Abstract P2-03-05: HOXB13/IL17RB-low breast cancers are predicted to respond to PIK3CA inhibitors independent of PIK3CA mutational status

Abstract Background: The Breast Cancer Index is a gene expression-based signature comprising two functional biomarker panels, the Molecular Grade Index (MGI) and the two-gene ratio, HOXB13/IL17BR (H/I). MGI measures tumor proliferation, while H/I measures estrogen signaling. Integration of MGI and H/I provides a single score that quantifies the risk distant recurrence, while H/I alone predicts benefit from extended endocrine therapy (EET). In multiple EET trials, we have shown that hormone receptor-positive (HR+) breast cancer (BC) patients with H/I-high biomarker expression benefit from EET, while those with H/I-low expression do not benefit from such therapy. Previous protein-protein interaction (PPI) dysregulation (dysreg) analysis of H/I-low BC tumors with drug-response-associated dysreg in BC cell lines revealed a significant correlation with response to two PIK3CA inhibitors, suggesting that PIK3CA pathway may represent an exploitable therapeutic vulnerability for H/I-low BC patients. However, given that PIK3CA is frequently mutated in ER+ tumors, the PPI dysreg analysis may merely reflect enrichment of PIK3CA genomic alterations within the H/I-Low BCs. Thus, to determine if a genetic abnormality within the H/I-low BCs accounts for the correlation with PIK3CA inhibitor drug response in BC cell lines, we performed targeted DNA sequencing of H/I-high and H/I-low BC tumors. Methods: DNA was extracted from 44 H/I-high and 30-H/I-low BC samples using a Qiaqen DNeasy Blood & Tissue Kit. Samples were then prepared using the Archer VariantPlex kits according to a modified protocol and sequenced in batches on the Illumina NextSeq platform. To determine if a genetic abnormality within H/I-low BCs accounts for the PPI dysreg correlation with PIK3CA drug response, the magnitude of the sample’s drug associated PPI dysregs was compared to its mutational status. Geneset enrichment (GSEA) functional profiles between BC tissue PPI dysregs and BC cell line drug response-associated PPI dysregs, identified two PIK3CA inhibitors as potential therapeutic drugs for H/I-low tumors. For each PIK3CA drug, the set of leading-edge genes associated with the genesets important to the observed correlation were isolated. Z-score transformed PPI dysreg count data across the BC samples was used to calculate the average dysreg scores for each sample based on the drug-associated leading-edge genes. The samples were sorted based on scores associated with each drug and GSEA was run on each ranked list looking for an association between the dysreg magnitude and PIK3CA mutation status. Results: Targeted DNA sequencing identified copy number variants for in H/I-high and low tumors, the most common of which are: ERBB2 (3/44 H/I-high; 1/30 H/I-low; p: 0.642), FGF19 (3/44 H/I-high; 1/30; p: 0.642), CCND1 (3/44 H/I-high; 1/30 H/I-low; p: 0.642). Sequencing identified a similar distribution of single nucleotide variants (SNVs) between the H/I-high and H/I-low groups. The most common SNVs identified in order of prevalence are PIK3CA (18/44, H/I-high; 6/30, H/I-low; p: 0.078), TP53 (12/44 H/I-high; 6/30 H/I-low; p: 0.585), CDH1 (4/44 H/I-high; 5/30 H/I-low; p: 0.471), BRCA2 (4/44 H/I-high; 2/30 H/I-low; p: 1). No significant difference in mutational prevalence was identified between the H/I-high and H/I-low BCs. No-significant associations between the PPI dysreg magnitude of the drug-associated leading-edge genes and PIK3CA mutational status was observed (AZD6482 FDR=0.736 and A66 FDR=0.95). Conclusion: No significant genetic alteration, including PIK3CA mutational status, was identified between H/I-high and H/I-low groups. Thus, protein-protein interaction (PPI) dysregulation analysis identifies H/I-low BC tumors as those that are predicted to response to PIK3CA inhibitors independent of PIK3CA mutational status. Citation Format: Baris Boyraz, Grace Kirkpatrick, Stefan T. Kaluziak, Robert Morris, Johannes Kreuzer, Wilhelm Haas, Anthony John Iafrate, Dennis Sgroi. HOXB13/IL17RB-low breast cancers are predicted to respond to PIK3CA inhibitors independent of PIK3CA mutational status [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P2-03-05.