B cell maturation antigen (BCMA)-based Immunotherapy for Multiple Myeloma

Abstract

Introduction:

B cell maturation antigen (BCMA) contributes to MM pathophysiology and is a target antigen for novel MM immunotherapy. Complete responses have been observed in heavily-pretreated MM patients after treatment with BCMA antibody drug conjugates (ADC), chimeric antigen receptor T, and bi-specific T cell engagers (BiTE®). These and other innovative BCMA-targeted therapies transform the treatment landscape and patient outcome in MM.

Areas Covered:

The immunobiological rationale for targeting BCMA in MM is followed by key preclinical studies and available clinical data on efficacy and safety of therapies targeting BCMA from recent phase I/II studies.

Expert opinion:

BCMA is the most selective MM target antigen, and BCMA-targeted approaches have achieved high responses even in relapse and refractory MM as a monotherapy. Long-term follow-up and correlative studies using immuno-phenotyping and -sequencing will delineate mechanisms of overcoming the immunosuppressive MM bone marrow microenvironment to mediate additive or synergistic anti-MM cytotoxicity. Moreover, they will delineate cellular and molecular events underlying development of resistance underlying relapse of disease. Most importantly, targeted BCMA-based immunotherapies used earlier in the disease course and in combination (adoptive T cell therapy, mAbs/ADCs, checkpoint and cytokine blockade, and vaccines) have great promise to achieve long term disease control and potential cure.

1. Introduction

Multiple myeloma (MM) is a plasma cell (PC) malignancy in the bone marrow (BM) characterized by heterogeneity among and within individual patients. The abnormal PCs proliferate in the BM and have a high propensity for clonal heterogeneity and evolve and expand in the setting of complex interactions with diverse lineage cell subsets and growth factors in the BM microenvironment. The multiple-step development of MM starts from a premalignant precursor condition called monoclonal gammopathy of undetermined significance (MGUS), which progresses to smoldering MM (SMM), active MM, and in some cases PC leukemia (PCL). This progression is associated with genomic clonal expansion/evolution, as well as increasing immunosuppression in the BM milieu. Despite the advent of high dose therapy and autologous stem cell transplantation and the introduction of novel agents including proteasome inhibitors (PIs), immunomodulatory drugs (IMiDs), and monoclonal antibodies, most MM eventually relapses due to the development of drug resistance [1]. Thus, novel treatment strategies are urgently needed, especially in high-risk relapse and refractory (RR) MM.

The ability of therapeutic monoclonal antibodies (mAbs) to activate immune effectors with remarkable MM cytotoxicity and with favorable toxicity profiles have allowed for their incorporation into treatments for RRMM, and more recently newly diagnosed disease. These therapeutic mAbs eliminate mAb-coated target MM cells via activation of host defense mechanisms regulated by Fcγ receptor-expressing effector cells. These effector cells in turn induce antibody-dependent cell cytotoxicity (ADCC), complement-dependent cytotoxicity (CDC), and/or antibody-dependent cell-mediated phagocytosis (ADCP) to eradicate target MM cells [2]. In late 2015, US FDA approved two therapeutic monoclonal antibodies (mAbs) daratumumab (directed against CD38) [3,4] and elotuzumab (directed against signaling lymphocytic activation family 7/SLAMF7) [5,6] in MM. Efficacy and durability were shown with these first two naked IgG1 mAbs, especially in combination with current standard of care myeloma therapies and even in elderly patients with MM refractory to both PIs and IMiDs, as well as in newly diagnosed subjects [7,8]. These mAbs have transformed the treatment paradigm in MM and other PC diseases [9].

Since ideal targets for effective immunotherapies should be selectively and strongly expressed on the surface of patient MM cells relative to normal cells, CD38 and SLAMF7 may not be optimal MM target antigens. Both molecules are also expressed on normal tissues including activated B and T lymphocytes, monocytes, natural killer (NK) cells, and other effector cells. Indeed, NK cell depletion occurs following daratumumab treatments [10] and may reduce the efficacy of this antibody [11]. Nonspecific toxicities also may limit their full clinical utility. Most recently, B-cell maturation antigen (BCMA), a cell surface protein universally and selectively expressed at high levels on malignant PCs, has emerged as an ideal antigen to be targeted by novel immunotherapeutic modalities in MM [12–15].

2. BCMA is the most specific survival receptor without kinase activity in MM

BCMA, a transmembrane glycoprotein [16] and non-tyrosine kinase receptor [17], belongs to the tumor necrosis factor receptor (TNFR) superfamily and is almost exclusively expressed in plasmablasts [18] and PCs [19,20]. In nonmalignant lymphoid tissue, BCMA protein is detected in the interfollicular area of germinal centers, but not in the follicular mantle zone [20]. Unlike the other two functionally-related TNFRs B-cell activation factor receptor (BAFF-R) and transmembrane activator and calcium modulator and cyclophilin ligand interactor (TACI), BCMA is uniquely induced in late memory B cells committed to PC differentiation and is present in all PCs. In contrast, other normal tissues have undetectable BCMA transcripts except for plasmacytoid dendritic cells (pDCs) which express significantly lower BCMA levels than PCs from the same individuals [13,21]. Multiple independent studies using tissue gene expression profiling and immunohistochemistry as well as flow cytometry analysis indicate that BCMA [22] transcript and protein are strongly expressed in PCs and only weakly detected in normal organs due to PCs [12,13,15]. BCMA−/− mice have impaired long-term PC survival but show no defects in short-term production of immunoglobulins, early humoral immune response, and B-cell development [19,23]. Thus, BCMA is critical for differentiation and optimal survival of long-lived PCs in the BM but is dispensable for B cell homeostasis.

