Clinical Deep-Dive · Hematology

CAR-T Therapy in DLBCL and Multiple Myeloma: Outcomes and Optimization

August 18, 2026

CARTITUDE-4 cilta-cel cohort

208 patients

CARTITUDE-4 median follow-up

33.6 months

Flatiron real-world cohort

932 patients (1445 eligible LOT)

Flatiron median follow-up

23.6 months

CARTITUDE-4 trial identifier

NCT04181827

Clinical Perspective

The available sources indicate relevance for hematologists, hematologic oncologists, and clinicians treating relapsed or refractory diffuse large B-cell lymphoma (DLBCL) and relapsed or refractory multiple myeloma (RRMM) because CAR-T therapy has changed the treatment landscape for DLBCL and cilta-cel is an approved therapy in RRMM for patients after prior lines of therapy ^[2]^[3]. The source-confirmed data and methodological descriptions support the use of careful comparative frameworks when interpreting single-arm CAR-T trial results in the context of heterogeneous real-world treatment patterns, and they support use of matching and inverse probability weighting approaches when constructing indirect comparisons between CAR-T trial cohorts and real-world physician's choice cohorts ^[1]^[3]. What remains unanswered is how best to standardize patient selection criteria, workflow steps, and operational pathways across centres to optimize both initial response rates and long-term disease control after CAR-T infusion, and how emerging in vivo CAR-T technologies will be integrated into existing treatment algorithms?

Dr. Shruti Pandey, MBBS, MD (General Medicine), DM (Clinical Haematology) · Hematology

Chimeric antigen receptor T-cell (CAR-T) therapy has transformed the treatment of relapsed or refractory diffuse large B-cell lymphoma (DLBCL), yet a substantial proportion of patients fail to achieve durable remission ^[2]. Registrational trials for ciltacabtagene autoleucel (cilta-cel) and idecabtagene vicleucel (ide-cel) were single-arm studies conducted in relapsed refractory multiple myeloma (MM) patients who were triple-class-exposed or triple-class-refractory ^[1]. In the CARTITUDE-4 study, cilta-cel demonstrated superior efficacy versus pomalidomide, bortezomib, and dexamethasone or daratumumab, pomalidomide, and dexamethasone in patients with relapsed or refractory multiple myeloma after 1-3 prior lines of therapy; the CARTITUDE-4 cohort included data from 208 patients who received cilta-cel with median follow-up 33.6 months, and a Flatiron real-world cohort included 932 patients (1445 eligible lines of therapy) with median follow-up 23.6 months ^[3].

Clinical Context

Chimeric antigen receptor T-cell (CAR-T) therapy has been described as transformative for relapsed or refractory diffuse large B-cell lymphoma (DLBCL), reflecting substantive changes in the therapeutic landscape for that indication, while also presenting ongoing outcome challenges because a substantial proportion of patients do not achieve durable remission ^[2]. The limitations to durable disease control in DLBCL treated with CAR-T are attributed in the literature to both inadequate initial response and to subsequent disease relapse, indicating two distinct phases where outcomes can be lost ^[2].

For relapsed or refractory multiple myeloma (RRMM), the available registrational evidence for two named CAR-T products—ciltacabtagene autoleucel (cilta-cel) and idecabtagene vicleucel (ide-cel)—originated from single-arm trials enrolling patients who were triple-class-exposed (TCE) or triple-class-refractory (TCR) ^[1]. Comparative effectiveness research researchers have emphasised that appropriate choice of data sources and careful matching of patient populations are critical when placing results from these single-arm CAR-T trials in comparative context, because direct randomized comparisons are lacking for these registrational programs ^[1].

A review of available trials and real-world cohorts identified nine clinical trials of NCCN-preferred regimens and five real-world data publications in previously treated MM populations, and found that none of the clinical trials specifically evaluated patients with TCE or TCR MM; among the real-world publications, cohorts varied, with two evaluating exclusively TCR MM, two analyzing mixed TCE/TCR populations, and one assessing exclusively TCE MM, and the real-world treatment patterns were heterogeneous ^[1]. This heterogeneity in real-world treatment patterns and the absence of randomized trial data in some of the key heavily pretreated MM populations underlines the methodological challenges when attempting cross-study or trial-to-real-world comparisons.

