This week features 20 clinical research articles, 3 review articles, and 18 basic research articles, covering all aspects of CAR-T cell therapy in hematological malignancies, solid tumors, and autoimmune diseases.
The weekly report is divided into four sections: Part I: Highlight Articles; Part II: Clinical and Translational Research — including new drug clinical trials, investigator-initiated trials (prospective/retrospective), translational research (bedside to bench), case reports and case series, and clinical prediction model development; Part III: Review Articles — including reviews on various aspects of CAR-T cell therapy; Part IV: Basic and Translational Research — including basic research, translational research (bench to bedside), technical updates, and preclinical studies aimed at optimizing CAR-T cell therapy and developing novel T-cell therapies.
Part I: Highlight Articles
Dual tumour-myeloid targeting of glioblastoma with GPNMB CAR-T cells
Journal & IF: Nature (56.1)
Corresponding author & institution: Sheila K. Singh (McMaster University, Hamilton, Ontario, Canada)
DOI: 10.1038/s41586-026-10641-1
Abstract: Glioblastoma is a lethal brain tumor against which current multimodal treatment regimens are almost invariably unable to prevent recurrence. The core reason for treatment failure is the high intratumoral cellular heterogeneity and a tumor microenvironment dominated by tumor-associated macrophages, which promote tumor proliferation and mediate immunosuppression. Several chimeric antigen receptor (CAR)-T cell therapies are under development for glioblastoma, but durable efficacy remains elusive: heterogeneous tumor antigen expression, antigen loss, and the inability to directly target the microenvironment barrier by targeting tumor cells alone all attenuate therapeutic responses. These limitations have prompted researchers to pioneer a novel strategy — no longer viewing glioblastoma as a single malignant cell population, but intervening in the tumor and immune system as a coupled whole. Using a multi-omics target screening platform, this study confirmed that GPNMB is a dual-compartment antigen simultaneously expressed on both tumor and myeloid cell subsets in glioblastoma. GPNMB-directed CAR-T cells demonstrated potent antitumor activity: in orthotopic patient-derived xenograft models and syngeneic glioma models, this therapy simultaneously eliminated GPNMB-positive tumor cells and immunosuppressive myeloid cells, achieving long-term disease control. This strategy integrates tumor killing with microenvironment remodeling, offering a novel antigen screening and targeted therapy approach for heterogeneous, myeloid-rich solid tumors.
GPNMB-directed CAR T cell therapy against MiT/TFE-family fusion-driven solid tumors
Journal & IF: Nat Cancer (28.0)
Corresponding author & institution: Douglas J. Mahoney (University of Calgary, Alberta, Canada)
DOI: 10.1038/s43018-026-01194-3
Abstract: The development of chimeric antigen receptor (CAR)-T cell therapy for solid tumors is constrained by the lack of safe, homogeneously expressed cell surface targets. This study identified glycoprotein NMB (GPNMB) as a protein regulated by MiT/TFE family fusion genes, showing high, homogeneous, and stable expression in primary and recurrent alveolar soft part sarcoma (ASPS) and translocation renal cell carcinoma. The authors constructed GPNMB-directed CAR-T cells (GCAR1), which demonstrated potent antitumor activity in patient-matched primary tumor cells, tumor organoids, and xenograft models. A post-hoc interim analysis of an open-label, single-subject first-in-human trial (NCT07104682) enrolled one patient with relapsed/refractory metastatic ASPS, showing that GCAR1 achieved disease stabilization for up to 3 months, with regression of most non-target lesions (meeting the primary endpoint), and was well tolerated. GCAR1 T cells underwent polyclonal expansion in peripheral blood and were detectable in vivo for up to 1 month. Spatial transcriptomics identified immunosuppressive microenvironment niches in treatment-resistant lesions; animal xenograft models confirmed that immune checkpoint blockade synergized with GCAR1. In summary, these data validate the conceptual feasibility of GCAR1 for GPNMB-positive solid tumors and provide a broad-spectrum rationale for targeting cell surface antigens regulated by oncogenic fusion genes with CAR-T therapy.
Long-term control of peritoneal metastases following claudin 18.2-targeted CAR T-cell therapy in advanced gastric cancer
Journal & IF: J Hematol Oncol (47.8)
Corresponding author & institution: Qi Changsong / Sun Yu (Peking University Cancer Hospital)
DOI: 10.1186/s13045-026-01826-2
Abstract: Gastric cancer patients with peritoneal metastases have a poor prognosis and limited treatment options. Claudin 18.2 (CLDN18.2)-targeted CAR-T cell therapy has demonstrated antitumor activity in advanced gastric cancer, but its efficacy against peritoneal metastatic lesions remains unclear. This article reports three cases from the Phase I satri-cel trial (NCT03874897) in gastric cancer patients with high CLDN18.2 expression and peritoneal metastases. All three patients achieved clinical and radiographic benefits with durable control of peritoneal metastases. One patient underwent conversion surgery 8 months after CLDN18.2-targeted CAR-T cell infusion; after subsequent resection of bilateral ovarian metastases, the disease did not progress further, achieving long-term control of peritoneal metastases. The other two patients had overall survival of 44 months and 35 months, respectively. Although circulating CAR-T cells were nearly undetectable in peripheral blood around day 30 post-infusion, CAR-T cell infiltration was still detected in the resected primary gastric tumor and ovarian metastatic tissues months after surgery. These results confirm that CLDN18.2 CAR-T cell therapy has significant and durable antitumor activity in gastric cancer with peritoneal metastases.
Part II: Clinical Research Advances
Circulating Tumor DNA Assessment of Disease Response in Large B-Cell Lymphoma: Lisocabtagene Maraleucel Versus Autologous Stem Cell Transplantation Standard Therapy
Journal & IF: J Clin Oncol (44.7)
Corresponding author & institution: Jeremy S. Abramson (Massachusetts General Hospital Cancer Center)
DOI: 10.1200/JCO-25-03051
Abstract: This article presents the circulating tumor DNA (ctDNA) correlative study results from the TRANSFORM trial (NCT03575351), comparing liso-cel with standard of care (salvage immunochemotherapy followed by high-dose chemotherapy and autologous stem cell transplantation [ASCT]) in second-line large B-cell lymphoma (LBCL). A total of 136 patients were enrolled. Using the ultra-sensitive PhasED-Seq technology, ctDNA was analyzed at prespecified time points (randomization, Day 43, Day 64, Day 126; 3 months post-liso-cel infusion, approximately 2 months post-ASCT) for correlation with clinical outcomes. In both treatment arms, ctDNA clearance (i.e., measurable residual disease [MRD] negativity) at each time point was associated with longer event-free survival (EFS); the liso-cel arm had a significantly higher proportion of patients achieving MRD negativity. Among patients achieving both complete response (CR) and MRD negativity, liso-cel provided greater benefit than ASCT, with longer EFS, progression-free survival (PFS), and duration of response. In patients who achieved CR after ASCT, subsequent ctDNA re-detection confirmed ctDNA’s ability to predict disease relapse. After adjusting for PET response status, MRD negativity remained significantly associated with EFS; interaction testing showed a significant interaction between PET response status and treatment regimen on EFS. In second-line LBCL, liso-cel achieves deeper and more durable ctDNA molecular clearance, consistent with its superior EFS and PFS over ASCT. ctDNA-based MRD detection has independent prognostic value beyond PET and can serve as a complementary biomarker for assessing treatment response and predicting tumor relapse.
