Therapeutic Frontiers Expand for Patients With Relapsed/Refractory Multiple Myeloma

Multiple myeloma is a clonal plasma cell neoplasm characterized by bone lesions, renal impairment, cytopenias, and immunodeficiency. Despite significant therapeutic advancements in the past 2 decades that have resulted in improved survival, myeloma remains an incurable disease. The immune environment in which the cancer cells thrive is known to be a key player in the evolution of monoclonal gammopathies from premalignant stages to advanced malignancy. Further, immune dysregulation—marked by T-cell exhaustion, tolerance induction by tumor microenvironment, and tumor escape from immune surveillance—is important in the pathogenesis. Therefore, various immune strategies have been developed, including immune-enhancing drugs such as immunomodulatory drugs, checkpoint inhibitors, monoclonal antibodies, and, more recently, chimeric antigen receptor (CAR) T-cell therapy and bispecific antibodies for T-cell redirection.

Engineering the Future of Multiple Myeloma Treatment

CAR T cells are T lymphocytes genetically modified by viral vectors or nonviral technology such as DNA transposons to express a synthetic receptor to target a specific antigen. The single chain variable fragment (ScFV) on the ectodomain of the CAR recognizes tumor-associated antigens on the surface of tumor cells, binds to them, and initiates a cascade of cytotoxic signaling that leads to tumor lysis.

The ectodomain is linked to the intracellular domains by a hinge/transmembrane region, commonly derived from CD8 or IgG4. The intracellular portion is the signaling domain. In the first generation of CARs, this included only the CD3ζ signaling domain, which lacked a proliferation profile. Second- and third-generation CARs now include 1 (second generation) or 2 (third generation) costimulatory domains that are typically 4-1BB, CD28, and/or OX-40 to promote efficient T-cell signaling and persistence. Fourth-generation CARs, which further affect the tumor microenvironment to induce cytokine production after the CAR recognizes the target antigens, and fifth-generation CARs are being developed to further improve CAR efficiency and longevity.

On the other hand, bispecific antibodies use patients’ unengineered T cells. The off-the-shelf antibody is designed so that 1 end binds to a multiple myeloma cell and the other end binds to a killer T cell. The first bispecific antibody for multiple myeloma was developed with an ScFV that attached to the tumor antigen and another that attached to CD3 of the T-cell receptor complex of the T cell with a linker. The half-life was short, and continuous infusion was required. Since then, bispecific antibodies are manufactured with an Fc segment that increases the half-life so that the agent can be administered weekly or less frequently; this is the treatment of choice in ongoing clinical trials. New agents in development include trispecific antibodies that may have a costimulatory protein or target dual myeloma antigens or antibodies that engage natural killer cells.

There are several tumor antigens being investigated as suitable targets for CAR T-cell and T-cell redirected therapies, such as CD38, CD138, SLAMF7, CD19, and more. However, the most widely studied target for both CAR T-cell therapy and bispecific antibody therapies is B-cell maturation antigen (BCMA).

BCMA is a cell surface receptor in the tumor necrosis factor receptor superfamily member 17. It is deemed an ideal antigenic target because it is expressed specifically on normal and malignant plasma cells but not on hematopoietic stem cells, and it has higher expression on myeloma cells than normal plasma cells. It plays a key role in B-cell maturation and differentiation and promotes myeloma cell growth by binding to its ligands BAFF and APRIL. Expression of BCMA increases with progression from monoclonal gammopathy of undetermined significance to advanced myeloma.

Targeting BCMA

Based on encouraging results from the first major global multicenter phase 1 anti-BCMA CAR T study (NCT02658929) conducted in relapsed or refractory multiple myeloma,¹ investigators initiated the pivotal phase 2 KarMMa trial (NCT03361748).² The results of this trial were updated at the 18th International Myeloma Workshop (IMW), held in Vienna, Austria, in September 2021.³

Idecabtagene vicleucel (ide-cel; Abecma), formally bb2121, is an anti-BCMA second-generation CAR construct with 41BB costimulatory domain. Among 128 patients enrolled in the KarMMa study, 84% were triple-class refractory. At a median follow-up of 24.8 months, overall response rate (ORR) was 73%, with complete response (CR) or stringent CR (sCR) reported in 33% of patients. Minimal residual disease (MRD) was negative in 79% of complete responders. Further, responses were attained at a median of 1 month (range, 0.5-8.8) and the median duration of response (DOR) was 10.9 months.

The median progression-free survival (PFS) was 8.6 months and median overall survival (OS) was 24.8 months. DOR and PFS were improved in the higher-dose ranges and in complete responders. Similar degrees of responses were observed in all subgroups, including Revised International Staging System for multiple myeloma III criteria, extramedullary disease, and high tumor burden. In terms of adverse events (AEs), cytopenias were observed in 97% of patients. Grade 3/4 neutropenia was seen in 89% of patients, grade 3/4 thrombocytopenia was seen in 52%, and grade 3/4 infections in 23%. Cytokine release syndrome (CRS) was seen in 84% of patients: 78% at grade 1/2, 6% at grade 3 or higher. CRS occurred at a median onset of 2 days and median duration was 5 days. Neurotoxicity was reported in 18% of patients, 4% of whom reported the AE as grade 3 or higher. Results of the study led to FDA approval of the first commercially approved CAR T-cell product in March 2021.

