Abstract 6929: Inter- and intratumoral PIK3CA subclonal diversity in breast cancer contextualized by single-cell multiomics
Jon Zawistowski,Isai Salas-Gonzalez,Tia Tate,Tatiana Morozova,Katherine Kennedy,Durga Arvapalli,Jamie Remington,Jeffrey Marks,E. Shelley Hwang,Gary Harton,Victor Weigman,Jay A. West
DOI: https://doi.org/10.1158/1538-7445.am2024-6929
IF: 11.2
2024-03-22
Cancer Research
Abstract:Abstract Rare clonotypes within pre-cancerous tissues can drive progression to cancer. However, the evolution of rare clonotypes in tumors or normal tissue cannot be defined in the absence of single-cell resolution. At this single-cell level, multiomic interrogation across the Central Dogma of Biology provides enhanced power to reconstruct such evolutionary trajectories, defining the mutational profile, cell identify, and receptor expression within each subpopulation. Leveraging a multiomic approach, we aimed to define how different mutations in the same oncogenic driver observed in the same tumor resection associate with copy number variation (CNV) across the genome. We analyzed individual ductal carcinoma in situ/invasive ductal carcinoma cells using a unified whole-genome and full-transcript RNAseq workflow (ResolveOMETM, BioSkryb Genomics) coupled with panel-level extracellular protein information through oligo-conjugated antibodies (BioLegend). We sequenced the exomes and transcriptomes of ResolveOME-amplified single cells from mastectomy samples from twelve patients. At the single nucleotide variant (SNV) level, we identified an allelic series of PIK3CA oncogenic driver mutations in the same tumor resection. A single amino acid, in-frame deletion of E109 dominated the sample, followed by H1047R and K111E in decreasing subclonal abundance. A fourth mutation, E345T, not present in the first sample, was detected as the sole PIK3CA variant in the second tumor sample. Intriguingly, each respective PIK3CA mutation class was associated with a distinct copy number alteration profile revealed by low-coverage whole-genome sequencing: Cells harboring the predominant ΔE109 mutation displayed chromosome 8p,16p, and 17 loss while the less abundant H1047R mutation was in single cells harboring 1q gain, 4q loss, and 22 loss in addition to the 8p and 16q loss present in the ΔE109 cells. PIK3CA K111E had a quiescent, 2n copy number profile. The transcriptomic arm of ResolveOME, containing an oligo-conjugated antibody readout of surface protein expression, jointly confirmed the epithelial identity for the cells harboring the oncogenic PIK3CA mutations. A subpopulation of cells harboring prototypical breast cancer CNV were typed as non-epithelial with increased stemness characteristics, indicative of the ability to resolve phenotypic cellular states. These results suggest a tight interrelationship between CNV and SNV influencing the relative rate of clonal expansion. They also provide the opportunity to explore CNV:SNV signature association with loci exclusive of PIK3CA, and to exploit the power of multiomic integration for lineage reconstruction and for defining common oncogenic signatures of the evolving tumors. Citation Format: Jon Zawistowski, Isai Salas-Gonzalez, Tia Tate, Tatiana Morozova, Katherine Kennedy, Durga Arvapalli, Jamie Remington, Jeffrey Marks, E. Shelley Hwang, Gary Harton, Victor Weigman, Jay A. West. Inter- and intratumoral PIK3CA subclonal diversity in breast cancer contextualized by single-cell multiomics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6929.
oncology