2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

SESSION

702. CAR-T Cell Therapies: Basic and Translational: Poster II

Program: Poster Abstract Session
Sunday, December 13, 2026
06:00 p.m. - 08:00 p.m.
MCCNO - Hall F
Technology and Procedures, Assays, Molecular Testing, Omics Technologies, Pathology, Bioinformatics, Cell Expansion, Computational Biology, Emerging Technologies, Gene Editing, Profiling, Treatment Considerations, Combination Therapy, Drug Development, Biological Therapies, Chimeric Antigen Receptor, (CAR)-T Cell Therapies, Immunotherapy, Adverse Events, Artificial Intelligence (AI), Deep Learning, Generative AI, Machine Learning, Biological Processes, Apoptosis, Pathogenesis, Genomics, Immune Mechanism, Metabolism, Microbiome, Molecular Biology, Multi-Systemic Interactions, Clinical Practice (Health Services and Quality), Quality Improvement, Diseases, Immunology, Autoimmune Disorders, Lymphoid Malignancies, Lymphoid Leukemias, Lymphomas, Plasma Cell Disorders, Myeloid Malignancies, Acute Myeloid Malignancies, ALL, Non-Hodgkin Lymphoma, B Cell Lymphoma, Aggressive Lymphoma, Research, Fundamental Science, AML, Translational Research, Clinical Research, Clinical Trials, Real-World Evidence, Registries, Survivorship, Study Population, Human, Adult, Animal Model
PAPER

4103 Lymphodepleting chemotherapy reveals baseline ecological vulnerability of the gut microbiome

Charan Thej Reddy Vegivinti, MD, MBBS, Abdur Rehman2, Lubna Rehman3, Faezeh Darbaniyan, PhD4, Donna Marie Weber, MD5, Melody Renee Becnel, MD6, Oren Pasvolsky, MD7, Sarvari V. Yellapragada, MD8, Mahmoud R. Gaballa, MD9, Sheeba Koshy Thomas, M.D.10, Mark R. Tanner, PhD, MBA, CMPP11, Paolo Strati, MD12, Ranjit Nair, MD13, Fateeha Furqan, MBBS14, Dai Chihara, MD, PhD15, Luis E. Fayad, MD16, Sairah Ahmed, MD17, Swaminathan P. Iyer, MD18, Jason Westin, MD, MS, FACP, FASCO19, Frederick L. Locke, MD20, Doris K. Hansen, MD21, Jing Christine Ye, M.D., MSc22, Krina Patel Jr, MD, MSc23, Sattva S Neelapu Sr, MD24, Robert Orlowski, MD, PhD25, Richard Champlin26, Christopher R. Flowers Jr, MD, MS27, Elizabeth Shpall, MD28, Eran Elinav, MD, PhD29, Christoph Stein-Thoeringer30, Robert Jenq, MD, PhD31, Michael D. Jain, MD, PhD32, Neeraj Saini, MD33
1 The University of Texas MD Anderson Cancer, Houston, Texas, United States
2 The University of Texas MD Anderson Cancer, Houston, Texas, United States
3 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
4 The University of Texas MD Anderson Cancer, Houston, Texas, United States
5 The University of Texas MD Anderson Cancer, Houston, Texas, United States
6 The University of Texas MD Anderson Cancer, Houston, Texas, United States
7 The University of Texas MD Anderson Cancer, Houston, Texas, United States
8 The University of Texas MD Anderson Cancer, Houston, Texas, United States
10 The University of Texas MD Anderson Cancer, Houston, Texas, United States
11 The University of Texas MD Anderson Cancer Center, Houston, Texas, Houston, Texas, United States
12 The University of Texas MD Anderson Cancer, Houston, Texas, United States
13 The University of Texas MD Anderson Cancer, Houston, Texas, United States
14 The University of Texas MD Anderson Cancer, Houston, Texas, United States
15 The University of Texas MD Anderson Cancer, Houston, Texas, United States
16 The University of Texas MD Anderson Cancer, Houston, Texas, United States
17 The University of Texas MD Anderson Cancer, Houston, Texas, United States
18 The University of Texas MD Anderson Cancer, Houston, Texas, United States
19 The University of Texas MD Anderson Cancer, Houston, Texas, United States
20 H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida, United States
21 H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida, United States
22 The University of Texas MD Anderson Cancer, Houston, Texas, United States
23 The University of Texas MD Anderson Cancer, Houston, Texas, United States
24 The University of Texas MD Anderson Cancer, Houston, Texas, United States
25 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States, The University of Texas MD Anderson Cancer, Houston, Texas, United States
26 The University of Texas MD Anderson Cancer Center, Houston, Texas, Houston, Texas, United States
27 The University of Texas MD Anderson Cancer, Houston, Texas, United States
28 The University of Texas MD Anderson Cancer Center, Houston, Texas, Houston, Texas, United States
29 German Cancer Research Center (DKFZ), Heidelberg, Baden-Württemberg, Germany
30 German Center for Infection Research, Tuebingen, Germany
31 City of Hope, Duarte, California, United States
32 H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida, United States
33 The University of Texas MD Anderson Cancer Center, Houston, Texas, Houston, Texas, United States, The University of Texas MD Anderson Cancer, Houston, Texas, United States