CARTITUDE-1 was a phase 1b/2 study (NCT03548207)⁴ that used a different CAR T product, ciltacabtagene autoleucel (ciltacel). Updated findings were presented from Usmani et al⁵ were presented at the 2021 American Society of Clinical Oncology (ASCO) Annual Meeting and at the IMW from Jagannath et al.⁶

Cilta-cel is a lentiviral vector-based anti-BCMA construct with a 4-1BB costimulatory domain. The BCMA-catching domain targets 2 different epitopes simultaneously, increasing the binding affinity, and is the same CAR construct as in the Chinese trial LEGEND-2 (NCT03090659).

In CARTITUDE-1, 97 patients with a median of 6 prior lines of therapy were enrolled. At baseline, 88% were triple-class refractory and 99% refractory to last line of therapy. At a median follow-up of 18 months, the ORR was 98% for all patients and included an sCR rate of 80%. Responses were attained at a median of 1 month and deepened over time. The median DOR was 21.8 months overall (95% CI, 21.8-not estimable) and was not reached in patients with sCR. MRD negativity was achieved in 92.0% of evaluable patients.

The 18-month PFS rate was 66% (95% CI, 54.9%-75.0%) and the 18-month OS rate was 80.9% (95% CI, 71.4%-87.6%). These results far surpass outcomes with other non–T-cell mediated novel agent therapies in triple-class refractory patients.

In terms of safety, cytopenias were universal and 92 of 97 patients experienced any-grade CRS; 95% were grade 1/2 and had a median time of onset of 7 days and duration of 4 days. All-grade neurotoxicity was reported for 21% of patients, 10% of whom had neurotoxicity of grade 3 or higher. Although most neurotoxic events occurred in the setting of CRS, 12 patients had late neurotoxicity, 6 of whom resolved, 1 had ongoing neurotoxicity, and 1 died because of neurotoxicity. There were 21 deaths on study: 2 occurred in fewer than 100 days, 10 deaths were because of disease progression, and 6 were because of treatment-related AEs. Late recovery (greater than 1 month) of grade 3/4 cytopenias from first onset was seen in 10% of neutropenia and 26% of thrombocytopenia. Cilta-cel is expected to gain FDA approval this year.

Comparing Outcomes

At the American Society of Hematology (ASH) Annual Meeting 2020, Shah et al7 presented an analysis that compared efficacy outcomes seen in the KarMMa trial with those reported from the MAMMOTH study,⁸ which was a retrospective observational study of conventional care regimens in patients with triple-class refractory multiple myeloma.

The MAMMOTH study, which has been used in other comparative studies, has been a benchmark for investigators to compare therapeutic maneuvers in patients with triple-class exposed relapsed or refractory multiple myeloma who have received various standard-of-care therapies. The analysis applied matching-adjusted indirect comparisons to assess the efficacy of ide-cel and conventional care and showed that, in a matched population, ide-cel treatment was associated with a significantly higher ORR, PFS, and OS than conventional care.

Cilta-cel was similarly compared with conventional treatment in the MAMMOTH study and was presented by Costa et al at ASCO 2021.⁹ The MAMMOTH data set was used to identify patients with multiple myeloma refractory to anti-CD38 monoclonal antibodies who would meet eligibility for CARTITUDE-1 and who received conventional therapy. The intention-to-treat population (ITT) in CARTITUDE-1 was defined as patients who underwent apheresis, and a modified ITT population was defined as subset of patients who received cilta-cel at the recommended phase 2 dose (RP2D).

ORR, PFS, and OS for both the ITT population and modified ITT population in CARTITUDE-1 vs matching MAMMOTH cohorts were found to be superior. Specifically, the ORR in the ITT cohort was higher in CARTITUDE-1 compared with the MAMMOTH counterpart (84% vs 28%). Patients in the CARTITUDE-1 ITT cohort vs MAMMOTH cohort had improved PFS and OS rates at 12 months, 73% vs 12% and 83% vs 39%, respectively. Comparing the modif ied ITT cohorts, patients in CARTITUDE-1 had superior ORR (96% vs 30%), 12-month PFS rate (79% vs 15%), and 12-month OS rate (88% vs 41%).

Therefore, in patients with relapsed or refractory multiple myeloma beyond therapy with immunomodulatory drugs, proteasome inhibitor, and anti-CD38 monoclonal antibody, treatment with CAR T, ide-cel, or cilta-cel is associated with higher response rate and superior PFS and OS when compared with conventional treatment.

