When Dr. Yixuan He and her Ph.D. student Jiawei Tu began analyzing genetic data from over 30,000 cancer patients across two leading U.S. cancer centers, they weren’t just looking for mutations—they were searching for patterns shaped by history, migration, and ancestry. At the European Society of Human Genetics conference, He presented groundbreaking findings showing that a patient’s genetic ancestry significantly influences cancer progression and survival, offering a powerful new lens for personalized medicine. The study examined nearly 1,900 tumor mutations across five cancers—breast, colorectal, glioma, pancreatic, and lung—drawing from patient data at Dana Farber Cancer Institute in Boston and MD Anderson Cancer Center in Houston. Crucially, the team accounted for socioeconomic status and air pollution exposure to isolate the genetic effects, ensuring the results weren’t skewed by environmental disparities.
While tumor sequencing is now routine in oncology, this study is the first to integrate ancestry data at such scale and rigor, revealing dozens of mutations that vary significantly across ancestral groups—about half of which are already targetable with existing therapies. The researchers developed a survival prediction score based on tumor genetics, which improved notably when ancestry information was added. In pancreatic cancer, the enhancement in predictive accuracy was especially pronounced, suggesting that ancestry isn’t just a background detail—it’s a clinically relevant factor. And the best part? No new tests are needed. Since genetic ancestry can be inferred from existing tumor sequencing data, and environmental factors like pollution exposure can be estimated from zip codes, the integration is both feasible and cost-effective.
The team uncovered new biological insights as well: for example, the CDK6 gene, which regulates cell proliferation, was found to be enriched in African American breast cancer patients, while American colorectal cancer patients with admixed ancestry showed frequent loss of SMAD2, another key cell cycle regulator. These findings could guide more precise, ancestry-informed treatment strategies. Even more compelling, the same ancestry-related genetic signals appeared consistently across both biobanks, despite their geographic and demographic differences—validating the robustness of the results. "By identifying specific genetic markers linked to ancestry, we can pinpoint targetable mutations to help doctors use treatments with better survival outcomes," He said. The next step is expanding the analysis to more cancer types and incorporating lifestyle factors like smoking, with plans to collaborate on validating the findings in diverse global populations. This work doesn’t just improve prediction—it moves medicine closer to equity, ensuring breakthroughs benefit everyone, not just a narrow segment of the population.