The dynamic landscape of immune environment for MM patients receiving T cell redirectiing therapies

H Haimeng Li (1Peking University Health Science Center, Beijing, China) X Xuelin Dou (1Peking University People's Hospital, Beijing, China) Y Yang Liu G Guang Hu (Ministry of Education Key Laboratory of Geriatric Diseases and Immunology, Suzhou Key Laboratory of Pathogen Bioscience and Anti-infective Medicine, Center for Systems Biology, School of Life Sciences, Suzhou Medical College of Soochow University) X Xiao-Dong Mo (1Peking University People's Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Hematopoietic Stem Cell Transplantation, Peking University, Beijing, China) N Nan Peng L Lei Wen (3Peking University People's Hospital, Beijing, China) Q Qian Li Y Yue Wu (Genomic Analysis Laboratory, Salk Institute for Biological Studies, La Jolla, CA, USA.) H Huixia Guo Y Yijuan Chen (5Department of Hematology, Fu Xing Hospital, Capital Medical University, Beijing, China) L Liru Wang (4Fu Xing Hospital Affiliated to Capital Medical University, Beijing, China) L Luyang Sun (State Key Laboratory of Fine Chemicals) J Jin Lu (Center for Biological Physics, Arizona State University)

Abstract

Abstract Background: Multiple myeloma (MM) is a major hematologic malignancy driven by clonal plasma cells in the bone marrow. BCMA-directed CAR T cells and BCMA × CD3 TCEs have yielded promising responses in heavily pretreated patients, yet their optimal sequencing remains unclear. Deeper insight into how these therapies differentially reshape the immune microenvironment is essential for identifying response biomarkers and advancing MM immunotherapy. Results: Comprehensive single-cell atlas of 102 patients and 178 longitudinal samples We profiled 934 k bone-marrow cells (scRNA + TCR + BCR) spanning pre-therapy, early (1 mo), mid (2–5 mo), late (≥6 mo), and post-relapse timepoints after BCMA-CAR-T (n = 59) or BCMA-TCE (n = 43) treatment, creating the largest immune atlas to date in MM T-cell–redirecting therapy. TCE efficacy is driven by expansion of endogenous CD8⁺ effector T cells, whereas CAR-T globally reprograms the T-cell pool TCE responders showed a stepwise rise in CX3CR1⁺ CD8_Tef cells; CAR-T instead converted endogenous CD8 compartments toward Tem, γδ T, and naive states, suggesting broader lineage diversification and potential durability. CAR-CD8_Tef_CX3CR1 emerges as the dominant responder subset after infusion CAR sequence–based sorting revealed that this subset accounts for most proliferative CAR-T clones in responders. STARTRAC analysis shows therapy-specific state transitions CAR-T promoted extensive clonal sharing between CD8_Tem, CD8_Proliferating, CD8_Tex, γδ T, and CD8_Naive, whereas TCE selectively expanded CD8_Tef without inducing further differentiation. Distinct TCR-clonal dynamics underline divergent immune activation Most clones were therapy-specific (Exclusive_Pre/Exclusive_Post). TCE yielded more Expanding CD8_Tef clones; CAR-T produced more Expanding CD8_Tem, CD8_Tex, and γδ T clones. Persisting clones were virtually absent in CD4 subsets. Tumor-reactive versus by-stander effector T cells defined by clonal size and expansion Tumor-reactive Tef (Ttr) exhibited heightened cytotoxic signatures. Endogenous Ttr preferentially expanded under TCE, whereas CAR-T responders were enriched for CAR-Ttr clones. Coordinated TME cellular programs exhibiting varied associations with responsiveness An NMF-based analysis revealed three efficacy-linked bone-marrow programs. The Cytotoxic-Program (TIME-Tef) comprises CD8_Tef_CX3CR1 and γδ T cells and correlates strongly with favorable responses. In parallel, an Immune-Response Program (TIME-Mono)—dominated by NK cells and CD14⁺ monocytes—also associates with clinical outcome but displays a distinct, therapy-dependent pattern Temporal kinetics reveal modality-specific biomarkers of response In TCE therapy, a monotonic rise of CD8_Tef_CX3CR1 from early to late timepoints stratified responders, with clonal expansion peaking in mid/late phases. After CAR-T, Tex, proliferating, and γδ T fractions surged Pre → Early → Mid, while CD8_Tef declined and CD4 memory/naive rebounded at Mid, mirroring tumor-burden reduction. After CAR-T, Tex, proliferating, and γδ T fractions surged Pre → Early → Mid, while CD8_Tef declined and CD4 memory/naive rebounded at Mid, mirroring tumor-burden reduction. Pretreatment exhausted-like Tex predicts TCE failure but not CAR-T outcome A TCE-specific Predictive Index that integrates CD8_Tex_TOX, CD4_Treg_FOXP3, and MALT_SLC4A10 predicts both inferior M-protein reduction (AUC = 0.82) and significantly shorter progression-free survival (P < 0.01). Infusion-product (IP) features forecast CAR-T durability High CAR-Treg abundance and excessive G2M/S-phase CAR-T cells predicted inferior responses, whereas a naïve/memory-biased IP signature marked long-acting CAR-T products and was detectable in patients ≥12 mo post-infusion. Relapse landscape highlights antigen escape and a stem-like myeloma clone In six paired BCMA-CAR-T relapses, expression of BCMA, GPRC5D, and SDC1 uniformly declined. BCR-sharing exposed a stem-like malignant cluster whose gene signature, selected by Cox + Lasso, predicted post–CAR-T relapse in COMPASS and multiple GEO cohorts. Conclusions: These findings delineate modality-specific immune-remodeling trajectories, uncover dynamic and baseline predictors of efficacy, and nominate a relapse-linked stem-like myeloma program, collectively guiding optimal sequencing and biomarker-driven personalization of BCMA-targeted immunotherapy in MM.

Article Details

Journal Blood
Volume / Issue Vol. 146, Issue Supplement 1
Published November 03, 2025
Pages 7533-7533
ISSN 0006-4971
Publisher Elsevier BV

Journal Info

Blood

Elsevier BV

ISSN: 0006-4971 Health Sciences

Authors (14)

H

Haimeng Li

1Peking University Health Science Center, Beijing, China

X

Xuelin Dou

1Peking University People's Hospital, Beijing, China

Y

Yang Liu

G

Guang Hu

Ministry of Education Key Laboratory of Geriatric Diseases and Immunology, Suzhou Key Laboratory of Pathogen Bioscience and Anti-infective Medicine, Center for Systems Biology, School of Life Sciences, Suzhou Medical College of Soochow University

X

Xiao-Dong Mo

1Peking University People's Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Hematopoietic Stem Cell Transplantation, Peking University, Beijing, China

N

Nan Peng

L

Lei Wen

3Peking University People's Hospital, Beijing, China

Q

Qian Li

Y

Yue Wu

Genomic Analysis Laboratory, Salk Institute for Biological Studies, La Jolla, CA, USA.

H

Huixia Guo

Y

Yijuan Chen

5Department of Hematology, Fu Xing Hospital, Capital Medical University, Beijing, China

L

Liru Wang

4Fu Xing Hospital Affiliated to Capital Medical University, Beijing, China

L

Luyang Sun

State Key Laboratory of Fine Chemicals

J

Jin Lu

Center for Biological Physics, Arizona State University