The CARTITUDE-4 study offers a contemporary example of how these issues are being addressed methodologically: CARTITUDE-4 results were compared with a Flatiron Health MM cohort using matched eligibility criteria and statistical adjustment approaches to attempt comparative effectiveness assessment between a clinical-trial CAR-T cohort and real-world physician's choice treatment ^[3]. The CARTITUDE-4 data cutoff and the real-world data window used for comparison are explicitly reported, and the CARTITUDE-4 cohort and Flatiron cohort sizes and follow-up durations are provided in the source material ^[3].

Information on population-level burden, incidence, prevalence, and economic impact for DLBCL or MM were not available in the provided source excerpts. Information on detailed contemporary standard-of-care sequencing and guideline-based algorithms across the full spectrum of lines of therapy was not available in the provided source excerpts.

Key Findings

Ciltacabtagene autoleucel (cilta-cel) is described in the available source as an approved therapy for relapsed or refractory multiple myeloma (RRMM) and was evaluated in CARTITUDE-4, a trial in patients with RRMM after 1-3 prior lines of therapy; in CARTITUDE-4, cilta-cel demonstrated superior efficacy versus either pomalidomide, bortezomib, and dexamethasone or daratumumab, pomalidomide, and dexamethasone ^[3]. To place CARTITUDE-4 outcomes in a comparative context with routine clinical practice, an indirect treatment comparison was conducted using de-identified data from the Flatiron Health multiple myeloma cohort registry covering January 2020 to May 2024 and CARTITUDE-4 data with a data cutoff of May 1, 2024 ^[3].

The indirect comparative methods included matching patients from the Flatiron database to CARTITUDE-4 key eligibility criteria and adjusting baseline covariates of prognostic significance using inverse probability of treatment weighting; outcomes specified for comparative effectiveness included progression-free survival (PFS), real-world PFS (RW PFS), time to next treatment (TTNT), and overall survival (OS), and sensitivity analyses were conducted to examine the robustness of findings ^[3]. These methods reflect a deliberate approach to reduce confounding when contrasting a clinical-trial CAR-T cohort with heterogeneous real-world treatment cohorts.

Cohort sizes and follow-up framed the comparative dataset: the CARTITUDE-4 cohort included data from 208 patients who received cilta-cel with median follow-up 33.6 months, whereas the Flatiron real-world cohort comprised 932 patients (1445 eligible lines of therapy) with median follow-up 23.6 months ^[3]. These explicit patient-count and follow-up metrics are reported in the source and underpin the comparative analyses described.

Context from earlier methodological reviews and research synthesis underscores why such trial-to-real-world comparisons require careful design: registrational CAR-T trials such as CARTITUDE-1 (cilta-cel) and KarMMa (ide-cel) were single-arm trials enrolling TCE or TCR MM patients, and a systematic search identified that existing NCCN-preferred regimen trials did not specifically study TCE/TCR populations, while real-world cohorts were heterogeneous, reinforcing the need for careful matching and data-source selection in comparative effectiveness work ^[1].

For diffuse large B-cell lymphoma (DLBCL), literature syntheses focused on CAR-T optimisation emphasize the importance of interventions across the therapeutic continuum, noting that limitations in durable disease control arise from inadequate initial responses and later relapse, and that strategies to improve outcomes may operate before leukapheresis, during manufacturing, prior to infusion, during in vivo expansion, and after remission, with in vivo CAR-T technologies highlighted as an emerging platform to expand access and potentially alter outcomes ^[2]. The review also highlights implementation challenges and evidence gaps across these strategy domains and calls for integration of evidence to create a more unified framework for clinical application ^[2].

Overall, the combined sources describe (1) the transformative but incomplete impact of CAR-T in DLBCL, (2) the single-arm registrational evidence base for key CAR-T products in heavily pretreated MM, and (3) contemporary efforts to use real-world data and statistical adjustment to contextualize CAR-T trial outcomes against physician's choice therapies in routine practice ^[1]^[2]^[3]. Specific comparative efficacy effect sizes or numerical hazard ratios beyond the cohort counts and median follow-up durations cited above were not reported in the provided source excerpts.

Safety & Tolerability

Safety information was not available in the provided source excerpts.