Short ramp-up glofitamab halves mortality risk after anti-CD19 CAR T-cell therapy failure in patients with diffuse large B-cell lymphoma: final results of the LYSA BiCAR phase 2 trial with a pre-specified external control arm
Journal & IF: J Hematol Oncol (47.8)
Corresponding author & institution: Guillaume Cartron (Montpellier University Hospital, France)
DOI: 10.1186/s13045-026-01816-4
Abstract: Background: Patients with diffuse large B-cell lymphoma (DLBCL) who experience treatment failure after anti-CD19 CAR-T cell therapy have extremely poor survival and no established standard of care. Our team previously reported interim data from the LYSA BiCAR Phase 2 trial using a short ramp-up glofitamab regimen in CAR-T failure patients; this article reports the final survival data and a prespecified comparative efficacy analysis against a concurrent external control cohort constructed from academic databases. Methods: BiCAR is a multicenter single-arm trial enrolling adult patients with CD20-positive DLBCL who are primary refractory to CD19 CAR-T therapy or have first relapse/progression after CAR-T. Patients received ofatumumab pretreatment followed by intravenous glofitamab with a rapid step-up dosing over 8 days to a 30 mg maintenance dose, administered every 21 days for up to 11 cycles. The external control cohort for comparative efficacy analysis was sourced from the French DESCAR-T registry and the ALYCANTE Phase 2 trial, with selection criteria: patients who experienced CAR-T failure, received subsequent non-bispecific systemic therapy, met the core BiCAR eligibility criteria, and started salvage therapy within 1 month of the BiCAR dosing period. To eliminate data source bias, individual patient-level raw data from the DESCAR-T registry and ALYCANTE trial were extracted for both the glofitamab-treated group (BiCAR cohort) and the control group. A propensity score model incorporating major prognostic indicators was constructed, using multiple imputation combined with stabilized inverse probability of treatment weighting for adjustment. Sensitivity analyses included multiple weighting schemes and enrollment restrictions; restricted mean survival time (RMST) was used to evaluate survival outcomes between the two groups. Results: The BiCAR trial enrolled 47 patients, of whom 46 received glofitamab. With a median follow-up of 30.4 months, median overall survival (OS) was 17.3 months, with a 2-year OS rate of 38.3%. The comparative efficacy analysis included 45 glofitamab-treated patients and 133 control patients. The primary weighted analysis showed: median OS 19.6 months in the glofitamab group vs 7.6 months in the control group, with a mortality hazard ratio of 0.49 (95% CI 0.31–0.79; P=0.007). Multiple sensitivity analyses yielded consistent conclusions. Over the 2-year follow-up period, glofitamab significantly extended RMST by 5 months (P=0.007). Conclusion: This study represents a high-credibility comparative efficacy analysis demonstrating that short ramp-up glofitamab significantly improves survival probability compared with concurrent non-bispecific antibody salvage regimens in DLBCL patients with first relapse or progression after CD19 CAR-T therapy, supporting this agent as the preferred salvage option for eligible CAR-T-failed DLBCL patients.
Idecabtagene vicleucel and endogenous T-cell phenotypes linked to progression-free survival in relapsed multiple myeloma
Journal & IF: Hemasphere (11.3)
Corresponding author & institution: Shari Kaiser (Bristol-Myers Squibb)
DOI: 10.1002/hem3.70423
Abstract: To individualize immunotherapy and improve treatment benefit in multiple myeloma (MM), the association between chimeric antigen receptor (CAR) cells, autologous T-cell phenotypes, and progression-free survival (PFS) needs to be further elucidated. This study used 258 bone marrow aspirate samples from 107 MM patients treated with idecabtagene vicleucel in the KarMMa trial, employing multi-dimensional flow cytometry and computational flow analysis to characterize immune cell dynamics and explore their prognostic value. Baseline screening data showed that higher proportions of PD1-positive cytotoxic T cells and ICOS-negative TIGIT-positive regulatory T cells (Tregs) were associated with longer PFS. At month 1 post-infusion, patients with CAR-T cell proportion >1%, CAR-T CD4/CD8 ratio >0.09, and higher abundance of non-activated non-cytolytic CAR-T cells had longer PFS. Notably, at the last post-infusion time point, 115 autologous T-cell subsets were identified, of which 47 had prognostic predictive value; a lower proportion of ICOS-positive TIGIT-negative Tregs was the most significant indicator associated with long-term PFS. In summary, this large-sample, full-course immune dynamic profiling dataset of anti-BCMA CAR-T therapy confirms that pre-apheresis patient T-cell composition, post-infusion CAR-T cells, and in vivo autologous T-cell phenotypes are all correlated with PFS. These findings have the potential to be translated into routine laboratory tests, helping to more clearly elucidate response and resistance mechanisms to CAR-T therapy in R/R MM patients.
CD7 chimeric antigen receptor T cells in patients with relapsed or refractory CD7-positive acute myeloid leukemia
Journal & IF: Leukemia (8.8)
Corresponding author & institution: Hu Yongxian / Huang He / Zhang Hongsheng (Bone Marrow Transplantation Center, First Affiliated Hospital of Zhejiang University School of Medicine / Yakobio)
DOI: 10.1038/s41375-026-03017-x
Abstract: Although CAR-T cell therapy has revolutionized the treatment of hematological malignancies, its application in acute myeloid leukemia (AML) faces numerous challenges. Approximately 30% of AML patients have tumor cells expressing CD7, making it a promising therapeutic target. This Phase I trial (NCT04599556) evaluated CD7-directed CAR-T cell therapy in R/R CD7-positive AML patients. Using a standard 3+3 dose-escalation design with two dose levels, patients received a single infusion of autologous or donor-derived CD7 CAR-T cells. The primary endpoint was the incidence of dose-limiting toxicities. Fourteen patients were enrolled. Treatment-related adverse events included cytokine release syndrome (92.9%), grade 3–4 cytopenias (100%), grade 1 neurotoxicity (7.1%), and viral reactivation (78.6%). The objective response rate was 92.3%, with an MRD-negative remission rate of 84.6%. Most patients achieved initial responses, but 7 patients experienced CD7-negative tumor relapse. With a median follow-up of 172.5 days, 5 patients remained in remission; the 1-year OS rate was 34.3%, and the 1-year leukemia-free survival rate was 34.1%. These preliminary results indicate that CD7 CAR-T cell therapy for R/R AML has a manageable safety profile and shows preliminary antitumor efficacy, supporting further validation in larger trials.
Kidney Transplantation Post-Chimeric Antigen Receptor T-Cell (CAR-T) Therapy in Multiple Myeloma
Journal & IF: Am J Kidney Dis (9.7)
Corresponding author & institution: Aisha Shaikh (Hospital of the University of Pennsylvania)
DOI: 10.1053/j.ajkd.2026.04.012
Abstract: Approximately half of multiple myeloma (MM) patients have renal impairment, and renal failure portends a poor prognosis. Kidney transplantation is rarely performed in MM patients due to concerns about tumor recurrence, infection, and graft rejection. However, with therapeutic innovations — particularly anti-BCMA CAR-T cell therapy, which can induce deep and durable hematologic responses — new opportunities for kidney transplantation have emerged for select patients. This article reports a 71-year-old patient with Stage III MM, cytogenetically high-risk with 17p deletion, who progressed to end-stage renal disease requiring long-term hemodialysis. The patient failed to achieve deep remission with multiple prior lines of therapy until receiving cilta-cel CAR-T cell therapy, achieving MRD negativity. One year post-CAR-T, the patient underwent living-donor kidney transplantation with basiliximab induction and tacrolimus plus mycophenolate mofetil maintenance. Post-transplant complications included hypogammaglobulinemia, cytopenias, BK viremia, and acute rejection, all managed with immunosuppression adjustment and supportive care. At 17 months post-transplant, graft function was stable (serum creatinine 1.8 mg/dL), and hematologic MRD-negative CR was maintained. This case demonstrates the feasibility of kidney transplantation after CAR-T therapy in MM patients, while also exposing the numerous clinical challenges of this treatment pathway.
Predicting CAR-T outcomes in R/R DLBCL: a multicenter real-world study of a 5-index model
Journal & IF: Front Immunol (7.0)
Corresponding author & institution: Li Ping / Liang Aibin / Qian Wenbin / Wang Li (Shanghai Tongji Hospital / Second Affiliated Hospital of Zhejiang University School of Medicine / Ruijin Hospital, Shanghai Jiao Tong University School of Medicine)
DOI: 10.3389/fimmu.2026.1863194
Abstract: Introduction: Chimeric antigen receptor T-cell (CAR-T) therapy has transformed the treatment landscape for R/R DLBCL, but predicting which patients will achieve long-term survival benefit remains a major challenge. Methods: This multicenter retrospective real-world study aimed to validate the predictive value of a previously established efficacy prediction model in Chinese patients with R/R DLBCL treated with CD19 CAR-T. The study included 92 patients from four domestic centers who received CD19 CAR-T therapy between August 1, 2021 and November 30, 2024. A 5-index prediction model was used to assess prognosis: double-expressor lymphoma, TP53 gene abnormality, ECOG performance status ≥2, maximum tumor diameter ≥5 cm, and number of prior therapy lines ≥4. The primary endpoints were ORR, CR rate, PFS, and OS. Results: With a median follow-up of 14.6 months, the 5-index model achieved a C-index of 0.767, indicating good predictive performance. The model effectively stratified patients into different risk groups, with significant differences in PFS (P<0.0001) and OS (P=0.0007) across strata, and demonstrated superior predictive ability over traditional prognostic scoring systems such as IPI and R-IPI. Discussion: This 5-index risk model demonstrates stable and reliable predictive capability in real-world clinical settings, can inform individualized treatment decisions for Chinese DLBCL patients receiving CAR-T therapy, and warrants further optimization and multi-region external validation.