Significantly, BCMA is overexpressed in the pathogenic PCs [13,15,24,25] and MM-supporting pDCs in the BM of MM patients compared to normal donors [21]. Relative to CD38 and SLAMF7, BCMA is expressed at high levels in all MM cell lines and malignant PCs, with increased expression levels associated with progression from MGUS to SMM to active MM [13,15,24]. In contrast, BAFF-R protein is hardly detectable in all MM cell lines and patient MM cells [24,26], and TACI is expressed at significantly lower frequency and levels compared with BCMA [24,26–29]. Significantly, BCMA is one of targets of the master transcriptional factor mediating PC survival interferon regulatory factor 4 (IRF4), which may interact with BCMA in MM cells [25]. Significantly, BCMA-dependent biological effects in MM cells are mainly regulated by the key growth and survival protein kinase B (AKT), MAPK, and nuclear factor (NF)-κB signaling cascades [21,27]. BCMA overexpression further promotes in vivo growth of xenografted MM cells harboring p53 mutations in mice. In addition to upregulated key survival proteins Mcl1, Bcl2, and Bcl-xL, significantly increased CD31/microvessel density (MVD) and vascular endothelial growth factor (VEGF) have been shown in BCMA-overexpressing versus paired control tumors. BCMA-overexpressing xenografted myeloma cells have increased gene expression crucial for osteoclast (OC) activation, adhesion, and angiogenesis/metastasis, as well as upregulation of genes mediating immune inhibition including programmed death ligand 1 (PD-L1), transforming growth factor β (TGF β), and interleukin 10 (IL-10). Thus. BCMA is critical in MM pathobiology and contributes to the MM cell-induced immunosuppressive BM microenvironment.

BCMA contains 184 amino acids and is a natural substrate for γ-secretase, a multi-subunit protease complex, resulting in its decreased level on the cell membrane and formation of soluble BCMA (sBCMA) [30,31]. Shedding of BCMA by γ-secretase controls PCs in the BM. It may represent a potential biomarker for B-cell involvement in human autoimmune diseases, since sBCMA is elevated and correlated with disease activity in systemic lupus erythematosus. Gamma-secretase directly cleaves BCMA, without prior truncation by another protease. This process is facilitated by the short length of BCMA’s extracellular domain containing 54 amino acids, suggesting sBCMA as a specific decoy for its ligands. Indeed, reduced membrane BCMA via γ-secretase-induced shedding led to decreased BCMA-mediated NF-κB activation induced by its cognate ligand a proliferation inducing ligand (APRIL). Soluble BCMA levels may also play a pathophysiological role in MM, since they can inhibit other BCMA ligand B cell activating factor (BAFF) from binding to its membrane-bound BCMA. This led to suppressed development and differentiation of normal B-cell and PCs, and thereby decreased polyclonal antibody levels [32]. Like other MM antigens, i.e., SLAMF7, CD138, CD38, sBCMA levels are significantly increased in MM patients versus healthy individuals, and levels in patient’s serum is further correlated with disease status and prognosis [33,34]. Indeed, serum BCMA levels may predict both progression free survival (PFS) and overall survival (OS) in MM patients. Furthermore, sBCMA may represent a novel prognostic and monitoring tool, especially for non-secretory MM. However, to date it remains unclear whether γ-secretase activity is increased in MM patient cells with increased BCMA levels, and whether elevated sBCMA in serum samples of patients with increased disease burden is associated with upregulation of γ-secretase.

BCMA binds to 2 cognate ligands, APRIL and BAFF, to deliver critical survival signals for BM long-lived PCs, but not memory B cells [35]. Specifically, APRIL binds to BCMA at >2-log higher affinity than BAFF [36], and is a more selective PC survival factor than BAFF due to its higher affinity of binding to BCMA predominantly expressed on all PCs. Upon binding to APRIL, BCMA transmits differentiation and survival signals to induce immunoglobulin isotype switching, especially IgA production, and viability of plasmablasts and PCs in the BM [37–39]. APRIL is mainly produced by myeloid lineage cells and OCs in the BM [37,40–43]. OCs, macrophages, and other myeloid precursor cells promote MM cell growth and survival in the BM [41,42,44,45] via APRIL in a paracrine manner. Importantly, MM cell lines xenografted in APRIL−/− mice have significantly reduced growth, indicating that APRIL also mediates in vivo MM progression [46]. Furthermore, binding of secreted APRIL from myeloid cells to MM cells depends on heparan sulfate on CD138 [47,48], supporting cross-talk between these PC-related receptors (BCMA, TACI, and CD138). In patient serum, APRIL, but not BAFF, correlates with pro-angiogenic cytokines such as VEGF and its receptor, MVD and syndecan-1, which parallel MM progression [49]. Overlapping downstream events following BCMA overexpression and APRIL binding to BCMA in MM cells include growth stimulation, evasion from apoptosis, as well as production of angiogenesis factors and key immunosuppressive molecules such as IL-10, PD-L1 and TGF-β (Figure 1). Thus, the APRIL/BCMA signaling cascade plays a significant role in modulating MM pathophysiology and immunosuppression in the MM BM microenvironment [14,21,41,42,50].

Figure 1. Biological sequelae following BCMA activation upon APRIL binding in MM.

BCMA (TNFRSF17, CD269) is selectively and highly expressed on all MM cell lines and patient MM cells. Upon APRIL binding to BCMA, three key growth and survival signaling cascades (MEK/ERK, PI3K/AKT, NFκB) are induced, leading to a variety of downstream gene expression. As listed in the right, the representative gene list includes those with critical functions in the pathogenesis of MM. APRIL, an important plasma cell-related factor secreted from various non-MM cells in the bone marrow, promotes MM pathophysiology mainly in a paracrine mechanism. Importantly, APRIL via BCMA induces expression of immune checkpoint molecule PD-L1, which binds to PD-1 on immune effector T/NK cells. Thus, APRIL/BCMA axis plays an important role to promote MM cell-induced immunosuppression in the bone marrow microenvironment. These data strongly support BCMA as a promising therapeutic target in MM. (adapted with permission from [42])

MM, multiple myeloma; OC, osteoclast; TAM, tumor-associated macrophage; MDSC, myeloid-derived suppressor cell; pDC, plasmacytoid dendritic cell

3. BCMA is an excellent target for novel MM immunotherapies

BCMA is a lineage-restricted differentiation antigen present in normal and malignant PCs, but absent in naive B cells and most memory B cells. As described above, BCMA significantly contributes to the multiple-step development and pathogenesis in MM. In addition, BCMA was identified as a target of donor anti-tumor immunity in patients with relapsed MM post allogeneic transplantation who respond to donor lymphocyte infusion [51]. All these studies support targeting BCMA in novel therapeutic strategies. Already several BCMA immune therapies BCMA immune have demonstrated promising responses in clinical trials, as described below and highlighted in Figure 2. These methods include antibody drug conjugates (ADCs), chimeric antigen receptor (CAR) T cells and Bispecific T-cell engagers (BiTEs). Mechanisms of their anti-MM activity are described in detail in the following sections.

