Resolving Intra-Tumoral Heterogeneity of Breast Cancer Using Single-Cell Multi-Omics

Yun Yan, MS (Advisor: Nicholas Navin, PhD)

Breast cancer is a highly heterogeneous disease characterized by diverse genetic, transcriptional, and epigenetic landscapes. Understanding how intra-tumoral heterogeneity (ITH) influences therapeutic response and drives tumor evolution remains a central challenge in cancer biology. Here, we leverage single-cell multi-omics technologies to dissect the cell states and gene expression programs underlying breast cancer heterogeneity, with a focus on triple-negative breast cancer (TNBC) and estrogen receptor-positive (ER?) breast cancer.

In the first part, we applied single-cell RNA sequencing and spatial transcriptomics to profile over 400,000 cells from 101 treatment-naïve TNBC patients. We identified 4 archetypes at the patient level and 13 metaprograms exhibiting distinct functional profiles along with 49 tumor microenvironment (TME) cell states at the single-cell level. Notably, we uncovered the immune features—including interferon signaling and HLA expression from cancer cells—that were associated with response to neoadjuvant chemotherapy. These findings provide a comprehensive view of the gene expression landscape of TNBC at a single-cell resolution and its relationship to therapeutic outcomes.

In the second part, we developed a single-cell multi-omics method, named wellDA-seq. This is a high-resolution, high-throughput method that simultaneously measures genome-wide copy number alterations and chromatin accessibility in single cells. Applying this technique to over 22,000 cells from ER-positive breast tumors, we revealed both genetically hardwired and epigenetically plastic cancer cell states. We found that most chromatin changes were confined to subclonal genomic regions. We identified that the origin-of-cell of the ER-positive cancer cells is luminal hormone-responsive cells. We also revealed several plastic phenotypes, including epithelial-mesenchymal- transition and interferon gamma response. The level of the global concordance between the genotypes and epigenetic profiles in each tumor is likely to be associated with the stage of tumor development. These results improve the understandings of how genomic alterations interface with epigenomic states to shape tumor evolution.

Together, this work advances our understanding of breast cancer heterogeneity by studying genomic, transcriptomic, epigenomic, and physical spatial information at single-cell resolution. This dissertation provides showcases of the potential of using single-cell multi-omics to uncover the cellular and molecular mechanisms of therapeutic response and tumor progression.

Advisory Committee:

• Nicholas Navin, PhD. Chair
• Jichao Chen, PhD
• Ken Chen, PhD
• Kunal Rai, PhD
• Linghua Wang, PhD