Transcriptomic landscape of peripheral T cells following CAR-T cell therapy in diffuse large B cell lymphoma
Journal & IF: Cancer Immunol Immunother (5.8)
Corresponding author & institution: Carolina Martínez-Laperche (Gregorio Marañón Health Research Institute, Madrid, Spain)
DOI: 10.1007/s00262-026-04444-6
Abstract: Introduction: CAR-T cell therapy has improved outcomes for R/R LBCL patients, but individual responses vary significantly. Baseline autologous T-cell status and post-infusion CAR-T cell differentiation levels are important contributors to this heterogeneity, yet the transcriptional programs regulating these processes have not been fully elucidated. Methods: This study performed RNA sequencing on CD3⁺ T cells from 26 LBCL patients at two time points: pre-leukapheresis (PA group, n=23) and day 14 post-CAR-T infusion (D14 group, n=9); differential gene expression analysis, gene set enrichment analysis (GSEA), and gene regulatory network inference were conducted. Results and interpretation: A total of 320 differentially expressed genes were identified in pre-apheresis autologous T cells, enriched in immune-related pathways including T-cell receptor (TCR) signaling, T-cell differentiation, IL-2-related signaling, TCR-mediated apoptosis regulation, and aberrant tumor pathways, as well as gene signatures indicative of immune dysfunction. Unsupervised clustering divided pre-apheresis T cells into two transcriptomic expression profiles, with dramatically different early disease progression rates between the two groups (91% vs 16%, P<0.001). This suggests that a T-cell state characterized by high activation but progressive functional exhaustion impairs CAR-T cell survival and persistence. At the pathway level, GSEA of pre-apheresis T cells showed significant enrichment of PI3K/AKT/mTOR, MYC target gene pathways, and apoptosis-related signatures. Conclusion: The transcriptomic heterogeneity of patient autologous T cells is highly correlated with early disease progression after CAR-T infusion, providing potential intervention targets for optimizing CAR-T persistence and enhancing antitumor efficacy.
Emapalumab and eltrombopag rescue for prolonged immune effector cell-associated hemophagocytic syndrome after CAR T-cell therapy in multiple myeloma
Journal & IF: Haematologica (8.2)
Corresponding author & institution: Tarek H Mouhieddine (Icahn School of Medicine at Mount Sinai, New York)
DOI: 10.3324/haematol.2026.300894
Abstract: CAR-T and bispecific antibodies are used in R/R MM, and immune effector cell-associated hemophagocytic syndrome (IEC-HS) is a novel toxicity with no standardized treatment. This article reports a high-risk RRMM patient who progressed to plasma cell leukemia with hypereosinophilic syndrome after anti-GPRC5D bispecific therapy and developed prolonged IEC-HS following cilta-cel CAR-T infusion. The patient met HLH-2004 and IEC-HS diagnostic criteria, with prolonged severe cytopenias requiring transfusion dependence, persistently markedly elevated ferritin, and a substantial increase in CXCL9 indicating IFNγ-driven persistent bone marrow inflammation. Prior glucocorticoids, anakinra, and ruxolitinib had limited efficacy; eltrombopag was added to improve platelets; based on high CXCL9 expression, emapalumab was initiated to target IFNγ, with IFNγ becoming undetectable within 1 week and CXCL9 normalizing within 3 weeks. The patient became transfusion-independent, and G-CSF use decreased significantly. Throughout, the patient maintained MRD-negative MM status, with 2-year follow-up maintaining stringent CR. This case confirms that CXCL9 can guide emapalumab use, and that combining emapalumab with eltrombopag can reverse IFNγ-driven refractory IEC-HS and restore hematopoiesis while preserving CAR-T antitumor efficacy; the presence of hemophagocytosis rather than hematopoietic failure in the bone marrow suggests that targeting inflammatory pathways rather than merely stem cell reinfusion is the core therapeutic approach, providing clinical evidence for individualized management of IEC-HS after cellular therapy.
A novel dose-dense strategy for CD19-directed CAR T-cell therapy is associated with durable responses without increased toxicity in patients with B-cell non-Hodgkin lymphoma
Journal & IF: Haematologica (8.2)
Corresponding author & institution: Jordan Gauthier (Fred Hutchinson Cancer Center)
DOI: 10.3324/haematol.2025.300321
Abstract: CD19-directed CAR-T cell therapy achieves durable remission in only 30%–40% of R/R B-NHL patients, and clinical efficacy urgently needs improvement. Dose-escalation studies have shown that increasing CAR-T cell dose enhances tumor control but causes intolerable severe adverse events. This study established the maximum tolerated dose of JCAR014 in B-NHL patients as 2×10⁶ CAR-positive cells/kg. The authors hypothesized that a second infusion at the same dose on Day 14 (dose-dense schedule) without repeat lymphodepletion would be safe and enhance antitumor efficacy. This article presents 8-year follow-up results from the dose-dense lead-in cohort of a Phase I/II trial (NCT01865617). Twenty patients were enrolled, with two CAR-T cell products prepared, each at 2×10⁶ CAR-positive cells/kg; 17 patients completed both infusions. In the dose-dense group, the incidences of any-grade CRS and neurotoxicity (NT) were 59% (10 patients) and 18% (3 patients), respectively; except for one case of grade 2 CRS, all adverse events occurred only after the first infusion. Despite no additional lymphodepletion, 94% (16 patients) showed CAR-T cell re-expansion after the second infusion. By Lugano criteria, the ORR was 47% (8/17) and CR rate 41% (7/17). Among all responders, the 8-year durable response rate was 63% (95% CI 37–100), significantly superior to the 26% 8-year durable response rate (95% CI 14–48) in the single-infusion group. In summary, early second dosing on Day 14 is feasible and safe, enabling durable antitumor responses in R/R B-NHL patients.
Female sex is associated with superior outcomes in patients treated with CD19-directed CAR T-cell therapy for large B-cell lymphoma
Journal & IF: Haematologica (8.2)
Corresponding author & institution: Muhammad Bilal Abid (Texas Tech University)
DOI: 10.3324/haematol.2026.300993
Abstract: Conventional immunotherapy has demonstrated that sex affects antitumor efficacy, but the impact of sex on outcomes of CD19 CAR-T therapy in LBCL remains unclear. This study used real-world data from 1,503 patients in the Center for International Blood and Marrow Transplant Research to compare CAR-T outcomes between male and female patients, with PFS as the primary endpoint and OS, relapse, and non-relapse mortality as secondary endpoints. The cohort included 601 females and 902 males, median age 62 years. At 24-month follow-up, males had higher cumulative incidence of relapse, with significantly inferior PFS and OS compared with females; after multivariable adjustment, male sex remained an independent risk factor for progression, death, and relapse. Age-stratified analysis using a 50-year cutoff to approximate menopausal status showed that sex differences were more pronounced in the under-50 population, with young females having substantially lower relapse risk and significant survival advantage; in the over-50 population, males had only mildly worse PFS, with no significant differences in relapse or OS. Mechanistically, estrogen enhances immune responses while testosterone is immunosuppressive, consistent with these findings. This is the first large-sample analysis to explore the impact of sex on CAR-T outcomes with age stratification, suggesting that sex, age, and hormonal status modulate CAR-T efficacy. Future CAR-T trials should incorporate sex-stratified analyses to inform individualized therapy. This is a retrospective observational study with limitations including inability to adjust for confounders such as baseline inflammation and tumor burden; prospective validation is needed.