Figure 2. Multiple promising BCMA-targeting immunotherapies in MM.

Shown here are examples of BCMA CAR T, BCMA-ADC, and BCMA-BiTE® treatment modalities, which have induced high frequency deep responses, even in of heavily-pretreated relapse and refractory MM patients. Those listed in bold were recently granted fast track status by US FDA and have entered phase 2 trials. LCAR-B38M, known as JNJ-68284528 in the US, has shown strong phase I results in China and been accepted for review by the China FDA. Listed are some of the many BCMA CAR T cell and BCMA bispecific antibody T cell engager (BiTEs) entering clinical trials. BCMA-NK Bi or Tri Ab, not shown here, can also potently induce effector cell-mediated lysis of MM cells in a BCMA-dependent manner. (adapted with permission from [50])

CAR T, chimeric antigen receptor T cell; ADC, antibody drug conjugate; BiTE®, bispecific T-cell engager; Bi, bispecific full-length immunoglobulin; MM, multiple myeloma cell; NK, natural killer cell; Mϕ, macrophage

4. Current BCMA-targeting immunotherapeutics in MM

4.1. Antibody-drug conjugates (ADCs)

ADCs, as opposed to naked mAbs, are composed of mAbs covalently linked to very toxic agents by chemical- or protein-based linkers. They represent a rapidly growing class of cancer therapeutics to treat advanced/metastatic disease. They are developed to specifically target cancer cells and thereby allow for selective delivery of the deadly payload to MM cells while reducing the risk of toxic side effects. Novel small compounds that are otherwise too fatal to be used systemically are excellent candidates for new ADC development. After delivery to target cells via specific mAbs, the potent payload is released and mediates cytotoxicity.

4.1.1. GSK2857916

Since BCMA its selectively expressed on malignant PCs and not on essential organs and normal CD34+ hematopoietic stem cells, the first MM ADC linked a novel anti-BCMA mAb J6M0 to a new class of potent microtubule inhibitors monomethyl auristatin E (MMAE) or monomethyl auristatin F (MMAF) [52]. J6M0 is a novel humanized mAb specifically binding to BCMA with high affinity (Kd, ~ 0.5nM), and contains an engineered Fc by removing fucosylated oligosaccharides. Such modification results in >1-log increased binding affinity to Fc-bearing NK cells. Compared with its normal Fc homolog, J6M0 augments BCMA-specific NK-mediated tumor cell lysis via ADCC and simultaneously recruits macrophages to trigger ADCP of MM cells both in vitro and in vivo. J6M0 blocks BAFF- and APRIL-induced signaling via BCMA, while the auristatin component (MMAE or MMAF) is released intracellularly via a lysosome-dependent mechanism. Both J6M0-ADCs were first tested side by side in co-cultures of MM cells and BM stromal cells (BMSCs) NK cells, monocytes, and PBMCs. The J6M0-MMAF named GSK2857916 has a protease resistant maleimidocaproyl (mc) linker, resulting in greater and more selective cytotoxicity against MM cells co-cultured with BMSCs than does its MMAE homolog. Moreover, GSK2957916 induces negligible by-standard killing against BCMA-negative cells surrounding MM cells in the BM milieu due to high structural stability and poor cell membrane permeability of MMAF to avoid drug leakage. The use of non-cleavable mc linker in GSK2857916 also limits non-specific toxicity and ensures high on-target potency. This novel defucosylated antagonistic BCMA-MMAF potently blocks proliferation and colony formation of MM cells, associated with induction of cell cycle arrest at G2/M and caspase-dependent apoptosis [13]. Effective in vivo anti-MM activity by GSK2857916 was also demonstrated in both xenografted and disseminated models of multiple human MM cell lines, with significantly prolonged (> 3.5 months) host survival, without host weight loss in the host mice receiving up to 9 doses. GSK2857916 is the first therapeutic ADC with 3 mechanisms of action (ADCC, ADCP, and apoptosis) against MM cells [14]. Furthermore, lenalidomide further enhances GSK2857916–induced MM cell lysis via enhanced Fc-dependent and -independent mechanisms via NK-binding [53] and caspase 3/7 [54], respectively.

GSK2857916 was next evaluated in a phase 1 study in patients with RRMM to determine its pharmacokinetic parameters, pharmacodynamic characteristics, therapeutic potential, and safety (). This international, multi-center, open-label, first-in-human clinical study enrolled patients with RRMM who failed at least 3 prior lines of therapy, including stem cell transplantation, alkylators, a proteasome inhibitor, an immunomodulatory agent, and anti-CD38 antibody [55]. Based on tolerability and clinical activity, 3.40 mg/kg every 3 weeks was selected as the dose schedule for the phase 2 dose-expansion study. Of note, BCMA receptor saturation was observed at doses of 1.92 mg/kg or higher. Importantly, GSK2857916 monotherapy showed 60% overall response rate (ORR) in 35 heavily pretreated RRMM patients, including 51% > very good partial response, with a median PFS of 7.9 months. Thrombocytopenia and corneal problems were the most frequently reported adverse events (AEs), as observed with MMAF and other MMAF-linked ADCs [56]. While higher comparative incidence of ocular AEs seems to be associated with maytansinoid- and MMAF-containing ADCs, the exact mechanisms of such toxicity are still undetermined. Multiple on- and off-target mechanisms of toxicity must be considered. Attention should be paid on these clinical features to facilitate early recognition and intervention in both preclinical and clinical investigations. These AEs are distinct from with the neurological and pulmonary toxicities attributed to MMAE, i.e., brentuximab vedotin (anti-CD30) [57]. GSK2857916 does not require a labor-intensive manufacturing process or treatment at a specialized center, and is not associated with potentially severe toxic effects, such as cytokine release syndrome (CRS) and encephalopathy. As noted above, the ability of GSK2857916 to deplete MM cells via at least 3 different mechanisms of action in preclinical studies may account for its promising clinical activity.

