic profiling of treatment-naïve breast cancer biopsies from patients receiving neoadjuvant chemotherapy

Abstract Background: Breast cancer is a heterogeneous disease, comprising approximately 70% hormone receptor-positive (HR+), 15% HER2-positive (HER2+), and 15% triple-negative breast cancer (TNBC) cases. As standard of care, HR+ tumors are treated with hormone therapy, often after surgery. However, chemotherapy may also be used before surgery (neoadjuvant setting). HER2+ breast cancer responds well (50-70% pathological Complete Response, pCR) to targeted anti-HER2 therapies. In contrast, TNBC is primarily treated with neoadjuvant chemotherapy with only 30-50% achieving pCR. Identifying biomarkers predictive of pCR or that indicate alternative therapeutic strategies can help tailor therapy. Recent studies utilizing proteogenomics data (DNA, RNA, protein, and phosphorylation site profiles from the same tumor) from clinical trials highlighted candidates including LIG1 loss in TNBC and GPRC5A/TPBG overexpression in HER2+ tumors as potential predictive biomarkers and therapeutic targets. This study applies proteogenomic profiling to pre-treatment biopsies from tumors treated with physician-selected neoadjuvant chemotherapy to further explore putative targets for precision oncology. Methods: Microscaled proteogenomic profiling was conducted on 51 treatment-naive primary breast tumor core needle biopsies. Subtyping was determined via IHC: HER2+ (IHC 3+), HER2 Equivocal (IHC 2+), HR+ (ER or PR positive), or TNBC (negative for all three receptors). All received physician-selected neoadjuvant regimens, and pCR was evaluated after surgical resection. Multi-omics profiling of tumor sample from each biopsy included whole-exome sequencing (WES), RNA-seq, and mass spectrometry-based proteomics and phosphoproteomics. Both data-dependent acquisition (DDA) and more sensitive data-independent acquisition (DIA) proteomics were used. Results: Proteomic profiles were obtained for 16 HER2+, 17 HR+, and 18 TNBC patients. WES data were acquired for 50 patients; RNA-seq for 26. DIA quantified more total proteins/sites than DDA, though measurements for top variable genes were highly correlated. To assess the potential for each proteomics data type to inform precision oncology, both were included in all subsequent analyses. The highest pCR rate (69%) was achieved in HER2+ tumors treated with dual antibody therapy where all 4 non-pCR tumors were HR+, a known resistance marker, and 3 were HER2 equivocal with lower HER2 protein expression. Both the DIA and DDA data for GPRC5A protein also confirmed significantly higher expression in non-pCR vs. pCR HER2+ tumors. HR+, HER2- tumors had the lowest pCR (12.5%), but two pCR cases co-treated with pembrolizumab showed high expression of cell cycle and immune proteins. TNBC showed 35% (6/17) pCR overall, improving to 56% (5/9) when treatment included pembrolizumab. Non-pCR tumors had lower cell cycle/immune protein levels and higher epithelial-mesenchymal transition protein levels versus pCR tumors. Detailed proteogenomic profiles for FDA-approved drug targets, TNBC-specific kinases, and cell membrane proteins enriched in the context of non-pCR TNBC cases will be presented. Conclusion: This study demonstrates the feasibility of proteogenomic profiling from clinical breast tumor biopsies and highlights mechanisms of anti-HER2 resistance, potential factors underlying exceptional chemotherapy response in ER+ cancers, and candidate alternative therapeutic targets for non-pCR TNBC, offering a template for future precision oncology efforts. Citation Format: E. Jaehnig, M. Holt, J. Zecha, M. S. Glover, W. Kelly, B. Nuttall, B. McCue, I. Marin, S. Carter, E. Bonefas, L. Healy, G. Miles, M. Anurag, S. Hilsenbeck, B. Zhang, S. Hess, A. Thompson. Proteogenomic profiling of treatment-naïve breast cancer biopsies from patients receiving neoadjuvant chemotherapy abstract. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS2-10-06.