A proliferation-inducing ligand enhances the anti-tumor effect of BCMA CAR T-cell through inhibition of soluble BCMA
Journal & IF: Biochim Biophys Acta Mol Basis Dis (5.0)
Corresponding author & institution: Pan Bin / Yan Zhiling (Affiliated Hospital of Xuzhou Medical University)
DOI: 10.1016/j.bbadis.2026.168342
Abstract: A proliferation-inducing ligand (APRIL) is a ligand for BCMA and may affect the efficacy of BCMA-targeted CAR-T cell therapy in MM. This study found that in patients responding to BCMA CAR-T therapy, plasma APRIL levels increased (>1 nM) in the first month of treatment, and APRIL concentration was negatively correlated with soluble BCMA (sBCMA) levels. APRIL binds to sBCMA, reducing the detection efficiency of sBCMA by ELISA, an effect distinct from that of γ-secretase inhibitors. APRIL also binds to membrane-bound BCMA, but with much lower affinity than its binding to sBCMA. In vitro experiments showed that sBCMA binds to BCMA CAR-T cells, weakening their binding and killing of myeloma cells; however, 2 nM and 10 nM APRIL blocked the interaction between sBCMA and BCMA CAR-T, restoring CAR-T binding and killing activity against myeloma cells. In a myeloma xenograft mouse model, APRIL administration in the first week after BCMA CAR-T infusion increased the proportion of IFNγ-positive CAR-T cells within tumors and enhanced in vivo antitumor efficacy. In summary, APRIL antagonizes the inhibitory effect of sBCMA on BCMA CAR-T cell binding and killing of myeloma cells, a mechanism that may help improve the clinical efficacy of BCMA CAR-T therapy.
Patient-Specific Determinants of Response to BCMA- and GPRC5D-Targeted CAR T-Cell Therapy in Multiple Myeloma: A QSP Analysis of Clinical Trial and Real-World Data
Journal & IF: Clin Pharmacol Ther (4.9)
Corresponding author & institution: Vasiliki Kostiou (Certara, Applied Biosimulation)
DOI: 10.1002/cpt.70378
Abstract: Although CAR-T cell therapy has shown encouraging efficacy in R/R MM, nearly all patients eventually relapse. Resistance and relapse are driven by both CAR-T cell-intrinsic factors and tumor-intrinsic factors. This study used detectable biomarkers to construct a mechanistic quantitative systems pharmacology (QSP) model simulating the entire process of MM progression and CAR-T therapy, aiming to predict and identify key factors influencing treatment response and relapse. The model incorporates core components that resolve disease dynamics and CAR-T cell expansion patterns, and can reproduce published pharmacokinetic and biomarker data from BCMA- and GPRC5D-targeted CAR-T therapies. Model validation was performed using clinical biomarker data from 29 real-world RRMM patients treated with commercial anti-BCMA CAR-T. Virtual clinical trials simulating the impact of baseline tumor burden and CAR-T cell parameters on efficacy suggested that poor prognostic factors fall into two categories: tumor-intrinsic factors (high tumor burden, low antigen expression) and CAR-T cell-related factors (low killing efficiency). Interestingly, simulations showed that patients with lower baseline normal plasma cell proportions had higher ORR. The model also predicted outcomes of dual BCMA/GPRC5D CAR-T combination therapy; sequential administration simulations suggested that anti-GPRC5D CAR-T followed by anti-BCMA CAR-T would yield better responses. This model provides an analytical framework for investigating efficacy mechanisms and multi-target combination strategies, while also aiding in optimizing clinical trial design and dosing regimens.
Impact of Disease Biology and Bridging Strategy on Outcomes After CAR-T Cell Therapy in Relapsed/Refractory Multiple Myeloma
Journal & IF: Transplant Cell Ther (4.7)
Corresponding author & institution: Irene Strassl (Department of Internal Medicine I: Stem Cell Transplantation, Hematology, Plasmatology and Medical Oncology, Ordensklinikum Linz, Linz, Austria)
DOI: 10.1016/j.jtct.2026.06.048
Abstract: Background: R/R MM patients typically receive bridging therapy to maintain disease stability during manufacturing of BCMA-targeted CAR-T cells. However, there is no unified standard defining optimal bridging regimens or treatment intensity, and it remains controversial whether patient outcomes are primarily determined by bridging therapy or by tumor biology. Objective: To evaluate the impact of different bridging regimens on outcomes after BCMA CAR-T therapy, adjusting for confounders including tumor biology, pre-CAR-T response, and baseline hematopoietic reserve. Design: This multicenter retrospective real-world study included 90 consecutive Austrian patients treated with BCMA-targeted CAR-T (idecabtagene vicleucel or cilta-cel) between January 2024 and July 2025. Bridging regimens were categorized as: polychemotherapy (PCHT), talquetamab-based (Tal), and other regimens. Kaplan–Meier curves, univariate and multivariate Cox proportional hazards models for PFS, and logistic regression for early adverse events were used. Results: With a median follow-up of 9.9 months, estimated median PFS was 17.1 months. Univariate analysis unadjusted for confounders showed inferior PFS in the polychemotherapy bridging group compared with other regimens; however, after adjusting for extramedullary disease/plasma cell leukemia (EMD/PCL) and IMWG high-risk status, this difference disappeared. CAR-T product type and number of prior therapy lines had no significant impact on PFS. In multivariate analysis, EMD/PCL was the strongest independent predictor of poor prognosis (adjusted HR 4.85), while IMWG high-risk status had no independent prognostic value. Secondary models showed that achieving at least very good partial response before CAR-T infusion only showed a trend toward PFS improvement without statistical significance. Talquetamab-bridged patients had numerically superior PFS without increased adverse events. CAR-HEMATOTOX score independently indicated poor prognosis. Key conclusions: Polychemotherapy bridging did not increase CAR-HEMATOTOX score, nor did it cause prolonged cytopenias, increased infectious complications, or reduced CAR-T expansion markers (ALC). Conclusion: This national real-world study confirms that tumor biology — particularly the presence of EMD/PCL — is the core determinant of CAR-T outcomes, with far greater impact than bridging regimen. The previously observed worse outcomes after intensive chemotherapy bridging essentially stem from patients’ baseline high-risk tumor characteristics rather than negative effects of the bridging therapy itself. Clinically, bridging regimens should be individualized based on tumor biology, and adequate tumor burden control before CAR-T infusion should be emphasized.
Allogeneic Hematopoietic Cell Transplant Following Standard of Care Brexucabtagene Autoleucel (Brexu-cel) in Adults with B-Cell Acute Lymphoblastic Leukemia (B-ALL): Results from the Real-World Outcomes Collaborative of CAR T in Adult ALL (ROCCA)
Journal & IF: Transplant Cell Ther (4.7)
Corresponding author & institution: Rawan G Faramand (Moffitt Cancer Center)
DOI: 10.1016/j.jtct.2026.06.039
Abstract: Background: Brexu-cel is an autologous anti-CD19 CAR-T cell therapy approved for adults with R/R B-ALL. Although efficacy is substantial, most patients relapse after treatment, and consolidation strategies to prolong remission duration are urgently needed; allogeneic HCT is one such option. This article reports outcomes of adult B-ALL patients who received consolidative allogeneic HCT after commercial brexu-cel therapy, using the ROCCA database. Methods: This multicenter retrospective analysis included adult R/R B-ALL patients from the ROCCA database who achieved CR after commercial brexu-cel and subsequently received consolidative allogeneic HCT. The ROCCA collaborative group covers 41 U.S. centers, with retrospective data on adult B-ALL patients treated with brexu-cel from 2021–2025; patients who underwent transplant after CAR-T failure were excluded. The primary endpoints were 1-year post-transplant OS and EFS; secondary endpoints included relapse, non-relapse mortality (NRM), and GVHD. Results: Of 399 patients receiving brexu-cel, 65 subsequently underwent consolidative allogeneic HCT, of whom 56 had first transplant and 9 had second transplant. First-transplant patients had a median age of 34 years and a median of 3 prior therapy lines, representing a heavily pretreated population. With a median post-transplant follow-up of 11 months, estimated 1-year post-transplant OS was 79% (95% CI 64%–88%), and 1-year EFS was 66% (95% CI 51%–77%); 1-year cumulative relapse rate was 19%, and NRM was 13%. Grade II–IV acute GVHD occurred in 18%, grade III–IV severe acute GVHD in 4%, and moderate-to-severe chronic GVHD in 12%. Univariate analysis showed age ≥40 years was a poor prognostic factor for OS (HR 4.31, 95% CI 1.40–13.3) and EFS (HR 3.36, 95% CI 1.43–7.91); myeloablative conditioning improved EFS (HR 0.37, 95% CI 0.15–0.93). In 9 second-transplant patients, 1-year OS and EFS were both 59% (95% CI 19%–85%), with 1-year NRM as high as 41% and no relapses during follow-up. Conclusion: This large real-world cohort confirms that consolidative allogeneic HCT after brexu-cel-induced remission is clinically feasible, with favorable survival, low relapse rates, and generally manageable toxicity, particularly suitable for transplant-naïve patients. These results support allogeneic HCT as an effective consolidation strategy after CAR-T-induced remission, while also suggesting the need for prospective studies to optimize patient selection criteria, conditioning regimens, and transplant strategies post-CAR-T.