Based on these data, GSK2857916 received breakthrough therapy status for RRMM by the FDA, as well as PRIME designation from the European Medicines Agency (EMA) in late 2017. Currently, more than 5 GSK2857916-related clinical trials are ongoing in patients with RRMM. These include: a phase 2 combination trials such as with Pembrolizumab (anti-PD-L1) (DREAMM 4); investigating the efficacy and safety of two doses of GSK2857916 in RRMM pretreated with anti-CD38 mAb; and evaluating GSK2857916 together with lenalidomide/dexamethasone or bortezomib/dexamethasone.

4.1.2. MEDI2228

MEDI2228 is another novel anti-BCMA pyrrolobenzodiazepine (PBD) dimer-linked ADC [58,59] designed to target MM cells and non-proliferating MM progenitor cells (MPCs). The MPC CD19+CD138- population, is present in MM patient BM and clonally related to the bulk of CD19-CD138+) MM cells, sharing the same light-chain restriction and genetic translocations [59]. Importantly, BCMA was detected on the surface of both the MM and the MPCs in all patient samples analyzed, including MM resistant to bortezomib, prednisone, and cyclophosphamide; bortezomib, prednisone, and dexamethasone; or mitoxantrone and dexamethasone treatment regimens. Since MPCs are not proliferative cells, they are less sensitive to anti-tubule drug (such as MMAF)-based ADCs, which mainly target mitotic cells. Importantly and in contrast, synthetic novel PBD dimers form sequence-selective, interstrand crosslinks in the minor groove of DNA, and are cytotoxic to both proliferating tumor and dormant tumor-progenitor cells. The structure of MEDI2228 includes a fully human anti-BCMA antibody site-specifically conjugated to a PBD tesirine via a protease-cleavable linker. This ADC is rapidly internalized into MM cells and trafficked to lysosomes, where the PBD warhead is released and causes cell death via accumulation of irreparable DNA damage. Furthermore, MEDI2228 is characterized by strong binding to membrane-bound BCMA but only affinity to recombinant monomeric human BCMA or sBCMA. MEDI2228 is highly active in MM cell lines (IC⁵⁰ range 6 to 210 ng/ml) regardless of BCMA levels and retains its activity in the presence of BMSCs. Some MM cell lines expressing higher BCMA are more resistant to MEDI2228, indicating that unidentified mechanisms also contribute to the sensitivity to MEDI2228 [57]. A single injection of MEDI2228 at very low doses (0.1 mg/kg) induces human MM regression in mouse xenograft models. Moreover, a similar dosing schedule also significantly inhibits MM tumor growth in disseminated mouse models of human MM.

The first Phase I clinical trial of MEDI2228 (), including both dose-escalation and -expansion, is currently ongoing in patients with RRMM.

4.1.3. AMG 224

AMG 224 is comprised of an anti-tubulin inhibitor maytansine derivative conjugated to antibody lysine residues via the non-cleavable 4-(N-maleimidomethyl) cyclohexane-1-carboxylate linker (anti-BCMA-mc-DM1). The trial of AMG 224 is currently not recruiting patients (https://clinicaltrials.gov/ct2/show/NCT02561962).

4.2. Chimeric antigen receptor (CAR) T cells

Cellular therapies provide the potential to overcome drug resistance and induce durable remissions with a single treatment. Importantly, chimeric antigen receptor (CAR) T cell therapy for cancer is a significant milestone in modern medicine. CAR is a fusion protein containing an extracellular antigen recognition domain typically from a single-chain variable fragment (scFv) of an antibody, intracellular signal domains, and costimulatory domains from immune effector cells. CAR T cells recognize and bind to antigen on tumor cells in a major histocompatibility complex (MHC)-independent manner. After CAR T cells bind to target tumor cells, an immune synapse is formed, followed by the activation and proliferation of cytotoxic T cells to produce immune cytokines and lysis of targeted tumor cells [60,61]. Autologous CAR T cells are generated from the patient’s own peripheral blood T cells and infused back to the patient to eliminate the targeted tumor. The FDA approval of CD19-targeted CAR-T therapies for B-cell malignancies including acute lymphoblastic leukemia (ALL) and diffuse B cell non Hodgkin lymphoma has confirmed the remarkable efficacy and potential curability of CAR-T immunotherapy [62].

More recently, anti-BCMA CAR T cells have been evaluated as adoptive cellular therapy for MM. Within the past 5 years, there are at least 13 phase II trials of BCMA CAR T cells and more than 30 phase I trials listed under https://clinicaltrials.gov. The following sections will include representative preclinical and clinical results to date.

4.2.1. CAR-BCMA T by the National Cancer Institute

The first anti-BCMA CARs were composed of the scFv portions derived from different mouse anti-human BCMA mAbs with the hinge and transmembrane region of human CD8α, the signaling moiety of the CD28 costimulatory molecule, and the signaling domains of the CD3ξ [12]. They were cloned into a γ-retroviral mouse stem cell-based splice-gag vector. After transducing T cells with these γ-retroviral vectors encoding these CARs, CAR-BCMA T cells specifically recognize BCMA on MM cell lines and patient MM cells. These CAR-BCMA T cells, including CD4+ and CD8+ subsets, induce BCMA-specific cytokine production, proliferation, cytotoxicity, and in vivo tumor eradication. Compared with CD8+ T cells, a higher percentage of CD4+T cells produced IL-2. These CAR-BCMA T cells produce IFNγ when stimulated with primary MM cells and kill patient MM cells. Importantly, soluble BCMA protein does not interfere with anti-BCMA CAR function in vitro and in vivo.

