Genetic Ancestry Tied to Differences in CRC Burden, Age at Diagnosis

European genetic ancestry was associated with 50% higher crude odds of colorectal cancer (CRC) compared with all other ancestries combined, but American admixed-Latino and East Asian ancestry carried higher age-specific CRC risk once the competing risk of death was taken into account, according to a cohort study published in JAMA Network Open.¹

The findings, drawn from more than 316,000 participants in the National Institutes of Health's All of Us Research Program, suggest that genetic ancestry captures disease patterns that self-reported race and ethnicity alone may miss, a distinction with implications for how health systems approach CRC screening and risk stratification.

Understanding the Role of Genetic Ancestry in CRC Disparities

CRC remains a leading cause of cancer mortality worldwide, with roughly 2 million new diagnoses annually. Genetics, biology, lifestyle, and social determinants of health impact CRC incidence and outcomes, with community behaviors and environmental factors further contributing to disparities in risk and prognosis.² Therefore, investigators emphasized that understanding how these factors interact may help explain the disproportionate CRC burden observed in certain subgroups.¹

Much of the existing literature on CRC disparities relies on self-reported race and ethnicity, which the study authors noted are social constructs that capture lived experience but imperfectly reflect genetic similarity and often exclude admixed individuals from disparities research altogether.

With genetic ancestry, computed from short-read whole-genome sequencing and expressed through principal components, the researchers could more precisely capture shared genetic background. Therefore, the investigators analyzed All of Us Controlled Tier data on 316,624 eligible participants collected from July 1986 through October 2023.

Genetic Ancestry Reveals Differences in CRC Burden, Age-Specific Risk

Of the study population, 2914 (0.9%) developed CRC. Also, 54.4% (n = 172,327) were of European genetic ancestry, and 45.6% (n = 144,297) were of non-European genetic ancestry. Specifically, 19.6% (n = 61,922) were African, 16.7% (n = 52,856) were American admixed-Latino, 3.1% (n = 9932) were East Asian, South Asian, and Middle Eastern, and 6.2% (n = 19,587) identified as other. Participants with European genetic ancestry were older (mean [SD] age, 58.4 [17.1] vs 49.7 [16.3] years) and more frequently had a cancer history (19.3% [n = 33,341] vs 8.7% [n = 12,510]) than those with non-European genetic ancestry.

The study's central finding was a divergence between overall CRC burden and age-specific risk. In crude comparisons, European ancestry carried 50% higher odds of any CRC (OR, 1.50; 95% CI, 1.39-1.62), including both colon and rectal cancer, whereas African (OR, 0.77; 95% CI, 0.70-0.85) and American admixed-Latino (OR, 0.63; 95% CI, 0.56-0.71) ancestries were associated with lower odds relative to all other groups combined.

By contrast, individuals of European ancestry were diagnosed at a median age of 63.4 years, which was older than the median age at diagnosis of American admixed-Latino (8.4 years earlier), East Asian (7.9 years earlier), African (5.0 years earlier), and other (4.7 years earlier) ancestry groups. When the researchers modeled hazard on an attained-age timescale, accounting for the competing risk of death, American admixed-Latino (HR, 1.30; 95% CI, 1.14-1.47) and East Asian (HR, 1.43; 95% CI, 1.06-1.94) ancestry groups showed significantly higher age-specific CRC hazard than European ancestry, a pattern that held after adjustment for clinical and demographic covariates.

The authors suggested that European populations may accumulate more cumulative cases later in life because of a larger older at-risk population, whereas some non-European groups, particularly American admixed-Latino individuals, may face earlier onset or higher relative risk at younger attained ages. Notably, the ancestry-based pattern differed from previously reported CRC incidence patterns based on self-reported race and ethnicity, further highlighting that genetic ancestry and social experience are related but distinct constructs.

Established Risk Factors Remain Key Predictors of CRC

Despite the ancestry signals, established clinical, behavioral, and family-history variables were among the strongest contributors to the study's multivariable and prediction models. In adjusted logistic regression, prior cancer history (adjusted OR [aOR], 1.61; 95% CI, 1.48-1.77), type 2 diabetes (aOR, 1.95; 95% CI, 1.79-2.13), family history of CRC (aOR, 2.82; 95% CI, 2.42-3.29), and family history of polyps (aOR, 1.52; 95% CI, 1.34-1.73) were among the strongest predictors.

Socioeconomic and access-related variables also diverged by ancestry group. Annual household income was inversely associated with CRC among European participants only (aOR, 0.97; 95% CI, 0.94-1.00). The county-level Social Vulnerability Index, however, was positively associated with CRC among non-European participants only (aOR, 1.33; 95% CI, 1.04-1.70), a signal that social determinants of health may operate differently across ancestry groups even after adjustment.

For risk prediction, a penalized least absolute shrinkage and selection operator model showed only modest discrimination (test area under the receiver operating characteristic curve [ROC AUC], 0.733). XGBoost, an extreme gradient boosting model that used raw, missingness-aware data, performed substantially better, with a test ROC AUC of 0.898 (95% CI, 0.882-0.912) for the full cohort and consistent performance across European (ROC AUC, 0.900), non-European (ROC AUC, 0.869), and race/ethnicity subgroups.

Smoking pack-years, family history of CRC, age, prior cancer history, and family history of polyps ranked as the top contributors to the model's predictions, with genomic principal components and self-reported race and ethnicity contributing more modestly. Rather than replacing established screening approaches, the investigators suggested that the model could potentially complement CRC risk stratification and screening by serving as a risk-enrichment tool.

Study Limitations and Implications for CRC Risk Stratification

The researchers acknowledged their study’s limitations, including that the All of Us dataset was not population-based and was restricted to participants with whole-genome sequencing and linked electronic medical record data, potentially introducing selection bias and limiting the generalizability of the findings. In addition, locus-specific ancestry, molecular tumor characteristics, and tumor stage, which could provide more granular patient insight, were unavailable. Still, they expressed confidence in their findings.

“These findings suggest that genetic ancestry is associated with meaningful differences in CRC burden and age-specific risk, and a multiethnic extreme gradient boosting model may complement CRC risk stratification and screening,” they wrote.

References

1. Chatzipanagiotou OP, Charalampous CM, Cordle A, et al. Genetic ancestry and colorectal cancer in the All of Us dataset. JAMA Netw Open. 2026;9(8):e2628789. doi:10.1001/jamanetworkopen.2026.28789
2. Eng C, Holowatyj AN. Colorectal cancer genomics by genetic ancestry. Cancer Discov. 2022;12(5):1187-1188. doi:10.1158/2159-8290.CD-22-0217