Multiple myeloma (MM) is the second most prevalent hematological malignancy (1). Despite significant improvement in treatment outcomes due to novel agents, such as immunomodulatory drugs, protease inhibitors and anti-CD38 monoclonal antibodies, the prognosis remains grim for refractory/relapsed multiple myeloma (RRMM) (2, 3). Currently, anti-BCMA chimeric antigen receptor (CAR)-T cell therapies have yielded significant clinical outcomes in RRMM patients, with anti-BCMA CAR-T cell products approved by the US FDA (idecabtagene vicleucel in March 2021 and ciltacabtagene autoleucel in February 2022 (4, 5)), that Phase II clinical trials demonstrated overall response rates (ORR) of 70%-95% (6, 7). However, due to the high relapse rate of post-CAR-T cell therapies, largely attributed to antigen escape (8), there is a need for more effective CAR-T-based regimens.

CD19, which is expressed on multiple differentiated MM cells and myeloma-like stem cells, is associated with enhanced MM drug resistance and tumor-propagation (9). Previous studies have reported the synergistic effects of CD19-targeted therapies with other MM therapies (10). The combination of anti-CD19 CAR-T cell therapy and autologous stem cell transplantation has shown significant clinical outcomes in RRMM patients (11). Additionally, the ultra-low expression of CD19 in MM cells can trigger elimination by anti-CD19 CAR-T cell therapies (12), establishing CD19 as a crucial biomarker of RRMM and a therapeutic target of CAR-T cell therapies. Multiple studies utilizing anti-BCMA and anti-CD19 CAR-T infusion have obtained significant clinical outcomes in proof-of-concept trials for RRMM patients (13–15).

This meta-analysis discusses the clinical efficacy and safety of combined anti-BCMA and anti-CD19 CAR-T cell therapies in RRMM. The strengths of this study include the large sample size, comprehensive range of clinical trials, valuable evaluation index system and in-depth subgroup analyses, collectively facilitating the application of combined infusion with CAR-T cell therapies.

Modern advancements in MM treatment have progressed from conventional chemotherapies to more specific immune-based treatments. As a state-of-the-art method, anti-BCMA CAR-T cell therapies have attained remarkable success in RRMM. BCMA, mainly expressed in malignant plasma cells and multiple B cells, is viewed as an eligible target for RRMM CAR-T cell therapies (27). Previous studies have demonstrated that over 81% RRMM respond to anti-BCMA CAR-T cell therapies, however, most responsive patients tend to relapse eventually (6, 7, 27). Bruno et al. found that BCMA expression varied greatly among different MM patients, and that nonresponsive MM patients had plasma cells with lower expression of BCMA (28). The downregulation of BCMA after infusion of anti-BCMA CAR-T cell therapies, aligning with the onset of MM disease progression, was also observed (28, 29). These findings demonstrate that BCMA escapes the existence and insufficiency of solo antigen-targeting CAR-T cell therapy in MM treatments. The role of CD19 in B-cell lineage differentiation is significant, and its expression generally decreases in MM cells (30). Previous studies have demonstrated that a small population of less-terminally differentiated CD19⁺ plasma cells may construct a drug-resistant, clonogenic disease reservoir that is maintained by components of the bone marrow microenvironment (31, 32). These lead to consider CD19 as a promising target alongside anti-BCMA CAR-T cell therapies. This meta-analysis of 12 clinical trials, inclusive of 347 participants, indicates that combined anti-BCMA and anti-CD19 CAR-T cell therapy deliver remarkable benefits with a controllable safety profile in RRMM.

This meta-analysis, focusing on the efficacy and safety of combined anti-BCMA and anti-CD19 CAR-T cell therapy in RRMM, highlights that all clinical trials show substantial effectiveness. The pooled ORR was 94%, with the highest response rate of 100% reported by Xiaolan Shi et al. (23–25), and other results hovered around 90%, demonstrating the considerable effectiveness of the combined CAR-T cell therapy in RRMM. ORR only reflects the quantitative indicators of therapeutic response, while PR, VGPR, CR, sCR and other indicators can reflect the quality of remission. Therefore, we further analyzed the incidence of the therapeutic effect of CR+sCR and identified that the CRR was 50%. Due to the low rate of CR obtained by conventional treatments, even autologous transplantation could only achieve a CR rate of 40% (33), hence these results suggest significant efficacy in MM treatment using combined anti-BCMA and anti-CD19 CAR-T cell therapy.

In terms of effectiveness, compared with single-target CAR-T therapies, the combined anti-BCMA and anti-CD19 CAR-T therapy showed better efficacy. The reported ORRs were 77% by Yang Q et al. (34), 85.2% by Zhang L et al. (35), and 87% by Hu D et al. (36), which are all lower than our result (94%). Moreover, the reported CRRs were 37% by Yang Q et al. (34), 47% by Zhang L et al. (35), and 44% by Hu D et al. (36), which are all lower than our result (50%).

Despite the excellent efficacy, toxicities of CAR-T cell therapy in RRMM after treatment are also important. CRS, a systemic inflammatory response, is the most common AE following CAR-T infusion (37). It can result in clinical symptoms such as neurotoxicity, fever and hypotension (38). In this study, the overall incidence of any grade CRS was 98%, primarily of CRS grade 1 or 2. In addition, neurotoxicity is also a common adverse reaction. Neurotoxicity is a complex syndrome, including encephalopathy, cognitive defect, dysphasia, seizure and cerebral edema (39), which is thought to be one of the primary challenges to the general application of CAR-T cell therapy (40). The overall incidence of grade 3 or 4 neurotoxicity was low (8%). Generally, considering the high efficiency of combined anti-BCMA and anti-CD19 CAR-T cell therapies, the safety profile was manageable and tolerable. In the future, a more mechanistic understanding of AE is important to raise the efficacy-to-toxicity ratios of combined CAR-T cell therapies.

