Efficacy and safety of BCMA nanobody CAR T-cell therapy in relapsed or refractory plasma cell myeloma Open Access

Relapsed or refractory (R/R) multiple myeloma (MM) remains a significant clinical challenge, often yielding poor outcomes despite chemotherapy or transplantation.¹ B-cell maturation antigen (BCMA), belonging to the tumor necrosis factor superfamily, is primarily expressed by malignant and normal plasma cells (nPCs), rendering it a potential therapeutic target for MM.² BCMA-targeted chimeric antigen receptor (CAR) T-cell therapy has emerged as a promising therapeutic option, demonstrating notable efficacy in the management of R/R MM.^3,4 Recently approved BCMA-targeting CAR T-cell therapies by the US Food and Drug Administration, idecabtagene vicleucel and ciltacabtagene autoleucel (cilta-cel), have shown promise in the treatment of heavily pretreated R/R MM.^5,6 Early studies reported overall response rates (ORRs) ranging from ∼60% to 80%. However, recent advancements in BCMA CAR T-cell design have markedly improved therapeutic outcomes.^7-12 For instance, dual epitope–binding BCMA CAR T-cell constructs, such as LCAR-B38M, which exhibit high-avidity antigen binding, have been shown to elicit strong clinical responses at lower cell doses.^4,13 

Classical CARs comprise 4 components: an extracellular antigen-recognition domain, a transmembrane domain, an intracellular costimulatory domain like 4-1BB and/or CD28, along with a CD3ζ signaling domain.¹⁴ CAR T cells typically use the single-chain variable fragments (scFvs) of a monoclonal antibody as their antigen-recognition domain. Nanobodies, also known as the variable domain of the heavy chain of heavy chain antibody (VHH), are derived from heavy chain-only antibodies found in animals such as those from the Camelidae family and sharks.^15,16 Nanobodies offer a small, stable, single-domain structure with high affinity and specificity comparable to scFvs.¹⁷ Notably, nanobodies exhibit reduced immunogenicity compared with mouse-derived monoclonal antibodies, making them potentially safer for use in CAR T-cell therapy.^18,19 

This study evaluates the efficacy and safety of a nanobody-based anti-BCMA CAR T-cell therapy (S103 CAR T cell) in patients with R/R plasma cell myeloma.

The antigen-binding domain of S103 CAR T-cell therapy is composed of 2 nanobodies, which are connected by short peptide linkers. These nanobodies are integrated into a second-generation lentiviral vector that includes a CD8α hinge and transmembrane region, a 4-1BB costimulatory domain, and a CD3ζ activation domain. The resulting recombinant lentiviral vector is referred to as the S103 CAR plasmid.

Peripheral blood mononuclear cells were obtained from healthy patients with R/R MM. CD3⁺ T cells were subsequently isolated using CD3 MicroBeads (Miltenyi Biotec, 200-070-202), and cultured in TexMACS GMP medium (Miltenyi Biotec, 170-076-306). Activation was achieved through the addition of Dynabeads Human T-Activator CD3/CD28 (Thermo Fisher, 40203D). Lentiviral transduction was performed after 2 days of culture and activation, and the CAR T cells were further cultured for 12 to 14 days before harvest. Transduction efficiency of the CAR T cells was assessed throughout the culture process using flow cytometry. For detection, BCMA-His protein served as the primary antibody, whereas an anti-His–labeled antibody (Abcam, ab1206) was utilized as the secondary antibody.

To evaluate the biological activity of S103 CAR T cells in vitro, a cytotoxicity assay was performed. S103 CAR T cells were cocultured with the NCI-H929-Luc cell line (BeNa Culture Collection) at an effector-to-target cell ratio of 3:1. Following overnight incubation at 37°C with 5% CO₂, the luciferase activity of the target cells was measured to assess their lysis efficiency.

Eighteen NPG mice (6 per group, balanced by sex) were injected with 1 × 10⁷ MM.1s-Luc cells per mouse. After 8 days, all mice received a 250 μL injection, and were divided into 3 experimental groups: the CAR T-cell group (0.5 × 10⁷ CAR⁺ T cells per mouse), the T-cell group (the same total number of nonmodified T cells), and the solvent control group (receiving cell preservation solution). Weekly immunoimaging assessments were conducted after injection to evaluate the in vivo antitumor efficacy of S103 CAR T cells.

