Vanderbilt's massive breast cancer atlas could reshape how tumors are treated

Researchers at Vanderbilt-Ingram Cancer Center have created a sweeping transcriptomic atlas of breast cancer — one of the largest ever assembled — by integrating RNA expression data from more than 7,600 tumors across 22 independent studies. Published in Cell Reports Medicine, the atlas groups tumors into interconnected molecular clusters, visualized like an archipelago, where each "island" represents biologically related cancers with shared treatment responses and clinical outcomes.

The findings challenge existing classification systems. Estrogen receptor-low tumors, for instance, behave more like ER-negative disease and may need different treatment approaches than currently used. The atlas also identifies HER2-positive subsets most dependent on HER2 signaling, a hidden neuroendocrine breast cancer subtype with worse outcomes, and an aggressive basal-luminal progenitor subtype of triple-negative breast cancer with potential vulnerabilities to targeted therapies.

Key Takeaways:

• Current breast cancer classifications miss clinically meaningful tumor differences
• ER-low tumors may be better treated as ER-negative disease
• FGFR1 emerges as a potential target for a newly identified neuroendocrine subtype
• A new subtyping tool works with standard bulk RNA sequencing for practical clinical use

Why it matters: This atlas offers oncologists a molecular roadmap to more precisely match patients to therapies — potentially improving outcomes for patients whose tumors are currently misclassified or underserved by existing treatment frameworks.