Genetic landscape of colorectal cancer (CRC) across genetic ancestries: Implications for early cancer detection (ECD).

person Alexander Ioannidis Department of Biomedical Data Science, Stanford Medical School and Institute for Computational and Mathematical Engineering, Stanford University, Stanford, CA info_outline Alexander Ioannidis, Preethi Srinivasan, Sara L. Bristow, Shifra Krinshpun, Omid Shams Solari, Samuel Rivero-Hinojosa, Vasily N. Aushev, Adham A Jurdi, Minetta C. Liu, Bree Mitchell, Alexey Aleshin, Johannes G. Reiter, Yoshiaki Nakamura, Takayuki Yoshino, Jeffrey Wall, Parvathi Myer, Carlos D. Bustamante

Authors person Alexander Ioannidis Department of Biomedical Data Science, Stanford Medical School and Institute for Computational and Mathematical Engineering, Stanford University, Stanford, CA info_outline Alexander Ioannidis, Preethi Srinivasan, Sara L. Bristow, Shifra Krinshpun, Omid Shams Solari, Samuel Rivero-Hinojosa, Vasily N. Aushev, Adham A Jurdi, Minetta C. Liu, Bree Mitchell, Alexey Aleshin, Johannes G. Reiter, Yoshiaki Nakamura, Takayuki Yoshino, Jeffrey Wall, Parvathi Myer, Carlos D. Bustamante Organizations Department of Biomedical Data Science, Stanford Medical School and Institute for Computational and Mathematical Engineering, Stanford University, Stanford, CA, Natera, Inc., Austin, TX, Department of Gastroenterology and Gastrointestinal Oncology, National Cancer Center Hospital East, Kashiwa, Japan, National Cancer Center Hospital East, Kashiwa, Japan, Galatea Bio, Hialeah, FL, Albert Einstein Cancer Center, New York, NY Abstract Disclosures Research Funding No funding sources reported Background: Blood-based ECD has the potential to reduce mortality rates by improving strategies for adoption of and adherence to screening. As health disparities in the incidence of CRC across minority populations are known to exist, we evaluated the genomic landscape in patients with CRC across genetic ancestries. Methods: De-identified patients (N = 16,337) with stage I-IV CRC had personalized, tumor-informed commercial ctDNA testing (Signatera TM ) based on whole exome sequencing. Genetic ancestry determined via supervised local ancestry inference (pmid36042219) included African (AFR, n = 1697), Ashkenazi Jewish (ASJ, n = 1192), East Asian (EAS, n = 2247), European (EUR, n = 9726), Latino (LAT, n = 1291), and South Asian (SAS, n = 184) ancestries. Somatic driver mutations and pathogenic germline variants (PVs) in CRC-associated genes were evaluated. Somatic single nucleotide variants were used for mutational signature inference. Results: Across ancestries, 47-67% of the patients were male and most were aged > 50 years. Among 35 to 50-year-olds, microsatellite instability (MSI) rates were higher in LAT (12.9%, p = 0.04) and SAS (28.3%, p < 0.001) vs EUR (8.2%). Among > 50-year-olds, MSI rates were lower in AFR (10.7%, p < 0.001), EAS (7.9%, p < 0.001), and LAT (10.4%, p < 0.001) vs EUR (16.0%). The POLE signature was more common in AFR (2%, p = 0.001) vs EUR (1.1%). There were significant differences in the frequency of driver mutations and PV rates between non-EUR vs EUR ancestry groups (Table, *p < 0.05, **p < 0.001). Gene-level enrichment was observed in MLH1 for SAS (8.7%, p < 0.001) and LAT (1.8%, p = 0.03), CHEK2 for AFR (0.4%, p = 0.03) and EAS (0.1%, p < 0.001), and PMS2 (1.4%, p = 0.006) for AFR. Among patients aged ≤50 years, similar findings were observed, with differences driven by MSI rates and germline findings in Lynch syndrome (LS) genes. Conclusions: This is the first study to investigate genomic differences between multiple ancestry groups in the largest cohort to date, including EAS, LAT, and SAS populations. The differences observed between ancestries could provide insight into differences in both hereditary risk and subsequent tumor initiation and growth. Findings from this study may provide information for developing risk stratification and prevention strategies for the early detection of cancer and provide the rationale for precision treatment based on ancestry. Rates of Somatic Driver Genes (MSI %, MSS/nonhypermutated %) Rates of Germline PVs by Penetrance APC BRAF KRAS PIK3CA TP53 LS High Moderate Low AFR 60%* 82%** 31%** 4%** 39%** 54%** 38% 19% 29% 66%* 4.7%* 5.8%* 0.4%* 0.9%** ASJ 48% 75% 41%* 5% 32%* 43% 42% 18% 27% 69% 3.3% 4.7% 0.9% 3.3% EAS 54% 84%** 41%* 4%** 24% 43% 35% 14%* 36% 79%** 3.7% 4.9% 0.1%** 0.2%** EUR 47% 76% 51% 7% 22% 42% 36% 17% 31% 69% 3.0% 4.3% 0.1% 2.5% LAT 48% 76% 27%** 4%* 36%* 44% 35% 17% 34% 69% 3.6% 5.3% 0.1% 1.9% SAS 56% 73% 3%** 2% 44%** 43% 44% 13% 18% 70% 12.0%** 14.1%** 0.5% 2.2%