Other CAR T trials were reported at ASH 2020 and are being studied in various phase 1/2 trials. Research is directed at improving the efficacy and persistence of CAR products, which vary by source of product (autologous vs allogeneic CAR T cells), choice of vector (lentiviral, retroviral, or nonviral DNA transposon technology), use of humanized ScFv to prevent immunogenicity, CD4/ CD8 ratio controlled to enrich for central memory phenotype to improve longevity of CAR T cells, dual target constructs to prevent relapses because of antigen escape, CARs against non-BCMA targets to treat BCMA negative relapses, and more.

Bispecific antibodies are in earlier stages of development than CAR T. The majority of antibodies target BCMA, although there are some targeting antigens other than BCMA that have great potential in patients who have relapsed post BCMA-targeted therapies with BCMA-negative plasma cells.

Teclistamab is an anti-BCMA/anti-CD3 bispecific antibody with intravenous and subcutaneous formulations. Results of the MajesTEC-1 study (NCT03145181) were published in Lancet.^10,11 Investigators treated 157 patients with median of 6 prior lines of therapy, of whom 82% were triple-class nail-related AEs in 18%. Dysgeusia was reported in 57% of patients. A phase 2 expansion study of talquetamab in the RP2D MonumenTAL-2 is recruiting. refractory, 90% were refractory to last regimen, and 85% were previously transplanted were enrolled in a dose escalation/expansion study. A total of 40 patients received the RP2D of 1500 μg/kg. At RP2D, the median time to response was 1 month and median time to CR was 3 months.

At a median follow-up of 7.2 months, median DOR was not reached (7.2-not reached). The ORR was 65% in the RP2D group, 58% had a very good partial response (VGPR) or better, and 40% had a CR or better. Importantly, the majority of patients in CR were MRD-negative at 10^-6.

Among responders, 85% were alive and progression-free at follow-up. The most common AEs of any grade were CRS, all grade 1 or 2, (70%) and neutropenia (65%). Grade 3 or 4 AEs occurred in 80% of patients, with the most common being neutropenia (40%), anemia (28%), and thrombocytopenia (20%). Infections occurred in 45% of patients and were grade 3 or higher in 23%.

Talquetamab is an anti-GPRC5D/ CD3 first-in-class duo antibody. Results from the phase 1 MonumenTAL-1 trial (NCT04634552) were presented by Chari et al at ASH 2020¹² and updated at the IMW by van de Donk et al.¹³

GPRC is highly expressed in poorrisk myeloma, and in hair follicles. In the MonumenTAL-1 trial, 174 patients with a median of 6 prior lines of therapy were enrolled, 102 to the intravenous arm and 72 to the subcutaneous formulation arm, in dose escalation and expansion cohorts. At baseline 71% of patients were triple-class refractory and 86% were refractory to last line of therapy; 21% of patients had received prior BCMA-targeted therapies.

The ORR was 70% at the RP2D, and 50% of responders had a VGPR or better, with a median time to first confirmed response of 1 month. Responses were durable and deepened over time, with 81% of responders continuing on treatment after a median follow-up of 6.3 months. CRS was reported in 79% of patients; 4% had CRS of grade 3/4. Median time to onset of CRS was a day after subcutaneous dose, and the duration was 2 days. Neurotoxicity was reported in 7% of patients (grade 1/2) and was mostly in the context of CRS. Grade 1/2 skin-related AEs were seen in 75% of patients and nail-related AEs in 18%. Dysgeusia was reported in 57% of patients. A phase 2 expansion study of talquetamab in the RP2D MonumenTAL-2 is recruiting.

Ongoing Research in Bispecific Antibodies

Various other bispecific antibodies are in clinical trials, including Regeneron 5458, another anti-BCMA/anti-CD3 bispecific antibody with very encouraging results reported at ASH last year.¹⁴ The ORR was 63% and responses were achieved by 1 month. The median DOR was 6 months, and among responders with more than 6 months of follow-up, 83% had ongoing responses for up to 13 months and 74% of responders remained on treatment.

TNB-383B is a fully human triple-chain BCMA × CD3 bispecific antibody with a unique anti-CD3 moiety for target lysis with minimal cytokine release and 2 anti-BCMA moieties. It is administered intravenously every 3 weeks without step-up dosing. Data for 58 patients from the ongoing first-in-human study were presented at ASH 2020.¹⁵ Safety data were comparable with results of other studies.

Cevostamab is another non–BCMA bispecific antibody. The target antigen is FcRH5, which is found on naïve and memory B cells and plasma cells. The anti-FcRH5/anti-CD3 is administered intravenously every 3 weeks, and data were presented at ASH 2020.¹⁶

Finally, CC-93269, a bispecific antibody with 1 CD3 and 2 BCMA binding sites, shows encouraging early data as well.

Future directions for bispecific antibodies include understanding resistance mechanisms, studying them in combination with various agents, and understanding sequencing strategies.

Because myeloma is marked by clonal heterogeneity, combinations of drugs with different mechanisms of action and nonoverlapping toxicities are frequently used with success. With the arrival of this new era of powerful immunotherapeutic tools such as CAR T-cell therapy and T-cell redirective agents, a sound understanding of their optimal use is key to maximizing their potential.