2975 - Microbiome Diversity as a Predictor of Radiation Response and Toxicity in Rectal Cancer

N. J. Eustace¹, K. Lee², M. Fakih³, A. Kaiser⁴, A. Liu¹, and Y. J. Chen¹; ¹Department of Radiation Oncology, City of Hope National Medical Center, Duarte, CA, ²TGen, Phoenix, AZ, ³Department of Medical Oncology, City of Hope National Medical Center, Duarte, CA, ⁴Department of Surgery, City of Hope National Medical Center, Duarte, CA

Purpose/Objective(s): Total neoadjuvant therapy (TNT) followed by surgical resection is recommended for T3/4 or N (+) rectal cancer. The gut microbiome (GM) can affect cancer therapies by direct interaction with treatment or by indirectly stimulating host response through immunomodulation. This prospective, longitudinal study aimed to evaluate the association between diversity and abundance of GM and chemoradiation (CRT) sensitivity and toxicity. Materials/

Methods: Our study examined 80 stool samples collected from 18 patients with T3/T4 or N(+) rectal adenocarcinoma requiring TNT between 2021 and 2023. Samples were collected before induction chemotherapy (if applicable), before CRT, the 3^rd and the last week of CRT, and 4-6 weeks post CRT. Radiation enteropathy was graded by CTCAE v5.0. Grades 0-1 are categorized as mild toxicity and Grades 2-5 are severe toxicity. Responders are cases with pathologic complete response (CR), tumor regression grade 1, or sustained clinical CR for 1 year. 16S rRNA gene sequencing was employed to characterize fecal microbial communities, enabling the evaluation of diversity metrics and the identification of discriminatory bacterial taxa across experimental cohorts.
Results: There were 13 cases (72.2%) with mild toxicity and 5 cases with severe toxicity (28.8%). There were 9 responders (50.0%) and 9 non-responders (50.0%). Examining individual microbiome samples collected longitudinally over time, Beta diversity analyses indicated no significant alterations in GM community composition throughout the course of TNT. Alpha diversity analysis (Shannon diversity) on baseline samples (before any treatment) showed statistically significant higher GM richness and evenness for responders compared to non-responders. Shannon diversity analysis also showed higher alpha diversity in cases with mild toxicity compared to severe toxicity although not statistically significant. By ANCOM-BC analysis, significant enrichment or depletion of bacterial genera between mild vs. severe toxicity and responders vs. non-responders were identified. Specifically, Lachnospira (related to short-chain fatty acid production) is depleted in the samples collected immediately before CRT of non-responders.
Conclusion: Our findings suggest that GM richness and evenness before TNT are associated with response to TNT and likely associated with milder toxicity. Phylogenetic differences in baseline GM may impact treatment response.