After thorough validation in the mouse model, we initiated the clinical study with the following specific protocol: autologous peripheral blood lymphocytes were collected from patients, who underwent lymphodepletion chemotherapy with IV fludarabine (30 mg/m² per day) and cyclophosphamide (FC; 300 mg/m² per day) from day −5 to day −3. The median administered dose of S103 CAR T cells was 1 × 10⁶/kg (range, 0.3× 10⁶ to 2 × 10⁶/kg). Efficacy evaluations were conducted before CAR T-cell infusion, during the first, second, and third months, and every 3 months thereafter. Assessments included immunoglobulin electrophoresis, free light chain (FLC) analysis in blood and urine, bone marrow aspiration and biopsy (including minimal/measurable residual disease [MRD] detection by flow cytometry), as well as whole-body positron emission tomography (PET)–computed tomography (CT) or localized CT imaging at 3-month intervals.

MRD and BCMA expression in MM were assessed using multiparametric flow cytometry with a 3-tube, 3-laser, 8-color panel: tube 1: cκ-FITC/cλ-PE/CD38-PerCP/CD19-PE-Cy7/CD138-APC/CD20-APC-Cy7/CD56-BV421/CD45-V500; tube 2: CD38-FITC/CD229-PE/CD27-PerCP/CD19-PE-Cy7/CD138-APC/CD20-APC-Cy7/CD56-BV421/CD45-V500; tube 3: Ki67-FITC/CD269(BCMA)-PE/CD38-PerCP/CD19-PE-Cy7/CD138-APC/CD56-BV421/CD45-V500. BCMA-PE (clone 19F2) was sourced from BioLegend (San Diego, CA). Polyclonal rabbit anti-human κ-fluorescein isothiocyanate (FITC) and λP-phycoerythrin (λ-PE) light chain antibodies were obtained from Dako (Copenhagen, Denmark), The cell cytoplasm κ-FITC and λ-PE are monoclonal rabbit anti-human κ-FITC or λ-PE light chain antibodies (in F[abʹ]2 form) produced by Dako, Denmark, whereas other monoclonal antibodies, lysing solution, and permeabilization reagents were supplied by BD (Cayey, Puerto Rico).^20,21 

BCMA expression on normal plasma cells (nPC) was measured using bone marrow samples from 20 patients without MM in complete response (CR). The relative fluorescence intensity (median fluorescence intensity [MdFI]) of BCMA was analyzed as the ratio of BCMA MdFI in plasma cells to that in non-B lymphocytes. A reference range of 2.7 to 7.2 was established based on the mean ±2 standard deviation from these 20 patients with nPCs.

The protocols of the clinical trial and the retrospective study were approved by the ethics committee of Hebei Yanda Lu Daopei Hospital and Beijing Lu Daopei Hospital. The studies were conducted in compliance with the Declaration of Helsinki. All patients or their guardians provided written informed consent in the clinical trials.

The S103 CAR was engineered with a dual-nanobody tandem as the antigen-recognition domain, and the 2 nanobodies have different binding epitopes (supplemental Figures 1 and 2), fused with the hinge and transmembrane regions from CD8α, the 4-1BB costimulatory domain, and the CD3ζ activation domain, forming the complete CAR structure sequence (Figure 1A). This construct was driven by the EF1α promoter to ensure robust expression.

To evaluate the performance of S103 CAR T cells, we analyzed their transduction efficiency and cell viability. The transduction efficiency ranged from 45.3% to 70.2%, with an average efficiency of 53.35% (Figure 1B). Cell viability was consistently high, with an average viability of 94.15% before harvest, comparable to that of mock T cells (94.03% on average; Figure 1C).

The in vitro cytotoxicity of S103 CAR T cells was assessed using the BCMA⁺ NCI-H929-LUC cell line, which exhibits luciferase activity. The cell lysis rates were 55.5% ± 18.5%, 86.6% ± 12.3%, and 97.1% ± 2.7% at effector-to-target ratios of 1:1, 3:1, and 10:1, respectively, significantly exceeding those of the control group. These findings demonstrate that S103 CAR T cells exhibit specific cytotoxic activity and high biological efficacy against NCI-H929-LUC cells in a dose-dependent manner (Figure 1D).