The most effective CAR-BCMA T cell in preclinical studies by Carpenter et al [12] were tested in the first-in-human CAR T cell therapy phase I clinical trial in RRMM (). A threshold of >50% BCMA expression, assessed by flow cytometry or IHC, was required for study enrollment. In 2016, the preliminary data showed safety and efficacy of CAR-BCMA T therapy in 12 patients [63]. This trial went on to enroll 24 heavily treated RRMM patients (range of prior lines of treatment: 3-19, median: 9.5) who received autologous CAR-BCMA T cells (ranging from 0.3 – 9×10⁶/kg) following lymphodepletion chemotherapy with cyclophosphamide (cy, 300 mg/m², 3 days) and fludarabine (fdb, 30mg/m², 3 days) [63,64]. Among 16 patients infused with highest dose level of CAR T cells, the ORR was 81%, including 2 sCR, 8 VGPR, and 3 PR. Minimal residual disease (MRD) negativity assessed by flow cytometry was seen in 11 patients. The median EFS for patients who received the highest dose level of CAR T cells was 31 weeks, with 6 patients ongoing responses and 10 patients with progressive disease. After CAR T cell infusion, the CD4/CD8 ratio significantly decreased, and proliferating and differentiating CD8+ CAR T cells showed effector memory phenotype. The toxicity profile included CRS, with severity correlated with tumor burden. Grade ¾ hematologic AEs of lymphodepletion and CAR T cell therapy included leukopenia or neutropenia in 15 patients (15/16, 93.8%) receiving highest doses of CAR T cells. No deaths were noted in this study.

This first reported BCMA CAR T cell therapy [64] demonstrates significant responses even in heavily-pre-treated RRMM patients with defective host immune system. Importantly, this study provided valuable experiences to inform further clinical studies of BCMA CAR T therapies in patients with RRMM. First, responders have increased number of CAR T cells peaking between 1 and 2 weeks after infusion. Higher CAR T cell number was associated with better responses, as seen in the CD19 CAR T clinical trials in young RR ALL patients [65]. Second, this BCMA CAR T therapy has considerable toxicities, and some patients require vasopressor treatment due to severe CRS-related toxicities. An increased risk of severe CRS correlates with higher percentages of PCs in the BM of MM patients. As in the anti-CD19 CAR T therapy for ALL, tumor burdens at the beginning of the treatment is associated with CRS-related toxicities, without unexpected off-target or off-tumor toxicities noted to date. Importantly, CRS and neurotoxicity were reversible in all patients and manageable by administration of anti-IL-6 receptor (IL-6R) inhibitor tocilizumab. Third, this CAR contains BCMA-recognition domains derived from mouse Ab that is potentially immunogenic and possibly susceptible to immunologic rejection. Fourth, decrease or loss of BCMA expression on MM cells was observed after BCMA CAR T cell infusion. The serum BCMA level significantly decreased in responders compared with non-responders. When the disease progressed, some BCMA-negative MM cells were detected. BCMA downregulation might therefore impact the durability of this CAR T cell therapy, as reported for CD19 escape in patients whose disease became resistant to CD19 CAR T treatment. Therefore, BCMA antigen loss can occur under the selective pressure of anti-BCMA CAR T cell treatment.

4.2.2. bb2121 (Bluebird Bio/Celgene)

BB2121 (Bluebird Bio/Celgene) [66] was designed to extend the initial clinical observations from the report by Ali et al. [63] in the context of CAR architecture using the 4-1BB co-stimulatory domain and lentiviral transduction, rather than CD28 co-stimulation and γ-retroviral vector transduction [12]. It had a murine scFv for BCMA as used in previous reports [63,67]. BB2121 CAR-transduced T cells by lentiviruses (bb2121) exhibited a high frequency of CAR + T cells and robust in vitro activity against MM cell lines, lymphoma cell lines, and primary chronic lymphocytic leukemia peripheral blood [66]. A single intravenous administration of bb2121 resulted in rapid and sustained elimination of the tumors and 100% survival in NSG mouse models of human MM, Burkitt lymphoma, and mantle cell lymphoma [66].

The phase I CRB-401 study of bb2121 () is a multicenter clinical trial with centralized manufacturing to evaluate the safety and efficacy of bb2121 as a potential therapy for RRMM. BCMA expression levels were not required for this study. Initial results from the dose-escalation phase have shown promising efficacy and safety. In all, 21 patients (3 prior treatments or double refractory, ≥50% BCMA expression) underwent leukapheresis and subsequent cyclophosphamide fludarabine lymphodepletion, and then received between 5 × 10⁷ and 1.2 × 10⁹ anti-BCMA CAR T cells. ORR across all patients and doses was 89% (75% PR, 27% CR, 95% CI: 65-99). All patients with CR were MRD-negative, and DOR was > 134 days (95% CI: 7-361). Anti-BCMA CAR T cell expansion was observed, and bb2121-transduced CAR T cells were measurable in patients’ blood for up to 24 weeks post-infusion. Fifteen total cases of CRS (71%) were observed, with two grade 3 cases, and no grade ¾ neurotoxicity (NTX) was observed. bb2121 continues to show promising efficacy at dose levels ≥ 150 × 10⁶ CAR T cells, with deep and durable ongoing responses and manageable CRS and NTX. The FDA granted a breakthrough designation to bb2121 for the treatment of patients with RRMM in November 2017. An update in December 2017 (n=43) [68] reported an ORR of 94% in the 22 evaluable patients treated with the highest CAR T cell dose range, with 10 of 18 (56%) CR or unconfirmed CR and MRD-negativity in 9 of 10 evaluable patients. CRS, primarily grade 1-2, was reported in 15 of 21 (71%) patients; 2 patients had grade ≥ 3 CRS that resolved in 24 hours. There were 2 deaths on study; both patients had achieved CR and had not progressed. One death on study, due to cardiopulmonary arrest > 4 months after bb2121 infusion in a patient with an extensive cardiac history, was observed while the patient was in sCR and was assessed as unrelated to bb2121. Additionally, median PFS was not yet reached after 40 weeks of follow up.