Introduction. Gut microbiome collapse, marked by loss of community diversity and intestinal mono-domination by a single bacterial taxon, is a recurrent dysbiosis after chimeric antigen receptor (CAR) T-cell therapy and is associated with cytokine release syndrome (CRS). Yet why comparable lymphodepleting conditioning destabilizes the microbiome in some patients but not others, remains unknown. Although we previously showed that pre-treatment antibiotics injure the microbiome in CAR T-cell recipients, antibiotics precede lymphodepletion in only a minority of patients, whereas fludarabine/cyclophosphamide (Flu/Cy) conditioning is given to nearly all. We therefore examined the effect of conditioning itself on the gut microbiome and tested whether susceptibility to collapse reflects the baseline ecological organization of the community—captured by dominance-sensitive structure—rather than the identity of any specific taxon.

Methods. We analyzed 461 longitudinal stool shotgun metagenomes from 169 large B-cell lymphoma and myeloma CAR T-cell recipients at MD Anderson and Moffitt Cancer Center across baseline (≤−4 days), peri-infusion (W0; −3 to +3 days), early (W1; +4 to +14 days), and late (W2; ≥+15 days) time windows. Taxonomy, function, metagenome-assembled genomes, strains, and dietary substrate were profiled. Mono-domination was defined as a single species with ≥30% relative abundance, and baseline organization was defined by evenness (Pielou's J′). Causality was tested in a meropenem-conditioned A20-lymphoma murine CAR T-cell model with or without dietary valerate.

Results. Flu/Cy conditioning produced a stereotyped, dose-independent ecological collapse. Mono-domination rose from 24% at baseline to 51% at W1 while Shannon diversity fell to its W1 nadir (P<10⁻⁴), with similar patterns across CAR products and cyclophosphamide doses. This was not a minor compositional shift: Conditioning erased a median of 32% and 35% of baseline species in lymphoma and myeloma patients, respectively. Peri-infusion mono-domination was associated with higher grade ≥2 CRS in lymphoma (50% vs 27%; odds ratio (OR) 2.65; P=0.029) and myeloma (50% vs 13%; OR 7.0, p=0.04) patients. Collapse was non-random and mechanistically coherent. Flu/Cy selectively depleted an obligate-anaerobe, fiber-fermenting guild (Faecalibacterium, Roseburia, Blautia) and favored aerotolerant organisms; dominator genomes encoded more oxidative-defense and respiratory systems than the lost anaerobes (median 5 vs 2 per genome; P=8×10⁻⁶), including catalase, cytochrome bd oxidase, and Mn/Fe-superoxide dismutase. Strain-resolved profiling showed domination arose through expansion of each patient's resident strains rather than acquisition of new organisms (92.9% within-patient strain persistence). Reduced dietary substrate and cachexia did not explain collapse, as food-derived DNA (MEDI) was not lower in dominated patients. Susceptibility was already evident before conditioning: low baseline evenness predicted subsequent domination (OR 0.22 per SD; P=0.001), outperformed species richness, and added information independent of antibiotics (no evenness-antibiotic interaction; P=0.84). Baseline evenness acted on CRS entirely through domination (mediation; direct effect P=0.95), and an evenness-based colonization-resistance score predicted domination in the held-out myeloma cohort (AUC 0.79). In in vivo CAR-T mouse models, dietary valerate supplementation preserved diversity and the Lachnospiraceae guild through conditioning (day +21 Shannon 3.20 vs 2.39; P=0.003), providing causal support.

Conclusions. Flu/Cy conditioning is a clinically uniform but ecologically disruptive stressor that depletes anaerobic gut microbes without recovery during the observed window. CAR T-cell-associated microbiome collapse is governed by pre-treatment dominance-sensitive community organization, not species richness or any single taxon; the dominant organism is typically an endogenous aerotolerant resident that expands after loss of anaerobe-mediated colonization resistance. Baseline evenness provides a single pre-infusion measure of microbiome resilience, and preserving ecosystem structure through antibiotic stewardship and short-chain fatty acid support is a rational strategy to reduce inflammatory toxicity after CAR T-cell therapy.

Charan Thej Reddy Vegivinti, MD, MBBS: No relevant conflicts of interest to declare.