Figure 1E illustrates the establishment of the S103 CAR T-cell animal model, the drug administration approach, and the testing protocol. The in vivo pharmacodynamics were evaluated using a transplanted tumor model generated by introducing MM.1S-LUC cells into NPG mice. The administration of CAR T cells was performed via a single intravenous injection, mimicking clinical treatment protocols. Results indicated that, starting from day 8 after administration, the CAR T-cell group achieved complete tumor clearance, with no tumor recurrence observed during the study period. These findings demonstrate that S103 CAR T cells effectively suppress tumor proliferation in vivo.

Between 1 May 2021 and 31 December 2023, a total of 27 patients with R/R MM were treated with S103-BCMA CAR T-cell therapy at our institution. Follow-up data were collected up to 30 June 2024. The cohort consisted of 13 males and 14 females, with a median age of 58 years (range, 30-73). The median time from initial MM diagnosis to relapse was 34 months (range, 10-108), and the median number of previous lines of therapy was 5 (range, 1-10). Among the patients, 4 were diagnosed with plasma cell leukemia, and 11 exhibited extramedullary disease (EMD), including 3 cases of paralysis caused by central nervous system (CNS) infiltration. Of the 27 patients, 14 had undergone previous transplantation, including 13 who relapsed after autologous transplantation, and 1 who relapsed after allogeneic transplantation. One patient presented with a rare anaplastic plasma cell tumor. In addition, 11 patients had high-risk features, such as t(4;14)(p16;q32), t(14;16)(q32;q23) chromosomal translocation as identified through fluorescence in situ hybridization analysis and/or TP53 mutations, as shown in Table 1 and supplemental Table 1.

The median BCMA expression rate in the myeloma cells of the 27 patients was 53.3% (range, 18.5%-100%), although 2 patients displayed no detectable BCMA expression.

One month after BCMA CAR T-cell infusion, the ORR was 96.3% (26/27), with a complete response (CR) and very good partial response (VGPR) rate of 59.2% (16/27). The single nonresponding patient achieved a partial response (PR) following a second infusion of BCMA CAR T cells. One patient succumbed to a lung infection ∼2.5 months after infusion. At the 3-month evaluation, the best ORR reached 100% (27/27), with a CR + VGPR rate of 81.5% (22/27), although 3 patients exhibited disease progression (PD).

Long-term follow-up revealed a median duration of remission of 11 months (range, 2-36). The 1-year overall survival (OS) and progression-free survival (PFS) rates were 61.1% and 57.2%, respectively, whereas the 2-year OS and PFS rates were 42.3% and 42.4% (Figure 2). Two patients relapsed at 11 and 14 months after CAR T-cell therapy, with the former achieving CR again after GPRC5D (G protein-coupled receptor, family C, group 5, member D) CAR T-cell therapy, and the latter achieving CR again after selinexor. Among the 11 deaths, 4 were attributed to infections during disease control, and 7 were due to PD (Figure 3; supplemental Table 1).

Notably, all high-risk cases involving chromosomal abnormalities [t(14;16), t(4;14), t(14;20)] and/or TP53 mutations (n = 11) showed effective responses with 72.7% (8/11) CR within 3 months after BCMA CAR T-cell infusion, 1 year OS of 62.3%, and 1 year PFS of 65.5% (Table 1).

In the plasma cell leukemia subgroup (n = 4), all patients achieved an ORR within 1 month. Three achieved CR + VGPR, whereas 1 experienced PD within 3 months. Of the 3 patients with CR + VGPR, 1 underwent allogeneic hematopoietic stem cell transplantation and remains disease-free after 1.5 years. Another patient died of infection at 3 months, and 1 experienced recurrence after 14 months (Table 1).

Among the 11 patients with EMD, the ORR at 1 month was 100%. At 3 months, 9 patients remained responsive, including 5 with CR + VGPR. One patient died of infection within 3 months, and 3 ultimately experienced PD. The 1-year OS rate was 42.4%, and 1-year PFS rate was of 49.9% (Table 1). In the subgroup with CNS infiltration (n = 3), all patients initially presenting with paralysis achieved PR at the 1-month evaluation. By the 3-month assessment, 2 had achieved CR, whereas 1 had progressed.