Results for the first 33 consecutive patients received a bb2121 infusion were just reported in NEJM [69] using the data-cutoff date of 6.2 months after the last infusion date to evaluate the safety as the primary endpoint. Hematologic AEs were the most common events of grade 3 or higher and 76% had CRS with 70% at grade 1 or 2; 6% at grade 3. Forty-two % had NTX at grade 1 or 2 and 3% (one) with NTX at grade 4. The ORR was 85% with 45% (15/33) patients achieving CR but 40% (6/15) of patients with CR later experiencing a relapse. The median PFS was 11.8 months (95% CI, 6.2 to 17.8). CAR T cells were detected up to 1 year after the infusion and responses were correlated with CAR T cell expansion. This treatment has also been granted PRIME eligibility from the EMA.

The phase 2 KarMMa trial () is ongoing, and will serve as the basis for regulatory submission to the FDA [68].

4.2.3. LCAR-B38M (Legend/GenScript Biotech; Janssen Research & Development, LLC)

LCAR-B38M - biepitopic CAR T cells-targeting BCMA (Legend/GenScript Biotech, Nanjing, China) – is under evaluation for efficacy in patients with RRMM in 4 independent institutional studies at participating hospitals (, n=57 LEGEND-2) [70,71]. This CAR uses 4-1BB instead of CD28 as a co stimulatory domain. Specifically, LCAR-B38M is a dual epitope-binding CAR T cell therapy directed against 2 distinct BCMA epitopes, which differed from CAR-BCMA from NCI and bb2121. No BCMA expression levels are required for enrollment in the trial, and all patients received cyclophosphamide before CAR T infusion. Data presented in June 2017 showed that 18 of 19 (95%) RRMM patients achieved CR or near CR status, without relapse at a median follow-up of 6 months [72]. At data cutoff in the paper by Zhao et al [70] for , 57 patients had received LCAR-B38M CAR T cells with and experienced ≥ 1 AEs. Grade ≥ 3 AEs were reported in 37 of 57 patients (65%), including leukopenia (17 of 57patients; 30%), thrombocytopenia (13 of 57patients; 23%), and aspartate aminotransferase increase (12 of 57patients; 21%). CRS occurred in 51 of 57 patients (90%), with 4 of 57 patients (7%) at grade ≥ 3. One patient reported NTX of grade 1 aphasia, agitation, and seizure-like activity. The ORR was 88% (95% confidence interval [CI], 76 to 95): 39 of 57 patients (68%) achieved a CR: 3 of 57 patients (5%) achieved a VGPR; and 8 of 57 patients (14%) achieved a PR. The median time to response was 1 month (range, 0.4 to 3.5). With a median follow-up of 12 months, the median DOR was 16 months (95 percent CI: 12-not reached), and a median PFS of 15 months for all patients was observed. Among the patients who achieved an MRD negative CR measured by 8-color flow cytometry, the median PFS was 24 months.

In the second 17 RRMM trial of LCAR-B38M [71], CAR T cells were infused after lymphodepleting chemotherapy, followed by 3 infusions versus 1 infusion of the total CAR T dose in 8 and 9 cases, respectively. No response differences were noted among the two delivery subgroups. The ORR was 88.2%, with 13 stringent CR (sCR) and 2 VGPR, and 1 non-responder. Only 6 of 17 experienced grade > 3 CRS, with 1 death from a very severe CRS. Notably, positive anti-CAR antibody constituted a high-risk factor for relapse/PD, and patients who received prior autologous hematopoietic stem cell transplantation had more durable responses. Thus, LCAR-B38M () achieved deep and durable responses, with a manageable and tolerable safety profile, in patients who failed a median of 3 prior therapies (range, 1-9). It has been accepted for review by the China FDA.

Strong LCAR-B38M data has further fueled the development of this anti-BCMA CAR T therapy by JNJ-68284528 (Janssen/Johnson & Johnson, Beerse, Belgium) in a Phase ½ clinical trial () in patients with high risk RRMM [73]. A Phase 2 trial is also ongoing in Chinese participants (). EMA granted PRIME designation to Janssen’s investigational CAR-T therapy on 4^th April 2019.

4.2.4. CART-BCMA (University of Pennsylvania/Novartis)

Lentivirally transduced T cells with a fully-human BCMA-specific CAR containing the hinge and transmembrane domain of CD8 and CD3ζ intracellular signaling domains coupled with 4-1BB costimulatory domain (CART-BCMA) [74] was generated, and evaluated in the study reported by Cohen et al () [75]. Twenty-five heavily-pretreated patients with RRMM subjects received 1) 1-5 × 10⁸ CART-BCMA cells alone; 2) cy 1.5 g/m² and 1-5 × 10⁷ CART-BCMA cells; and 3) cy 1.5 g/m² and 1-5 × 10⁸ CART-BCMA cells. One patient died of candidemia and progressive MM following treatment for severe CRS and encephalopathy. Responses were seen 4 of 9 patients (44%) in cohort 1, 1 of 5 patients (20%) in cohort 2, and 7 of 11 patients (64%) in cohort 3. There were 5 partial, 5 VGPR, and 2 CR, 3 of which were ongoing for >11 months, including 1 ongoing sCR at 2.5 years with MRD-negative BM by flow cytometry. This report showed that CART-BCMA infusions, with or without prior lymphodepleting chemotherapy, are clinically active in heavily pretreated RRMM patients. Residual MM cells in some responders have reduced BCMA expression, which then increased at disease progression. CD4:CD8 T cell ratio and % CD45ROCD27+CD8+ T cells in the pre-manufacturing leukapheresis product were correlated with responses and CART-BCMA numbers. Unlike the previously-reported NCI study [64], low BCMA expression or high tumor burden did not exclude patients. It remains to be determined if BCMA levels predict degree of response and resistance mechanisms to CAR-BCMA T cells.

