Main Session
Sep 30
PQA 08 - Gastrointestinal Cancer, Nonmalignant Disease, Palliative Care

3485 - Gut Microbiome and Metabolic Traits in Immunotherapy-Based Total Neoadjuvant Therapy for MSS Locally Advanced Rectal Cancer: A Longitudinal Cohort Study

02:30pm - 03:45pm PT
Hall F
Screen: 3
POSTER

Presenter(s)

Weiqing Lu, PhD, MS - Fudan University Shanghai Cancer Center, Shanghai, Shanghai

W. Lu1, X. Wu1, Y. Wang2, H. Zhang3, W. Yang1, M. Zhou1, Y. Chen1, R. Wu3, Y. Wang3, J. Wan1, F. Xia1, Z. Zhang1, and L. Shen1; 1Department of Radiation Oncology, Fudan University Shanghai Cancer Center, Shanghai, China, 2Department of Radiation Oncology, Fudan University Shanghai Cance Center, Shanghai, China, 3Department of Oncology, Shanghai Medical College, Fudan University, Shanghai, China

Purpose/Objective(s): While immunotherapy-based total neoadjuvant therapy (iTNT) may enhance treatment efficacy for microsatellite stable (MSS) locally advanced rectal cancer (LARC), its added toxicities remain poorly predictable. Current clinical tools lack precision in stratifying therapeutic responses and adverse events. Emerging evidence suggests gut microbiome dynamics interact with radiotherapy, chemotherapy, and immunotherapy, potentially serving as predictive biomarkers. This study employs longitudinal stool metagenomics and untargeted metabolomics to characterize gut microbial-metabolic profiles in LARC patients undergoing iTNT.

Materials/Methods: This prospective cohort included 48 MSS LARC patients (2021-2023) receiving short-course radiotherapy (25Gy/5Fx) combined with 6 cycles of capecitabine/oxaliplatin and anti-PD-1 antibody. Clinical outcomes (clinical/pathological complete response, MRI-based tumor regression grade and acute toxicities (CTCAE v5.0) were systematically assessed. Longitudinal stool samples (baseline, mid-treatment, post-treatment) underwent metagenomics and untargeted metabolomic. Analyses were performed to compare taxonomic diversity, species-level abundance, and metabolite intensities across treatment phases, clinical response and hematologic/gastrointestinal toxicities.

Results: At baseline, responders exhibited gut microbiome enrichment in Firmicutes, dominated by beneficial bacteria producing short-chain fatty acids, with LefSe analysis confirming discriminative power (LDA>2.0, p<0.05). Non-responders showed Proteobacteria/Bacteroidetes dominance, enriched in Klebsiella and pks+ pathogenic bacteria. Metabolically, responders demonstrated alkaloid biosynthesis and vitamin/cofactor metabolism upregulation (FDR<0.05), whereas non-responders overexpressed carbohydrate/lipid metabolism pathways. Longitudinally, iTNT increased microbial alpha diversity (Shannon index) with progressive expansion of beneficial Firmicute and reduction of pathogenic Proteobacteria. Co-occurrence network analysis revealed higher baseline ecological stability in responders, which intensified post-treatment. Severe toxicity cohorts showed persistent enrichment of pro-inflammatory pathogenic bacteria across timepoints. Conversely, mild-toxicity patients exhibited anti-oxidative pathways.

Conclusion: The efficacy and toxicity of iTNT in LARC are associated with gut microbiome and metabolic signatures. Baseline and longitudinal enrichment of SCFA-producing taxa positively correlates with treatment response, while persistent colonization by opportunistic pathogens may drive therapeutic resistance and severe toxicities. Microbiome-associated metabolic reprogramming, particularly in vitamin/cofactor metabolism and anti-inflammatory pathways, emerges as a critical modulator of clinical outcomes.