There was no statistical difference between patients with or without extramedullary infiltration, patients with myeloma or plasma cell leukemia, and patients with or without high-risk chromosomal abnormalities (Table 1).

The 2 patients with undetectable BCMA expression by flow cytometry exhibited durable responses, both achieving CR by the 3-month evaluation.

Among the treated patients, 4 experienced grade 2 cytokine release syndrome (CRS), and 1 patient developed grade 3 CRS. In addition, grade 1 neurotoxicity was observed in a single patient. Within 3 months after infusion of BCMA CAR T cells, 26 of 27 patients experienced febrile episodes. Other commonly reported adverse events included vomiting, abdominal pain, and fatigue.

Infections were a notable complication following CAR T-cell therapy, with 4 cases of viral infections, 7 cases of bacterial infections, and 5 cases of fungal infections. After corresponding antibiotic, antiviral, and antifungal treatments, as well as immunoglobulin infusion to supplement immunoglobulin deficiency caused by CAR T-cell therapy clearing plasma cells, most patients’ infections can be controlled. Unfortunately, 4 patients succumbed to infection-related complications. Among them, 1 case was infected shock in +2 months, 1 case was infected in +5 months (unknown), 1 case died from COVID-19 infection in +11 months, and 1 case died from infection in +16 months (unknown).

A 60-year-old man, initially presented in October 2017 and was diagnosed with MM, immunoglobulin G kappa subtype, stage IIIA, characterized by TP53 mutation and t(4;14) chromosomal translocation. Initial treatment consisted of 4 cycles of BCD chemotherapy, which achieved VGPR. Subsequently, in 2018, the patient underwent haploidentical hematopoietic stem cell transplant with a 3/6 HLA-matched donor, using a BUCY + ATG conditioning regimen. At 3 months after transplant, the patient achieved CR; however, at 6 months, PD was noted. Retreatment with the initial regimen induced PR.

In September 2019, bone marrow FCM evaluation revealed 12.1% myeloma cells, and magnetic resonance imaging (MRI) identified an abnormal signal lesion in the T12 vertebral body, with extensive skeletal destruction. The patient underwent 3 courses of VRD and 5 cycles of IRD, achieving CR. In January 2021, relapse occurred, and 6 cycles of D-KPd were administered, resulting in VGPR. However, by September 2021, PET-CT imaging revealed new sites of bone destruction with elevated glucose metabolism, multiple pathological spinal compression fractures, and myeloma involvement of the liver, cervical lymph nodes, and esophageal wall. Treatment with doxorubicin hydrochloride liposomes, dexamethasone, and selinexor was initiated.

The patient was first admitted to our hospital on 9 October 2021. An M-protein level of 20.3% was shown, elevated serum FLC kappa (171 mg/L), and an increased kappa-to-lambda ratio of 85.07. Bone marrow pathology confirmed anaplastic plasmacytoma (Figure 4), MRD test results of 0.09% malignant monoclonal plasma cells, and partial expression of BCMA (60% expression rate). On 14 October, lymphocytes were collected for CAR T-cell manufacturing. By 23 October, the patient developed numbness from the navel to the inguinal region, progressive weakness in the left lower limb (muscle strength decreased to grade 2), difficulty urinating, and inability to walk independently. MRI revealed cervical and thoracic spinal canal stenosis with spinal cord compression and injury, consistent with CNS involvement. A 3-day course of FC chemotherapy was administered, and on 28 October, BCMA CAR T cells were infused at a dose of 2 × 10⁶/kg. One month after CAR T-cell infusion, the patient experienced fever, spinal and rib pain, nausea, diarrhea, elevated transaminase levels, coagulation abnormalities, and hematuria. All symptoms disappeared two weeks after infusion, MRD and morphology of bone marrow aspiration were both negative. At the 1-month evaluation, MRD remained negative, and bone marrow pathology revealed only a small number of suspected abnormal plasma cells. The M-protein level decreased to 3.5%, and the patient was assessed as partial remission. After CAR T-cell infusion, paralysis symptoms gradually resolved, and the patient regained normal walking ability within 1 month. MRI showed significant improvement in spinal cord compression, and PET-CT showed significant disappearance of lesions throughout the body (Figure 5).