In terms of safety, the AEs of CAR-T cell infusion mostly presented as CRS and neurotoxicity (41). The reported incidences of any grade CRS were 82% by Hu D et al. (36) and 80.3% by Roex G et al. (5), which is lower than our result (98%). Otherwise, compared to single-target CAR-T therapy, there was little difference in the incidence of other AEs. The reported incidences of grade 3 or higher CRS were 14% by Yang Q et al. (34), 11% by Hu D et al. (36), 14.1% by Roex G et al. (35), which is all higher than our result (9%). The reported incidences of neurotoxicity were 10% by Hu D et al. (36), 10.5% by Roex G et al. (5) and 8% by Xiang, X (42), which is slightly higher than our result (8%).

The results of subgroup analysis provide valuable insights for the selection of RRMM patients suitable for the combined anti-BCMA and anti-CD19 CAR-T cell therapy. Subgroup analysis of ORR by characteristics of the included participants showed that MM patients with a better disease status tended to acquire better efficacy. Specifically, patients under 55 years old can obtain a higher ORR compared to those aged 55 years and above. Similarly, a greater ORR was observed in patients who had received ≥50% of prior ASCT, compared to those who received less than 50%. This could be attributed to the sequential treatment of ASCT followed by CAR-T therapy, as reported by Xiaolan Shi et al. (23–25), presumably placing patients in a good basic condition when starting the CAR-T treatment, thus enhancing the remission effect. Also, compared to the proportion with extramedullary disease ≥20%, a higher ORR was observed with the proportion <20%, probably because myeloma cells proliferate and accumulate in other organs, leading to a worse survival state and a poorer response to the therapy. Another insightful subgroup analysis was performed based on the percentage of ISS III in patients, showing a greater ORR for ISS III <50% as opposed to ISS III ≥50%.

The variations in costimulatory molecules may also have an impact on safety which is reflected in the incidence of CRS. The CRS subgroup analysis revealed that CAR-T cell therapy with 4-1BB in the CAR construct attained a lower risk of CRS than other costimulatory molecules. The second-generation CAR-T is created by combining the cytoplasmic domain of the CD3ξ chain with the signaling domains of co-stimulatory receptors, such as CD28 and 4-1BB, to give T cells a high potential for proliferation, activity, and persistence (43, 44). It has been established that costimulatory molecules play a crucial role in affecting T cell activation, proliferation, survival and cytokine release which regulate tumor immunity (45). Studies have shown that in preclinical studies, CARs with 4-1BB costimulatory domains typically lead to a relatively gradual and long-lasting response (46) and produce less cytokine release compared (47) with CD28, which may explain the smaller incidence of CRS in 4-1BB than other costimulatory molecules in our subgroup analysis. We also performed a subgroup analysis of ORR based on the antigen recognition domain origin of the CAR. The findings revealed that the effectiveness was similar in the Human group and Murine group, with no statistically significant difference, indicating that the species origin of the CAR antigen-recognition domain is not a major determinant of ORR of dual-target CAR-T therapy. However, Hu D et al. (2023) reported that humanized CAR-T cells were superior to those produced from murine as there are inherent limitations in the use of murine scFv-based CARs (36). Consequently, additional clinical studies are required to address this controversy.

In the subgroup analysis of conditioning treatment, nine of the trials were treated with Flu + Cy and three trials were treated with Busulfan + Cy. The results showed better efficacy in the Busulfan + Cy group. This may be due to Busulfan + Cy not only having myelosuppressive effects but also being anti-tumor to some extent (48), so using this as conditioning treatment before combined CAR-T cells infusion may have better efficacy. To explore whether infusion time would affect the results, we conducted a subgroup study. Among the 12 trials, except for one that did not mention the infusion time, five trials received BCMA CAR-T cells and CD19 CAR-T cells on the same day (three trials were infused in one day, and two trials were infused in three days). The remaining six trials received CD19 CAR-T cells on Day 0, 40% BCMA CAR-T cells given on Day 1, and 60% BCMA CAR-T cells given on Day 2. Although CAR-T cells were not infused on the same day, they all completed the infusion within a short period. The results showed that the difference caused by infusion time was not statistically significant.

In summary, although considerable toxicity was observed, the combined anti-BCMA and anti-CD19 CAR-T therapy has a significant effect on the treatment for RRMM. Subgroup analysis revealed possible factors influencing efficacy and safety. These findings may inform the development of next-generation combined CAR-T cell therapy and the optimization of their clinical applications. Further, more well-designed clinical trials with large sample sizes are required to establish the role of combined CAR-T therapy in RRMM patients.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.

HX: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing. CG: Conceptualization, Project administration, Resources, Supervision, Writing – review & editing. PX: Conceptualization, Data curation, Software, Writing – review & editing. DZ: Project administration, Resources, Writing – review & editing. YX: Funding acquisition, Writing – review & editing. BC: Funding acquisition, Software, Writing – review & editing. HB: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing.