结直肠癌T细胞图谱中与遗传祖先相关的特征

Introduction: Tumor-infiltrating lymphocytes (TILs) are positive prognostic indicators in colorectal cancer (CRC), yet factors driving their variability beyond microsatellite instability (MSI) are poorly defined. Prior studies suggest that genetic ancestry may influence systemic and tumor-associated immune responses. We tested the hypothesis that global genetic ancestry is independently associated with distinct T-cell features in the CRC tumor immune microenvironment. Methods: In 230 patients with CRC from the Latino Colorectal Cancer Consortium, we quantified tumor-associated T-cell repertoires using immunoSEQ TCRbeta assays for T-cell receptor (TCR) abundance and clonality (log-transformed), pathologist-scored TILs per high-powered field (dichotomized ≥2 vs <2; N=180), and whole-exome sequencing (WES)-inferred TCRalpha (TCRA) fractions (dichotomized ≥0.03 vs <0.03; N=192). Genetic ancestry proportions (European [EUR], African [AFR], East Asian [EAS], Indigenous American [NAT], South Asian [SAS]) were estimated from germline/normal genotypes from WES via supervised ADMIXTURE (using 1kGP/HGDP references). Correlations between TCR features were evaluated using Spearman coefficients. Logistic/linear regression examined the associations between immune features (TCR abundance, TCR clonality, TILs, and TCRA fractions) and individual ancestry proportions, adjusting for age, sex, tumor location, and MSI status. To account for the constrained sum of ancestral components, we applied an additive log-ratio transformation (ALR), expressing AFR, EAS, SAS, and NAT ancestries as log ratios relative to EUR. These ALRs were fitted into a multivariable logistic/linear regression model, and the overall contribution of genetic ancestry to each immune feature was tested using a 4-df likelihood-ratio test (4-df LRT). Results: Various T-cell quantification metrics were positively correlated, with pathologist-reviewed TILs showing significant correlations with TCR abundance, clonality, and TCRA fractions (ρ=0.29, 0.29, and 0.22, respectively). CRC patients with higher TCRA fractions were more likely to have higher EUR ancestry (Odds Ratio [OR]: 12.85; 95% Confidence Interval [CI]: 1.09-151.70, p=0.043) and lower NAT ancestry (OR: 0.06, 95% CI: 0.005-0.70, p=0.025). Higher TCR clonality showed a suggestive association with lower EUR proportions (p=0.076) and higher NAT proportions (p=0.084). We found no association with overall genetic ancestry and any T-cell feature in the compositional analysis (4-df LRT>0.05). Conclusion: Our findings suggest that specific genetic ancestry components may be associated with T-cell landscapes in CRC. These results indicate that genetic background may influence the host anti-tumor immune response in CRC, highlighting the importance of integrating genetic ancestry into personalized risk stratification and precision medicine approaches.