At the 2-month evaluation, bone marrow pathology was normal, and M-protein levels further decreased to 2.7%. At 1 year after treatment, both bone marrow and M-protein assessments were negative, and the FLC ratio normalized. The patient has remained disease-free for 28 months as of the latest follow-up.

BCMA is predominantly expressed on malignant and nPCs as well as certain mature B cells, making it an attractive therapeutic target for MM.^2-4 Recently, 2 BCMA-targeted CAR T-cell therapies, idecabtagene vicleucel and cilta-cel, have been approved by the US Food and Drug Administration, and demonstrated efficacy in heavily pretreated patients with R/R MM.^5,6 Early BCMA CAR T-cell therapies relied primarily on mouse-derived scFvs, achieving ORRs of 60% to 80%, though efficacy in high-risk groups, such as those with EMD, remained suboptimal.^3,4 Recent advances have sought to enhance BCMA CAR T-cell therapy efficacy by improving the antigen-binding domain or substituting mouse-derived scFvs with human-derived scFvs to mitigate immunogenicity.^9,10 Further, dual-target CAR T-cell approaches combining BCMA and CD19 have also shown improved therapeutic outcomes, especially for the high-risk group.¹² Dual epitope–binding BCMA CAR T-cell constructs, such as LCAR-B38M, have demonstrated enhanced clinical responses at lower cell doses with reduced toxicity due to high-avidity binding.^4,13 

CAR T cells commonly use scFvs derived from monoclonal antibodies for antigen recognition. However, nanobodies (single-domain structures derived from the VHH found in camelids and sharks) offer a smaller, more stable alternative with comparable specificity and affinity, but reduced immunogenicity, potentially enhancing safety in CAR T-cell therapies.^15-19 In this study, anti-BCMA CAR T cells (S103) were developed using a CAR composed of 2 VHH antibodies fused to a CD8 hinge and transmembrane domain, paired with intracellular signaling domains of 4-1BB and CD3ζ.

The antigen-binding domain of S103 consists of a dual-nanobody structure boasting diverse binding epitopes. CAR T cells equipped with this dual-nanobody structure exhibit remarkable advantages when pitted against their traditional scFv-derived counterparts. Firstly, the dual-nanobody and dual-binding epitope construct can simultaneously recognize 2 different epitopes of tumor antigens, leading to a remarkable enhancement in the binding affinity to tumor cells. In contrast, traditional scFv typically recognizes only a single epitope, resulting in relatively weak binding. Furthermore, the dual-nanobody and dual-binding-epitope approach reduces nonspecific binding to normal tissue cells through dual-epitope recognition, thereby lowering the off-target effect. In comparison, traditional scFv is more likely to cause off-target issues because it recognizes only a single epitope. More importantly, in terms of pharmacokinetics and stability, nanobodies have a unique structure with a small molecular weight and high stability, enabling them to maintain activity in the body for an extended period. In contrast, scFv has an unstable structure, and is prone to aggregation and degradation. Finally, in terms of immunogenicity, nanobodies have a high degree of humanization, which elicits a weak immune response in the human body, enabling them to function persistently.²² The CAR T-cell product that has a similar structure to S103 is the commercially available cilta-cel. The long-term follow-up data of cilta-cel have effectively substantiated the aforementioned assertion.²³ 

A phase 1 clinical trial was conducted to evaluate S103 in patients with R/R MM. The results indicated that S103 CAR T-cell therapy was highly effective, particularly in high-risk subgroups. The ORR was 96.3% (26/27), with a CR + VGPR rate of 59.2% (16/27) at 1 month. At 3 months, the best overall response included an ORR of 100% (27/27) and a CR + VGPR rate of 81.5% (22/27). High-risk cytogenetic abnormalities, such as t(14;16), t(4;14), and t(14;20), along with TP53 mutations, are recognized as independent poor prognostic factors in plasma cell disorders.^24-27 Among the 11 patients with these abnormalities, S103 CAR T-cell therapy showed significant efficacy, achieving a 72.7% CR (8/11) within 3 months after BCMA CAR T-cell infusion, with a 1-year OS of 62.3% and 1-year PFS of 65.5%.