4.2.5. bb21217 (Bluebird Bio/Celgene)

The new CAR T bb21217 built on the bb2121 platform aims to improve cell persistence by modifying the manufacturing process to include a PI3K inhibitor, thereby increasing memory-like T cells to enhance the persistence of cytotoxic T cell function [76]. The phase I dose-escalation CRB-402 study () includes 12 patients, 58% of whom had high-risk disease. The median number of prior regimens received was 7, and 83% of patients had received a stem cell transplant. The preliminary results show 83.3% ORR at the first dose level tested, with CR/sCR in 25% patients and > VGPR in 50%patients. Sixty-seven % patients developed CRS, and NTX occurred in 25%patients. Responses are deepening over time, and CR was achieved as late as month 10. MRD negativity was achieved in 4 of 4 responders, and 2 non-responders were MRD-positive.

4.2.6. JCARH125 (MSKCC)

JCARH125 is a human-derived single- scFv product with a lentiviral vector and 4-1BB costimulatory domain, whose structure and optimized manufacturing process enrich for central memory T-cell phenotype [77]. The phase I/II EVOLVE study () included 44 patients with highly RRMM (median of 9 prior therapies, 64% with high-risk cytogenetics) receiving various doses of JCARH125. Thus far, ORR was 82%, with CR/sCR in 27% and > VGPR in 48% of patients. CRS was seen in 80% of patients, with 9% having a grade ≥ 3 event. NTX occurred in 18% and 7% were grade ≥ 3. A multicenter phase I/II trial of JCARH125 is underway.

4.2.7. P-BCMA-101 (Poseida Therapeutics)

Preclinical studies show that P-BCMA-101, manufactured using the “piggyback” approach as a nonviral system for DNA delivery plus mRNA including a small human fibronectin domain for BCMA, resulted in higher levels of stem cell memory T cells [78]. Rather than using a traditional antibody-based binder, P-BCMA-101 utilizes an anti-BCMA Centyrin™ fused to a CD3ζ/4-1BB signaling domain. Centyrins are fully human and have high binding affinities, but are smaller, more stable, and potentially less immunogenic. In addition, no viral transduction was used to engineer autologous CAR T cells, in CAR T described above. Preliminary results from the initial clinical study () in 23 patients with RRMM suggest that P-BCMA-101 may lead to superior response rates compared to other CAR-T therapies at similar doses, with a favorable safety profile [78]. P-BCMA-101 showed a 100% ORR in patients receiving the planned Phase 2 dose, with all patients showing rapid and deep responses and no CRS or NTX seen. P-BCMA-101-transduced T cells are less costly due to elimination of the need for virus and high percentages of stem cell memory T phenotype (T_SCM). P-BCMA-101 also contains a switch to rapidly decrease or eliminate therapy when dangerous side effects occur. P-BCMA-101 is the first anti-BCMA CAR-T therapy to receive regenerative medicine advanced therapy (RMAT) designation from the FDA in late 2018.

4.2.8. Descartes-08 (Cartesian Therapeutics)

Descartes-08 is CD8+ anti-BCMA CAR T-cells are modified transiently using mRNA transfection in an attempt to reduce the risk of CRS, NTX, and prolonged normal plasma cell aplasia seen in permanent modified BCMA CAR T [79]. Descartes-08 can be generated faster than traditional virus-transduced T cells and manufactured at clinical scale showing high (>90%) purity, post-cryopreservation viability, and transfection efficiency. Descartes-08 demonstrated robust dose- and time- dependent killing of BCMA+ MM cell lines, including those with acquired resistance to IMiDs, as well as primary MM cells co-cultured with autologous BMSCs. Compared with transfected pan- (CD3+) anti-BCMA CAR T-cells, Descartes-08 (CD8+) demonstrated enhanced transfection and killing with reduced secretion of IFNγ, an inflammatory cytokine highly correlated with CRS. Descartes-08 also demonstrated robust dose-dependent efficacy in a disseminated human MM model (i.v. MM1SLuc cells in NSG mice). Descartes-08 is currently in phase I trial () to determine whether this transiently transduced CAR T can improve the benefit: risk profile of CAR T-cells in MM and enable their use in patients with earlier-stage disease.

4.2.9. APRIL CAR T cells (Autolus Limited)

One new format of CAR T cell therapy is the APRIL-CAR T, which simultaneously targets BCMA and TACI receptors on MM cells. Downregulated BCMA proteins were reported in recent BCMA CAR T trials as one mechanism whereby MM cells relapse after initial responses [64,75]. The cytolytic activity of this novel APRIL CAR was significant, even at low antigen levels [80]. In an animal model study, elimination of MM cells was observed in the bone marrow with concurrent detection of APRIL-CAR T cells. A Phase I/II trial of APRIL CAR T cells (AUTO2) is currently recruiting patients with RRMM ().

4.3. Bispecific T cell engager (BiTEs)

Bispecific T-cell engagers (BiTEs) are comprised of two scFvs, one that binds to CD3 molecules on T-cells and the other to a surface antigen on the target MM cells to bridge T cells to tumor cells. BiTEs can elicit cytotoxicity without requiring antigen-presenting cells, MHC-I/peptide complex, and co-stimulatory molecules. Blinatumomab, the first CD19/CD3 BiTE^® (Amgen, Thousand Oaks, CA, USA), has been FDA-approved for the treatment of RR CD19+ ALL and other hematological malignancies.

