Short-Term Longitudinal Analysis of Gut Microbiota Dynamics During Anti-CD19 CAR-T Cell Therapy in Diffuse Large B-Cell Lymphoma Patients

Ziyuan Shen,1 Xing Xing,2 Kang Rong,3 Ziyang Geng,4 Ning Yang,5 Linyan Xu,3,6 Hongfeng Ge,7 Wei Sang3,6,8 1Clinical Research Institute, the Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, People’s Republic of China; 2Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA; 3Department of Hematology, The Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, People’s Republic of China; 4The Second Clinical Medical College, Xuzhou Medical University, Xuzhou, Jiangsu, People’s Republic of China; 5Department of Gastroenterology, The First People’s Hospital of Xuzhou, Xuzhou, Jiangsu, People’s Republic of China; 6Blood Diseases Institute, Xuzhou Medical University, Xuzhou, Jiangsu, People’s Republic of China; 7Department of Hematology, Bozhou People’s Hospital, Bozhou, Anhui, People’s Republic of China; 8Cell Research and Translational Medicine Center, The Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, People’s Republic of ChinaCorrespondence: Wei Sang, Department of Hematology, The Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, People’s Republic of China, Email xyfylbl515@xzhmu.edu.cn Hongfeng Ge, Department of Hematology, Bozhou People’s Hospital, Bozhou, Anhui, People’s Republic of China, Email ghfzz@163.comPurpose: Alterations in gut microbiota may influence immune response and treatment outcomes in patients with diffuse large B-cell lymphoma (DLBCL). However, the dynamics during anti-CD19 CAR-T cell therapy remain unclear.Methods: We conducted a short-term longitudinal microbiome analysis in DLBCL patients (n=12) undergoing CAR-T cell therapy targeting CD19. Stool samples were collected at baseline, 1 week, and 2 weeks post-infusion. 16S rRNA gene sequencing was used to assess microbial diversity, taxonomic composition, and functional pathways. Correlation analyses were then conducted between microbial taxa and inflammatory biomarkers.Results: Alpha diversity indices showed no statistically significant differences across time points. Beta diversity analysis revealed distinct clustering between baseline and week 1 samples in sPLS-DA, although PERMANOVA did not reach statistical significance. At the phylum level, Bacteroidota abundance significantly increased at week 2 compared with baseline (P = 0.008), accompanied by a marked reduction in the Firmicutes/Bacteroidota ratio. Genus-level heatmap and LEfSe analysis identified enrichment of Parabacteroides, and Prevotella at week 2, whereas baseline samples were enriched in Clostridium sensu stricto 13 and Fusobacterium. Functional prediction indicated that lipoic acid metabolism pathways were significantly upregulated at weeks 1 and 2 compared with baseline (both P < 0.05). Correlation analysis demonstrated that specific bacterial taxa, including Parabacteroides and Prevotella, were positively associated with lymphocyte counts and inversely correlated with C-reactive protein levels.Conclusion: Gut microbiota alterations following CAR-T infusion, characterized by increased Bacteroidota abundance, specific taxonomic shifts, and enhanced lipoic acid metabolism, may provide early microbial signatures for monitoring immune modulation in DLBCL patients.Keywords: anti–CD19 CAR–T, diffuse large B–cell lymphoma, microbiome dynamics