Abstract PS5-20: Protein interactome dysregulation analysis reveals putative therapeutic targets for BCI(H/I)-low breast cancers patients

Abstract

Abstract Background: Hormone receptor-positive (HR+) breast cancer (BC) is a disease of late recurrence. HR+ BC patients who are disease free after five years of adjuvant hormonal therapy, have an ongoing risk of disease recurrence and death for at least 15 years from diagnosis. Multiple trials have demonstrated consistent absolute benefits with extending adjuvant endocrine therapy to 10 years in HR+ patients. In several extended endocrine therapy trials, we have shown that HR+ patients with BCI(HoxB13/IL17BR) [BCI(H/I)]-high biomarker expression benefit from extended endocrine therapy, while those with low BCI(H/I)-low expression do not benefit from such therapy. Currently, there are limited therapeutic options for BCI(H/I)-low patients who are at high risk for late disease recurrence. Previous cell line-based, global scale protein-protein interaction (PPI) dysregulation mapping studies have identified therapeutic pathways within cellular circuits carrying the perturbed protein-protein interaction. To identify exploitable therapeutic vulnerabilities in BCI(H/I)-low BC patients, we performed comparative multiplexed quantitative mass spectrometry-based interactome dysregulation analysis of BCI(H/I)-low with BCI(H/I)-high BC samples and correlated interactome perturbations with drug responses. Methods: 21 BCI(H/I)-high and 23 BCI(H/I)-low frozen tumor samples from ER+ BC patients were identified in the MGH BC tissue repository, and proteomes were mapped using TMT-11 reagents, and the Synchronous Precursor Selection supported MS2/MS3 method on an Orbitrap Fusion and an Orbitrap Lumos. Dysregulations of PPIs were mapped through using bivariate outlier testing on each identified protein pair based on mapping Mahalanobis distances. . Protein identifiers (IDs) were converted to their appropriate gene IDs and gene set enrichment analysis (GSEA) was performed. Using previously established PPI data from 41 human breast cancer cell lines, we performed comparative PPI dysregulation GSEA and leading edge GSEA of the BCI(H/I)-high and -low samples. To identify therapeutic target candidates for BCI(H/I)-low patients, tumor interactome dysregulations were correlated with drug response dysregulations determined in drug screen of 195 cancer therapeutics against 41 breast cancer cell lines. Results: Across 44 samples, we have quantified more than 11,608 proteins with an average of 7,890 proteins per sample. Protein concentration co-regulation analysis resulted in identifying more than 11,908 protein-protein associations (ρ >0.65, FDRq <0.0001). PPI dysregulation analysis revealed an average of more than 1596 dysregulated PPIs per tumor sample. GSEA of the 21 BCI(H/I)-high proteome samples revealed a positive association with the Bowie tamoxifen response, and the α- and γ-interferon response C2 gene sets, while the 23 BCI-H/I-Low samples revealed a positive association (p=0.03, FDRq=0.052) with the hedgehog signaling and the early estrogen Hallmark gene sets. Protein-protein interaction GSEA identified 85 significant gene sets (p <0.016, FDRq <0.25) that were more frequently dysregulated in the BCI-H/I-low samples as compared with the BCI(H/I)-high samples. Dysregulated pathways in BCI-H/I-low samples included cell cycle regulation, anaphase promoting complex-mediated degradation of mitotic proteins, and ERK and RAS signaling. Correlation of the BCI(H/I)-low interactome perturbations with cell line model drug responses to a library of 195 therapeutics revealed a significant correlation with PI3K inhibitor drug response. Conclusion: Quantitative mass spectrometry-based protein-protein interaction dysregulation identified the PI3K- pathway as a putative exploitable therapeutic vulnerability in BCI(H/I)-low HR+ BC patients at risk for late disease recurrence. Citation Format: Dennis Sgroi, Johannes Kreuzer, Robert Morris, Xcanda Herrera Lopez, Wayland Chiu, Wilhelm Haas. Protein interactome dysregulation analysis reveals putative therapeutic targets for BCI(H/I)-low breast cancers patients [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS5-20.