EMD, particularly large lesions, is a hallmark of advanced MM, and is often associated with resistance to standard therapies and poor prognosis.^28,29 In this study, 11 patients with EMD achieved an ORR of 100% within 1 month after S103 CAR T-cell infusion, including 5 cases of CR + VGPR. At 3 months, 9 patients remained responsive, with 5 maintaining CR + VGPR. Among 3 patients with CNS infiltration causing paralysis, all achieved PR within 1 month, and 2 achieved CR by 3 months.

Plasma cell leukemia, known for its poor prognosis and resistance to conventional therapies, such as proteasome inhibitors, immunomodulatory drugs, and anti-CD38 antibodies, is associated with rapid PD and short survival.^30-32 In this study, 4 patients with plasma cell leukemia received S103 CAR T-cell therapy, with all achieving clinical responses at 1 month, including 3 cases of CR + VGPR. At 3 months, 3 maintained CR + VGPR, with 2 achieving disease-free survival for over 1 year.

Anaplastic plasmacytoma, a rare and aggressive plasma cell disorder with poor prognosis, is associated with rapid PD and complications such as spinal cord infiltration and paraplegia.^33,34 One patient with anaplastic plasmacytoma achieved rapid disease control, full recovery from paralysis, and disease-free survival for 2 years following FC preconditioning and S103 CAR T-cell therapy.

Interestingly, 2 patients with BCMA expression below the threshold detected by flow cytometry still achieved CR with S103 CAR T-cell therapy, and sustained long-term disease-free survival. These findings suggest that BCMA expression levels may not significantly impact CAR T-cell efficacy in MM, consistent with previous studies.^3,35 

In terms of safety, there was no significant occurrence of CRS or neurotoxicity after infusion of S103 CAR T cells, although the occurrence of bacteria, fungi, and viruses can be controlled through corresponding anti-infective treatments and immunoglobulin supplementation. Four patients died due to infection during the follow-up process. Two cases occurred within 5 months, which may be related to the patient’s advanced age and uncontrolled underlying infection after multiline chemotherapy. Two patients occurred 11 months later at the local hometown. Both deaths from infection occurred after the patients returned home after completing their CART treatment. Therefore, patients should be required to receive regular infusion of immunoglobulin during the follow-up at discharge, which may help improve the long-term infection control after BCMA CART.

Because of the limited clinical data of this study, statistical analysis of different subgroups did not find statistical differences, and some conclusions have limitations. It is necessary to further expand the number of samples and conduct long-term follow-up to get more academic conclusions.

Overall, this study demonstrates that S103 CAR T-cell therapy with a dual VHH structure offers significant therapeutic benefits for high-risk plasma cell disorders, including cytogenetic high-risk MM, EMD, plasma cell leukemia, and anaplastic plasmacytoma. These findings highlight its potential to improve outcomes across diverse clinical scenarios.

The clinical trial demonstrated that S103 CAR T-cell therapy, incorporating dual-nanobody VHHs targeting BCMA, achieved a high ORR with a manageable safety profile in the treatment of R/R plasma cell myeloma. Notably, its efficacy extended to high-risk populations, including those with extramedullary lesions, high-risk cytogenetic abnormalities, plasma cell leukemia, and anaplastic plasmacytoma.

The authors thank all the medical staff in Hebei Yanda Lu Daopei Hospital for their clinical and technical assistance, especially their care for patients; and Hebei Senlang Biotechnology Co, Ltd, for their contributions to the preclinical investigations.

Contribution: X.Z., L.W., and P.H.L. developed the concept and designed the study; X.Z., L.W., Y.L., Q.W., X.N.H., and J.F.Y. conducted the clinical trials of chimeric antigen receptor T cells; L.N.Z. and X.G.Z. conducted a pathological analysis; X.Z., L.W., H.W., and L.Z. drafted and wrote the manuscript, and made major contributions to the analysis of clinical data; X.Z. and L.W. revised the manuscript; and all authors contributed to the article and approved the submitted version.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Xian Zhang, Hebei Yanda Lu Daopei Hospital, Yanjiao Economic and Technological Development Zone, Sipulan Rd, Langfang 065201, China; email: zhxian2@126.com.