Abstract
INTRODUCTION: Disparities in incidence and outcome of rectal cancer are multifactorial in etiology but may be due, in part, to differences in gut microbiome composition. We used serial robust statistical approaches to assess baseline gut microbiome composition in a diverse cohort of patients with rectal cancer receiving definitive treatment. METHODS: Microbiome composition was compared by age at diagnosis (< 50 vs ≥ 50 years), race and ethnicity (White Hispanic vs non-Hispanic), and response to therapy. Alpha diversity was assessed using the Shannon, Chao1, and Simpson diversity measures. Beta diversity was explored using both Bray-Curtis dissimilarity and Aitchison distance with principal coordinate analysis. To minimize false-positive findings, we used two distinct methods for differential abundance testing: LinDA and MaAsLin2 (all statistics two-sided, Benjamini-Hochberg corrected false discovery rate < 0.05). RESULTS: Among 64 patients (47% White Hispanic) with median age 51 years, beta diversity metrics showed significant clustering by race and ethnicity (p < 0.001 by both metrics) and by onset (Aitchison p = 0.022, Bray-Curtis p = 0.035). White Hispanic patients had enrichment of bacterial family Prevotellaceae (LinDA fold change 5.32, MaAsLin2 fold change 5.11, combined adjusted p = 0.0007). No significant differences in microbiome composition were associated with neoadjuvant therapy response. CONCLUSION: We identified distinct gut microbiome signatures associated with race and ethnicity and age of onset in a diverse cohort of patients undergoing definitive treatment for rectal cancer.