The association between hyponatremia and immune effector cell-associated neurotoxicity syndrome after chimeric antigen receptor T cell therapy
Journal & IF: Transplant Cell Ther (4.7)
Corresponding author & institution: Nobuharu Fujii (Department of Hematology, Okayama University Hospital)
DOI: 10.1016/j.jtct.2026.06.045
Abstract: ICANS is a high-risk fatal adverse event of CAR-T therapy. Hyponatremia often accompanies CRS and neurological inflammation. This single-center retrospective study explored the association between hyponatremia and ICANS. A total of 121 adult non-Hodgkin lymphoma patients treated with CAR-T at Okayama University from 2019–2025 were included. CRS and ICANS were graded per ASTCT criteria; serum sodium, osmolality, ADH, and IL-6 were dynamically monitored. ICANS occurred in 18.2% of patients, with a median onset on Day 6. There was no difference in serum sodium between groups at baseline; from Day 3 onward, the ICANS group had persistently lower sodium and significantly lower nadir values. Persistent hyponatremia (defined as a decrease lasting >3 days after CRS onset) was an independent predictor of ICANS, with a multivariate HR of 6.5. Mechanistically, this hyponatremia was consistent with the syndrome of inappropriate antidiuretic hormone secretion: low plasma osmolality with elevated ADH and urine sodium. IL-6 both induces inflammation and stimulates ADH secretion, disrupting the blood-brain barrier and jointly driving ICANS and hyponatremia. The limitation of this study is its single-center design. The findings confirm that persistent sodium decline after CRS can warn of ICANS, supporting early sodium monitoring for risk stratification; multi-center validation is needed.
Comparison of Allo-HSCT outcomes after CAR-T therapy versus chemotherapy in pediatric patients with relapsed/refractory B-ALL: a retrospective study
Journal & IF: Oncologist (4.7)
Corresponding author & institution: Hu Shaoyan / Sun Ruijuan / Chen Jing / Xiao Peifang (Children’s Hospital of Soochow University / Lu Daopei Medical / Shanghai Children’s Medical Center Hematology Oncology Center)
DOI: 10.1093/oncolo/oyag247
Abstract: Background: HSCT is a curative option for R/R B-ALL. Both conventional chemotherapy and CAR-T therapy are important strategies for achieving MRD negativity before transplantation, but which is superior remains unclear. Patients and methods: This multicenter study included 269 R/R B-ALL patients undergoing allogeneic HSCT, of whom 142 received CAR-T bridging and 127 received chemotherapy bridging before transplant. Results: Three-year post-transplant OS: CAR-T group 66.8%, chemotherapy group 72.3%; 3-year relapse-free survival: CAR-T group 65.3%, chemotherapy group 65.9%. There were no statistical differences in 3-year GVHD-free relapse-free survival or cumulative relapse rate between the two groups. Among patients in CR1 before transplant, the proportion was significantly higher in the chemotherapy bridging group than in the CAR-T bridging group (37.5% vs 13.6%, P<0.001); in this subgroup, the chemotherapy group had superior OS, relapse-free survival, and GVHD-free relapse-free survival. Conclusion: In pediatric R/R B-ALL, CAR-T as a bridging regimen to transplant yields comparable long-term post-transplant survival to chemotherapy bridging, but with a higher incidence of moderate-to-severe chronic GVHD. CAR-T can serve as a feasible bridging option, and enhanced GVHD monitoring is needed in clinical practice.
Relationship between early use of tocilizumab during chimeric antigen receptor T-cell therapy for multiple myeloma and cardiovascular risk and progression-free survival
Journal & IF: Eur Heart J Open (4.3)
Corresponding author & institution: Sophia Golec (Tufts Medical Center)
DOI: 10.1093/ehjopen/oeag097
Abstract: Background: CAR-T therapy is approved for R/R MM. Due to age and prior cardiotoxic therapies, MM patients have a high burden of cardiovascular comorbidities. Previous retrospective studies have shown that the incidence of major adverse cardiovascular events (MACE) in patients with severe CRS can reach 20%–30%. Early use of the IL-6 receptor antagonist tocilizumab to intervene in CRS may reduce MACE risk. Objective: This observational study aimed to investigate the association between early tocilizumab use and MACE. Methods: Single-center retrospective chart review of adult MM patients treated with CAR-T from 2017–2023. MACE was a composite endpoint including myocardial injury, heart failure, stroke, arrhythmia, and cardiovascular death, with a median follow-up of 12 months. Secondary endpoints were PFS and OS. Cumulative incidence function was used to calculate MACE incidence with non-cardiovascular death as competing risk, with Gray’s test for between-group comparisons; Kaplan–Meier curves and log-rank tests were used for PFS and OS. Results: A total of 145 patients were included; 120 (82%) developed CRS; of these, 107 (89%) received tocilizumab within 24 hours of CRS onset. Within 1 year of treatment, 14 patients (9.7%) developed MACE, all occurring only in patients who had CRS. The 6-month cumulative MACE incidence was 8% (95% CI 5%–14%), and 12-month incidence was 11% (95% CI 7%–19%); 12-month PFS rate was 58% (95% CI 50%–68%). Conclusion: The occurrence of CRS had no significant impact on PFS or OS.
Response-adapted chimeric antigen receptor T cell (CAR-T)–sparing consolidation radiotherapy in high-risk large B-cell lymphoma (LBCL): Results of the prospective RESTART protocol
Journal & IF: Br J Haematol (3.6)
Corresponding author & institution: N. G. Mikhaeel (Guy’s Cancer Centre, London, UK)
DOI: 10.1111/bjh.70653
Abstract: Early relapse remains a challenge in CAR-T therapy for R/R LBCL, particularly in high-risk cases or patients with poor early response. This prospective RESTART study explored a risk-stratified combination strategy of radiotherapy and CAR-T. Based on lesion distribution and tumor biology, patients were assigned to radiotherapy-dominant Path A or systemic bridging Path B. The core innovation of this protocol is response-adapted consolidation radiotherapy: patients with Deauville score (DS) 4–5 at 1 month post-CAR-T, or DS 3 with baseline high-risk lesions (maximum diameter ≥5 cm or SUVmax ≥15), received consolidation radiotherapy. Radiotherapy was delivered at 6–8 weeks post-infusion using techniques that spare circulating CAR-T cells by reducing peripheral blood radiation exposure. A total of 190 patients were enrolled, of whom 55 (29%) received radiotherapy. In-field local control rate was 89%, and 92% in the consolidation-only radiotherapy population. With a median follow-up of 12.4 months, the overall cohort had 1-year PFS of 57% and 1-year OS of 71%. Even in patients with multiple high-risk factors, those receiving consolidation radiotherapy had 1-year PFS of 54% and OS as high as 95%, with no observed increase in systemic relapse due to radiotherapy-induced CAR-T damage. This study confirms that response-adapted, CAR-T-sparing consolidation radiotherapy is clinically feasible and can improve outcomes in high-risk LBCL patients.
