The impact of patient biology on racial disparities in breast cancer outcome
Hormone receptor positive (HR+) breast cancer is the most common subtype of breast cancer diagnosed globally. Despite effective targeted therapies, HR+ breast cancer remains a leading cause of cancer-related death in women. Long-standing epidemiological research identifies significantly worse outcomes for Black women diagnosed with HR+ breast cancer relative to White women. While structural factors such as access to healthcare and education level contribute to this outcome disparity, it persists even in analyses where these factors are controlled. In-depth analyses of the somatic molecular biology that may underlie these outcome disparities are hampered by a lack of datasets that represent Black patient populations. Here, we generate a HR+ breast cancer patient transcriptomic dataset that overrepresents Black women and controls for access to healthcare and education level. We find that signatures relating to the tumor microenvironment, i.e. collagen deposition and prognostically unfavorable T-cell landscapes are enriched in HR+ tumors from Black women. Importantly, we find, using experimental model systems in vitro and in vivo, that race-aligned collagen deposition patterns are at least partially attributable to tumor cell-intrinsic signaling and critical for Black breast cancer metastasis. We also find that unfavorable T-cell signatures in HR+ tumors from Black women, which have previously been attributed to race and ancestry, are more strongly poverty-aligned. Using multiple independent datasets, we identify STAT4 as a potential master regulator of this poverty-associated tumor immune signature. Together, these findings provide new evidence that somatic molecular biology of breast cancer patients can be modified by multiple structural factors such as self-identified race and poverty burden to promote poor patient outcomes. Integrating an understanding of structural factors into molecular cancer research is critical for implementing truly personalized, and maximally effective, oncology systems.