Study Design

The CARTITUDE-4 comparative dataset and methods are described in the source as follows: CARTITUDE-4 (NCT04181827) evaluated ciltacabtagene autoleucel (cilta-cel) in patients with relapsed or refractory multiple myeloma after 1-3 prior lines of therapy, and the study demonstrated superior efficacy versus pomalidomide, bortezomib, and dexamethasone or daratumumab, pomalidomide, and dexamethasone ^[3]. For the indirect comparative effectiveness analysis described in the source, de-identified patient-level data from the Flatiron Health multiple myeloma cohort registry covering January 2020 to May 2024 were used and compared with CARTITUDE-4 data with a data cutoff of May 1, 2024; key eligibility criteria from CARTITUDE-4 were applied to select comparable patients in the Flatiron dataset ^[3].

Baseline covariates of prognostic significance were adjusted using inverse probability of treatment weighting to balance the trial and real-world cohorts prior to outcome comparisons, and the comparative effectiveness outcomes explicitly included progression-free survival (PFS), real-world PFS (RW PFS), time to next treatment (TTNT), and overall survival (OS), with sensitivity analyses performed to explore robustness of the findings ^[3]. The CARTITUDE-4 cohort included data from 208 patients who received cilta-cel with median follow-up 33.6 months, and the Flatiron real-world cohort included 932 patients (1445 eligible lines of therapy) with median follow-up 23.6 months ^[3].

Separately, methodological analysis of the registrational CAR-T evidence base noted that registrational trials for cilta-cel and ide-cel were single-arm studies enrolling relapsed refractory MM patients who were triple-class-exposed or triple-class-refractory, and that a systematic review identified nine clinical trials of NCCN-preferred regimens and five real-world data publications but found no clinical trials that specifically evaluated TCE or TCR MM populations, highlighting the implications for comparative effectiveness research ^[1].

[Detail] was not available in the provided source excerpts for the following methodological items: specific statistical model parameters beyond inverse probability weighting, full inclusion/exclusion criteria text beyond the statement that CARTITUDE-4 eligibility criteria were used for matching, randomization details for comparator arms (if any), prespecified primary endpoint effect sizes or hazard ratios from CARTITUDE-4 as presented in the trial report, and funding or sponsor declarations in full.

FAQ

Q1 What is ciltacabtagene autoleucel (cilta-cel) approved for?

Ciltacabtagene autoleucel (cilta-cel) is described in the provided source as approved for relapsed or refractory multiple myeloma (RRMM), and CARTITUDE-4 evaluated cilta-cel in patients with RRMM after 1-3 prior lines of therapy where it demonstrated superior efficacy versus pomalidomide, bortezomib, and dexamethasone or daratumumab, pomalidomide, and dexamethasone. Information on the precise approved label wording, age restrictions, or exact approval date was not available in the provided source excerpts.

Q2 How does CAR-T therapy work?

Information on the mechanistic details of how CAR-T therapy works beyond the designation 'chimeric antigen receptor T-cell (CAR-T) therapy' was not available in the provided source excerpts; the sources describe CAR-T as a therapeutic modality that has transformed treatment for relapsed or refractory DLBCL and as the subject of optimization strategies at multiple points across the therapeutic continuum but do not provide a mechanistic stepwise description in the provided excerpts.

Q3 What is the recommended dose of cilta-cel?

Dosing information was not available in the provided source excerpts.

Q4 What are the side effects of CAR-T or cilta-cel?

Safety information was not available in the provided source excerpts.

CAR-T Therapy in DLBCL and Multiple Myeloma: Outcomes and Optimization

Direct Clinical Takeaway & Evidence Summary

CAR-T Therapy in DLBCL and Multiple Myeloma: Outcomes and Optimization

Clinical Perspective

Clinical Context

Key Findings

Safety & Tolerability

Study Design

FAQ

References

Get weekly clinical updates for Hematology in your inbox.

Related Clinical Updates

FDA Approves Acalabrutinib + Venetoclax for CLL/SLL: PFS HR 0.65 in AMPLIFY

AURORA: Bitopertin Reduced Protoporphyrin‑IX 49.8% in Erythropoietic Protoporphyria

narsoplimab: FDA Approves for TA-TMA; 61% TMA Response

Verified HCP Portal