A federated digital twin reveals cytomegalovirus reactivation impairs CAR-T cell therapy via IL-15-mediated cytokine competition in B-Cell lymphoma
Journal & IF: Front Immunol (7.0)
Corresponding author & institution: Mini Ghosh (Nanjing Drum Tower Hospital)
DOI: 10.3389/fimmu.2026.1844927
Abstract: Introduction: Among CMV-seropositive B-cell lymphoma patients receiving CAR-T therapy, 30%–40% experience CMV reactivation, which is highly associated with treatment failure. However, the immunological causal mechanism by which CMV impairs efficacy remains unclear — whether due to direct viral cytotoxicity, T-cell exhaustion, or cytokine resource competition; existing predictive models also lack mechanistic support to guide clinical intervention. This study constructed a privacy-preserving mechanistic digital twin model to validate the "cytokine pool depletion hypothesis": CMV-specific CD8⁺ T cells compete with CAR-T cells for the limited homeostatic cytokine IL-15. Methods: A system of ordinary differential equations was established to describe the competition between the two T-cell populations for IL-15. A hierarchical Bayesian federated averaging algorithm was used to train the model on multi-center data from 414 patients across 5 centers without pooling raw patient data; an independent 6th-center cohort of 89 patients served as the prospective validation set, not participating in model training. In silico counterfactual simulations and antiviral prophylaxis simulations were performed. Results: This digital twin model could predict clinically significant CMV reactivation (viral load ≥10,000 IU/mL, the threshold referenced from ECIL-7 guidelines for T-cell-depleting immunotherapy to guide preemptive antiviral intervention) at Day 28 post-CAR-T infusion, with an AUROC of 0.91 (95% CI 0.88–0.94), significantly superior to existing clinical risk scores. Two prespecified secondary thresholds were also analyzed: any detectable CMV reactivation (≥1,000 IU/mL) and severe CMV reactivation (≥50,000 IU/mL). Quantitative results from the fitted mechanistic model confirmed that IL-15 resource competition is the core mechanism by which CMV reactivation impairs CAR-T function. Counterfactual analysis showed that CMV reactivation reduced CAR-T expansion peak by 41.8% (p<0.001). Global sensitivity analysis identified pre-infusion CMV-specific T-cell precursor proportion (ξ) and resource competition coefficient (α) as the two core drivers of this effect, explaining 38% and 29% of outcome variance, respectively. Digital twin-based risk-stratified individualized antiviral prophylaxis simulations showed that the strategy could reduce the estimated 6-month disease progression risk by 32% while reducing overall drug exposure by 36%. In the prospective validation cohort, model-predicted CAR-T kinetic impairment was an independent poor prognostic factor for PFS (HR=3.4, 95% CI 1.5–7.8, p=0.004). Comparative analysis with three alternative mechanistic models further confirmed that the cytokine pool depletion hypothesis outperformed mere T-cell exhaustion and mere mediator-mediated hypotheses. Conclusion: Within the framework of this fitted mechanistic model, clinical data quantitatively confirm that IL-15 resource competition is the core mechanism by which CMV reactivation impairs CAR-T function; multi-model comparison further supports the cytokine pool depletion hypothesis, though causal determination still requires randomized intervention studies. This study demonstrates that privacy-preserving mechanistic digital twins can be used for early clinical risk stratification and individualized intervention, while providing a replicable paradigm for large-scale collaborative systems immunology research. The clinical value of this model still requires randomized controlled trials with prospective model-guided treatment for final validation.
Part III: Review Articles
A guide to CAR T cell therapies: development, current status and future prospects
Journal & IF: Nat Rev Immunol (47.1)
Corresponding author & institution: Padmanee Sharma (MD Anderson Cancer Center, University of Texas)
DOI: 10.1038/s41577-026-01322-1
Abstract: Since the first product was approved in 2017, CAR-T cell therapy has become one of the most potent means of reshaping antitumor immune responses. Built upon decades of fundamental research in T-cell biology and synthetic immunology, CAR-T cell therapy has revolutionized the treatment of B-cell malignancies, enabling durable remissions in B-cell leukemias, lymphomas, and multiple myeloma. Next-generation CAR designs are expanding the therapeutic scope to autoimmune diseases and solid tumors. Technological innovations including gene editing, allogeneic CAR-T manufacturing, and in vivo delivery are continuously improving the manufacturability, safety, and accessibility of CAR-T therapy; however, overcoming tumor antigen heterogeneity, the immunosuppressive tumor microenvironment barrier, and maintaining long-term CAR-T cell persistence remain urgent challenges. This review summarizes the key foundational discoveries underlying CAR-T therapy, systematically reviews current CAR molecular design principles, clinical development progress, and various emerging optimization strategies — aimed at enhancing therapeutic efficacy, broadening indications, and achieving durable immune antitumor control across different disease types.
The evolving role of cytokines for CAR-T cell manufacturing and beyond
Journal & IF: Nat Biomed Eng (26.3)
Corresponding author & institution: Sebastian Kobold (LMU Hospital, Munich)
DOI: 10.1038/s41551-026-01703-w
Abstract: CAR-T cells are a clinically validated therapeutic modality in cancer and beyond in recent years. However, widespread industrial implementation faces multiple technical, logistical, regulatory, and economic challenges. Ex vivo production is a major bottleneck, and cytokines are an indispensable component of this process that directly determines product attributes. To better understand the complexity of cytokine use throughout the manufacturing process, we systematically reviewed the literature, covering detailed manufacturing parameters from 292 published clinical reports. We found that the duration and dose of IL-2 use are progressively declining, driven by its association with unfavorable cell phenotypes. In preclinical studies, this trend has prompted evaluation of alternative cytokines that may maintain T-cell naive phenotypes, support expansion, and enhance CAR-T efficacy. This article describes the evolution of cytokine use, reveals the lack of standardization in clinical CAR-T manufacturing parameters, and summarizes preclinical and next-generation technology concepts that may improve manufacturing efficiency, reduce costs, and improve outcomes. Our observations may further guide the development of cytokine-armoring strategies to promote CAR-T expansion and persistence in patients; meanwhile, technological innovations such as short-process manufacturing and in vivo engineering may reduce dependence on cytokines during the ex vivo culture phase.
Neurological complications of current and emerging CAR-T cell therapies
Journal & IF: Neurotherapeutics (7.4)
Corresponding author & institution: Jorg Dietrich (Massachusetts General Hospital Cancer Center, USA)
DOI: 10.1016/j.neurot.2026.e00951
Abstract: Mobilizing the patient’s own immune system to enhance antitumor responses has long been a goal in oncology. Multiple immunomodulatory or immunostimulatory therapies have been FDA-approved or are under development, employing various strategies to activate the patient’s immunity. CAR-T cell therapy is one such approach: genetically engineering T cells to target tumor-specific antigens. Several CAR-T products are now clinically approved, achieving significant and durable responses in hematological malignancies. However, alongside antitumor efficacy, there are risks of complications, with neurological adverse events being particularly frequent. Building on the success of cellular therapy in hematological tumors, researchers are expanding this technology to solid tumors (including CNS tumors) while also exploring applications in non-oncological diseases such as autoimmune diseases. As novel CAR-T therapies are developed and brought to clinical use, assessing their neurotoxicity risk and developing prevention and treatment strategies is of great importance. This review summarizes the acute and potential long-term neurological complications of these potent novel cellular therapies, with emphasis on next-generation CAR-T technologies, their associated neurotoxicity, and interventions that may mitigate neurological damage.
Part IV: Basic and Translational Research
CAR T cell therapy selectively depletes disease-driving mutant calreticulin cells in xenotransplants and human organoid models of myelofibrosis
Journal & IF: Sci Transl Med (15.6)
Corresponding author & institution: Martin A. Pule (UCL Cancer Institute)
DOI: 10.1126/scitranslmed.adz3553
Abstract: Targeted immunotherapy has rewritten the treatment outcomes for lymphoid malignancies, but its efficacy in myeloid tumors is limited due to the lack of suitable therapeutic targets and the immunosuppressive properties of the tumor microenvironment. Myeloproliferative neoplasms (MPN) are chronic myeloid hematological malignancies, one-third of which are driven by calreticulin gene mutations. Mutant calreticulin (mutCALR) binds and activates the thrombopoietin receptor (TpoR), exposing the mutant calreticulin-thrombopoietin receptor complex on the cell surface of pathogenic cells, creating a therapeutic vulnerability. This study constructed a CAR-T cell therapy that specifically targets mutCALR-positive cells in vitro and in vivo. This CAR-T therapy selectively eliminated mutCALR-positive stem cells from myelofibrosis patient samples without damaging normal hematopoietic stem cells, while extending survival in mutCALR-positive leukemia xenograft mouse models. To model the fibrotic bone marrow microenvironment, human chimeric organoid models were constructed, confirming that CAR-T cell killing efficacy was not diminished even in a fibrotic tumor microenvironment. The study also established a method to upregulate mutCALR expression on CD34-positive cells from patients with advanced-phase/blast-phase MPN (defined as peripheral blood or bone marrow blasts >10%), further enhancing CAR-T targeted killing. This study presents a potentially curative approach for mutCALR-driven malignancies and establishes an innovative research system for evaluating targeted immunotherapies for hematological tumors in tumor microenvironments that closely resemble true pathological states.
Lymph Node-Targeting Nanotherapy Combined with Photothermal Therapy to Potentiate Chimeric Antigen Receptor-T Cell Treatment of Oral Cancer
Journal & IF: ACS Nano (17.3)
Corresponding author & institution: Zhang Ming / Liu Laikui / Tang Benzhong (School of Biomedical Engineering and Informatics, Nanjing Medical University / School of Stomatology, Nanjing Medical University / The Chinese University of Hong Kong)
DOI: 10.1021/acsnano.6c09366
Abstract: Oral squamous cell carcinoma (OSCC) is one of the most common malignancies of the head and neck. Immunotherapy has become a promising treatment for metastatic OSCC due to its considerable clinical benefits; however, the immunosuppressive tumor microenvironment, short duration of efficacy, and insufficient T-cell tumor infiltration greatly limit its efficacy. To address these challenges, this study constructed a synergistic strategy combining photothermal therapy (PTT) with CAR-T cell immunotherapy. The conjugated polymer nanoparticles (CPNPs) prepared in this experiment are highly efficient near-infrared II (NIR-II) photothermal agents that enable localized PTT and induce strong immunogenic cell death (ICD) to activate T cells. In addition, a lymph node-targeting nano-delivery system (ApoA1@PNPs) was constructed to enhance in vivo CAR-T cell generation efficiency; MUC1-specific CAR-T cells were also constructed to strengthen tumor antigen recognition. This combination strategy promoted efficient CAR-T infiltration into primary tumor lesions and induced long-lasting systemic antitumor immunity. This integrated PTT-CAR-T strategy overcomes the core shortcomings of conventional CAR-T therapy in oral squamous cell carcinoma and has considerable clinical translation potential. This dual synergistic therapeutic approach achieves better tumor lesion control and is expected to improve patient clinical outcomes.
High-Precision Intrinsic Interactome Elucidation of Chimeric Antigen Receptors via Photocatalytic Micromapping (μMap-CAR)
Journal & IF: J Am Chem Soc (16.6)
Corresponding author & institution: David W.C. MacMillan (Merck Center for Catalysis, Princeton University)
DOI: 10.1021/jacs.6c08969
Abstract: CAR-T cell therapy is a potent treatment for several hematological malignancies. However, its limited efficacy in more tumor types, especially solid tumors, highlights the need to optimize CAR molecular design to improve killing potency, persistence, and safety. The extensive research on the T-cell receptor (TCR) and its interactome has provided important theoretical bases for key CAR structural design. To further advance CAR-T therapy, more functionally relevant proteins need to be identified to support CAR engineering. Currently, the in situ molecular microenvironment around CAR molecules remains unclear, primarily due to the lack of analytical techniques combining high precision and high sensitivity. This study established the μMap-CAR high-resolution photocatalytic proximity labeling platform, which, after multiple methodological optimizations, can directly resolve the interactome of CARs on the surface of living T cells. The platform performed stably in CAR-T cell models under both resting and simulated activated states. μMap-CAR showed excellent sensitivity, resolving differential interactome changes after CAR intracellular domain modifications and directly linking signaling pathway perturbations to CAR structural elements. This study is the first to generate a high-resolution native CAR interactome map in primary T cells, identifying interactors common across different donors; candidate interactors were validated using super-resolution microscopy and CAR-T activation interference experiments. This platform is a versatile and powerful tool for CAR interactome analysis, while also providing actionable targets for proximity-based directed CAR engineering.
Dual-modular-nanobody CAR-T cell technical platform against the solid tumor microenvironment
Journal & IF: J Hematol Oncol (47.8)
Corresponding author & institution: Lu Xiaoling / Jiang Xiaobing / Yang Xiaomei (School of Basic Medicine, Guangxi Medical University / Department of Neurosurgery, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology / School of Basic Medicine, Guangxi Medical University)
DOI: 10.1186/s13045-026-01821-7
Abstract: Background: CAR-T therapy has shown remarkable efficacy in hematological malignancies, but faces two major challenges in solid tumors: tumor antigen heterogeneity and the immunosuppressive tumor microenvironment (TME). Systemic CTLA-4 blockade can enhance antitumor immunity but often causes severe adverse reactions. To overcome these barriers, this study constructed a nanobody-based dual-module CAR-T platform targeting fibroblast activation protein (FAP) on cancer-associated fibroblasts and releasing anti-CTLA-4 nanobodies locally in the tumor stroma. Methods: FAP/CTLA-4 dual-module CAR-T cells were constructed; antigen-specific cytotoxicity, cytokine secretion, and exhaustion levels were assessed in vitro; xenograft models were used to evaluate in vivo antitumor efficacy, with tumor growth, mouse survival, and T-cell tumor infiltration monitored (ethics approval: 202001011). One patient with R/R glioblastoma was enrolled and received intrathecal infusion of these CAR-T cells, with continuous monitoring of clinical efficacy, CSF cytokine levels (ethics approval: 2022-0553-01), and safety; transcriptomic sequencing and multiplex immunofluorescence staining were used to analyze changes in the tumor and immune microenvironment. Results: In vitro experiments confirmed that the engineered CAR-T cells had potent cytotoxicity, secreted large amounts of cytokines, and showed lower exhaustion levels. In in vivo animal models, the cells induced tumor regression, prolonged survival of tumor-bearing animals, and promoted T-cell infiltration into tumor tissue. In the glioblastoma patient receiving intrathecal administration, tumor status was stably controlled, CSF cytokine levels increased, and overall safety was favorable. Transcriptomic sequencing and multiplex immunofluorescence staining revealed that the tumor microenvironment was remodeled into an immune-activated phenotype. Conclusion: FAP/CTLA-4 dual-module nanobody CAR-T can relieve the immunosuppressive effects of the solid tumor microenvironment through local immunomodulation, demonstrating considerable therapeutic potential in both preclinical animal models and one patient with R/R glioblastoma.
Large language model-guided CAR-T in silico platform for cytokine optimization in liver cancer with low antigen density
Journal & IF: J Hematol Oncol (47.8)
Corresponding author & institution: Li Yanruide (University of California, Los Angeles)
DOI: 10.1186/s13045-026-01822-6
Abstract: CAR-T therapy has achieved outstanding efficacy in hematological malignancies, but its efficacy in solid tumors such as liver cancer is limited, primarily due to tumor antigen heterogeneity, low target antigen expression, and the immunosuppressive TME. Cytokine engineering can improve CAR-T persistence and effector function; however, the optimal cytokine to arm varies with tumor type, target antigen expression level, and microenvironment context, and systematically comparing them through animal and cellular experiments is time-consuming and labor-intensive. This study established an LLM-based CAR-T in silico simulation platform to systematically evaluate cytokine engineering strategies (IL-2, IL-7, IL-12, IL-15, IL-18) for GPC3-targeted CAR-T in liver cancer. Using cytokine screening as a biologically grounded benchmark experiment, the study validated whether the platform could reproduce established CAR-T cytokine biology and support novel target predictions. In silico simulations predicted IL-15 as having the strongest potentiating effect, particularly against GPC3-low tumor cells. Based on simulation results, cytokine-armed GPC3 CAR-T cells were constructed and validated in vitro and in vivo. IL-15-armed CAR-T showed superior proliferation, persistence, and serial killing activity against GPC3-low liver cancer cells. In human liver cancer xenograft models, IL-15-enhanced CAR-T achieved better tumor suppression than conventional CAR-T and other cytokine-armed CAR-T. The model successfully identified IL-15 as a known effective cytokine, validating the reliability of this LLM-guided CAR-T in silico analysis pipeline. In summary, this study establishes a biologically constrained, LLM-guided rational cytokine screening system for CAR-T engineering, and confirms IL-15 as a potent potentiating factor for targeting low-antigen-expressing liver cancer.
Circumventing Ewing sarcoma tumor microenvironment resistance by IL1RAP CAR-modified TGFβ1-imprinted natural killer cells in combination with IL-15 agonist and anti-GD2 antibody
Journal & IF: J Immunother Cancer (11.7)
Corresponding author & institution: Mitchell S Cairo / Wen Luo (New York Medical College)
DOI: 10.1136/jitc-2025-014633
Abstract: Background: Patients with metastatic, R/R Ewing sarcoma (ES) have an extremely poor prognosis. Natural killer (NK) cells have potent cytotoxicity against ES but are inhibited by TME-mediated resistance mechanisms. This study constructed a combination immunotherapy regimen attacking ES resistance from multiple dimensions: constructing CAR targeting the novel ES target IL1RAP to confer tumor-specific recognition to NK cells; TGFβ1 imprinting to counteract TGFβ-mediated immunosuppression; combining with the anti-GD2 antibody dinutuximab to enhance NK cell ADCC; and pairing with the IL-15 agonist NKTR-255 to prolong NK cell persistence and enhance ADCC. Methods: Antigen-presenting feeder cells co-expressing IL-21 and 4-1BB ligand were used to expand peripheral blood mononuclear cells to obtain conventional NK cells and TGFβ1-imprinted NK cells (imNK). mRNA encoding anti-IL1RAP CAR was electroporated into NK or imNK cells. In vitro cytotoxicity assays were performed with or without NKTR-255 and dinutuximab to evaluate killing activity of anti-IL1RAP CAR-NK/imNK cells against ES cells; ES xenograft mouse models were constructed for in vivo validation of the CAR-NK/imNK combination; single-cell RNA sequencing and CyTOF analysis of intratumoral cells were performed to elucidate the molecular mechanisms of response and resistance. Results: In vitro experiments confirmed that anti-IL1RAP CAR-NK cells significantly and specifically enhanced NK killing; in vivo models inhibited IL1RAP-positive ES tumor growth and lung metastasis. TGFβ1 imprinting significantly enhanced CAR-NK in vitro killing activity and tumor tissue infiltration, significantly reducing tumor burden and prolonging survival in orthotopic ES mouse models. Compared with single-agent or two-agent combinations, the triple combination of TGFβ1-imprinted CAR-NK (CAR-imNK), NKTR-255, and dinutuximab showed optimal antitumor efficacy against IL1RAP/GD2 double-positive ES. Single-cell mechanistic analysis of xenograft tumors revealed that after CAR-imNK + NKTR-255 + dinutuximab treatment, ES cell apoptosis increased, NK inhibitory receptor ligand expression on tumor cells was upregulated, and mouse macrophage recruitment to the ES microenvironment was enhanced. Conclusion: This preclinical study data confirm that an innate immunotherapy combination centered on tumor-targeting, TGFβ1-imprinted IL1RAP CAR-NK cells, combined with IL-15 agonist and anti-GD2 antibody, is a highly promising novel strategy for metastatic, R/R ES.
ENPP3 CAR T cells combined with CD206 modulation suppress adrenocortical carcinoma
Journal & IF: J Immunother Cancer (11.7)
Corresponding author & institution: Rosa Nguyen (National Cancer Institute, Bethesda, Maryland, USA)
DOI: 10.1136/jitc-2025-013726
Abstract: Background: Adrenocortical carcinoma (ACC) is a rare, aggressive malignancy with poor prognosis and limited curative treatment options. Although CAR-T cell therapy has shown promise in solid tumors, research on CAR-T in ACC is scarce. This study used patient-derived xenograft (PDX) models of ACC to screen for immunotherapy targets and develop novel CAR-T therapies. Methods: Target screening and TME profiling were performed using bulk RNA-seq and single-cell RNA-seq data from ACC patient samples in public databases; surface proteomic analysis and flow cytometry on PDX models validated candidate antigens. CAR-T targeting the antigen were constructed and evaluated for killing capacity in vitro and in vivo, with functional characterization by flow cytometry and cytokine detection. Results: This study identified ENPP3 as a shared immunotherapy target, expressed in 4 of 7 (57%) ACC PDX models. In vitro, ENPP3-targeted CAR-T cells eliminated >85% of ENPP3-positive ACC cells, but antitumor efficacy was markedly attenuated in PDX mouse models. ENPP3 CAR-T functional impairment was directly associated with immunosuppressive features of the TME, with CD206-positive tumor-associated macrophages (TAMs) as a key inhibitory factor. Combination with agents that modulate CD206-positive TAMs restored CAR-T function; the combination significantly enhanced antitumor efficacy in both ENPP3-high and ENPP3-low xenograft models (mean tumor weight difference −270.1 mg ±117.4, P<0.05). Conclusion: ENPP3 can serve as a novel target for CAR-T therapy in ACC. Combining ENPP3-targeted CAR-T with CD206-modulating agents relieves TME immunosuppression, potently inhibits tumor growth, and exerts significant anti-ACC effects. This synergistic strategy of CAR-T combined with modulation of immunosuppressive CD206-positive TAMs may also provide new insights for immunotherapy of other solid tumors beyond ACC.
Linperlisib enhances MUC1-Tn CAR T cell efficacy by inhibiting EGR1/DUSP2 axis to prevent CAR T cell exhaustion
Journal & IF: Leukemia (8.8)
Corresponding author & institution: Huang Yuxian (Zhujiang Hospital, Southern Medical University)
DOI: 10.1038/s41375-026-03021-1
Abstract: CAR-T therapy for T-ALL has suboptimal efficacy, primarily due to target-mediated fratricide caused by shared target antigen expression on tumor and normal T cells, and effector T-cell exhaustion. Mucin 1-Tn antigen (MUC1-Tn) is highly expressed in multiple malignancies including T-ALL. This study first validated MUC1-Tn protein expression in T-ALL cell lines and patient primary bone marrow cells, then constructed MUC1-Tn-targeted CAR-T cells. In vitro cytotoxicity assays and xenograft models confirmed that these CAR-T cells effectively lysed T-ALL tumor cells. Given the critical role of PI3Kδ signaling in regulating T-cell function and tumor immunosuppression, MUC1-Tn CAR-T cells were combined with the PI3Kδ inhibitor linperlisib. Linperlisib combination significantly enhanced MUC1-Tn CAR-T anti-leukemia activity and persistence, as evidenced by downregulation of exhaustion markers, reduced proportions of terminally differentiated cells, and sustained tumor control under secondary tumor challenge. Furthermore, linperlisib promoted CAR-T mitochondrial fusion and improved respiratory function. Mechanistically, linperlisib inhibited dual-specificity phosphatase 2 (DUSP2) and its upstream transcription factor early growth response protein 1 (EGR1), and this regulatory pathway is the core reason for extending CAR-T in vivo survival.
Humanized biparatopic nanobody-based CAR-T cells overcome antigen-heterogeneity in multiple myeloma
Journal & IF: J Transl Med (9.7)
Corresponding author & institution: Jincai Zhou / He Xiaowen (Origene Biotechnology)
DOI: 10.1186/s12967-026-08533-z
Abstract: Background: BCMA-targeted CAR-T therapy has antitumor activity in MM, but tumor BCMA expression heterogeneity and soluble BCMA (sBCMA) lead to frequent relapse. Optimizing BCMA-targeted CAR-T requires improving persistence, enhancing antitumor efficacy, tolerating low antigen expression, and resisting sBCMA-mediated inhibition. Methods: Anti-BCMA nanobodies (VHH) were screened from a humanized phage display library to construct monospecific and biparatopic CARs; patient-derived myeloma primary cells and xenograft tumors were used in preclinical models to evaluate function and select the optimal construct. Results: In vitro, biparatopic Nab5822 CAR-T cells demonstrated superior killing against antigen-heterogeneous myeloma cells, with higher levels of IL-2, IFNγ, TNFα, and other cytokines.