4.3.1. AMG 420 (BI 836909) (Amgen)

AMG 420 (formally named BI 836909) is the first BiTE® with two linked scFvs to bind concomitantly to MM via BCMA and T cells via CD3 [81]. This molecule contains the scFv targeting BCMA in N-terminal and the same scFv targeting CD3ε as for blinatumomab in C-terminal, followed by a hexa-histidine (His6-tag). Due to its small size (55 kDa), AMG 420 (BI 836909) is highly potent and efficacious at cross-linking MM cells and T cells, thereby inducing formation of cytolytic synapse. Multiple key cytokines (IFNγ, IL-2, IL-6, TNFα, IL-10) related to T effector function are released, ultimately leading to activation of T cells mediating specific lysis of BCMA+ MM cells, but not BCMA-negative cells. The anti-MM effect of AMG 420 is not significantly affected by the presence of BMSCs or sBCMA and APRIL (up to 150 and 100 ng/ml, respectively). It potently induces autologous patient MM cell lysis, regardless of disease status. AMG 420 treatment decreased tumor growth in a subcutaneous NCI-H929 MM xenograft model at doses of 50 µg/kg/day and higher, and prolonged survival in an orthotopic L-363 xenograft MM model at doses of 5 µg/kg/day and higher.

The Phase 1 trial () was designed to determine the maximum tolerated dose and dose-limiting toxicities of AMG 420 in 42 patients with RRMM who had been treated with > 2 prior lines of therapy including PIs and IMiDs [82]. Patients received escalating doses of AMG 420, 0.2 to 800 µg daily continuous infusion for four weeks of every six-week cycle. AMG 420 (400µg daily dose) reduced tumor burden in 70% of patients and achieved MRD-negativity in 40% patients. Six patients remained in continuing response 7.5 months after entering the trial. A total of 20 patients experienced serious AEs during the trial, with 17 requiring hospitalization. These included infections, peripheral nerve damage, liver failure, cardiac failure, fluid accumulation, biliary obstruction, spinal cord compression, renal failure, and weight loss. CRS was reported in 16 patients. Two patients died from AEs unrelated to AMG 420 treatment. Importantly, this first-in-man study of AMG 420 showed no major toxicities at doses up to 400 µg/daily, which is the recommended dose for further phase 2 investigation.

AMG 420 is the first BiTE to show efficacy in myeloma and has been granted fast track status by FDA. Since AMG 420 is an off-the-shelf therapy, patients can be promptly treated, whereas BCMA-targeting CAR T-cell therapies require T cells to be collected, modified, expanded, and reinfused. AMG 420 also had a lower rate of CRS than BCMA-targeting CAR T cells, potentially making it more suitable for older patients. Concerns about the relatively high infection rate may be due to the need to administer AMG 420 via continuous infusion.

4.3.2. AMG 701 (Amgen)

An extended half-life anti-BCMA BiTE® (BCMA HLE-BiTE®) AMG 701 produces effective in vitro and in vivo anti-MM activity and can therefore was given once weekly in patients with RRMM [83,84]. Furthermore, AMG 701 both potently induces T cell-directed lysis of BCMA-positive MM cells and triggers immunomodulatory effects to overcome the immunosuppressive BM microenvironment. These results provide the rationale for clinical trials of AMG 701, alone and in combination, to improve patient outcome in MM [84], and Phase I trial is currently ongoing in patients with RRMM.

4.3.3. TNB-383B (TeneoBio/AbbVie, Inc.)

This novel bispecific molecule is composed of two anti-BCMA heavy chain variable domains linked to a unique anti-CD3 T-cell recruiting arm to maximize MM cell killing while markedly reducing cytokine release [85,86]. TeneoBio’s human anti-BCMA VH domains showed in vitro cytotoxicity against BCMA-expressing human MM cells in immunodeficient mice, and Phase I clinical studies will begin in collaboration with AbbVie, Inc. in early 2019.

4.4. APRIL targeting reagent

Therapeutic agents blocking APRIL/BCMA are under investigation as well [21,41,87]. Besides APRIL binding to TACI on MM cells, APRIL binds to inhibitory regulatory T cells via TACI to significantly enhance their survival and immune inhibition [88]. A novel mouse anti-human APRIL antibody hAPRIL01A (01A) inhibits the binding of APRIL to BCMA and TACI [21,42,88]. Importantly, 01A inhibited APRIL- and OC-induced proliferation of MM cells, and further induced apoptosis of MM cells in co-cultures. 01A also enhances the cytotoxicity mediated by IMiDs and PIs in co-cultures of MM cells with BCMA-negative BM accessory cells and effector cells. Furthermore, APRIL-induced expression of genes mediating immunosuppression, such as PD-L1, TGF-β, and IL-10, are decreased in MM cells following 01A treatment. A phase I clinical trial of the fully humanized 01A mAb BION-1301 is ongoing in patients with RRMM [89].

4.4. BCMA-peptide vaccine and adoptive cell therapy

Cancer vaccines may increase the likelihood of response or improve response duration. They are and will be being clinically evaluated as combination therapies with other effective therapies that have immunomodulatory mechanisms of action.

To expand the spectrum of BCMA-targeting therapy, novel engineered peptides specific to BCMA, BCMA_72-80 (YLMFLLRKI) and BCMA_54-62 (YILWTCLGL), were recently identified for use in vaccination and adoptive T cell therapy in MM [90]. In recent preclinical studies, these engineered BCMA peptides display improved binding affinity/stability to HLA-A2 molecules compared to their native peptides and induce highly functional BCMA-specific cytotoxic T lymphocyte (CTL) expressing activation and co-stimulatory molecules. Importantly, heteroclitic BCMA_72-80 triggered expansion of BCMA peptide-specific memory CD8⁺ CTL with Th1-specific cytokine and lytic activities against MM. These results provide the framework for therapeutic application of these highly immunogenic heteroclitic BCMA peptides in MM patients as vaccines and/or as stimuli for expansion of autologous antigen-specific memory CTL for adoptive immunotherapy. A clinical trial will vaccinate patients with heteroclitic BCMA_72-80 to induced peptide specific memory cells, which will then be harvested and expanded ex vivo. These memory heteroclitic BCMA_72-80 specific CTLs will be administered as adoptive immunotherapy, followed by vaccination with heteroclitic BCMA_72-80 to promote ongoing central memory anti-MM immunity.

Conclusion

We are now entering a new era integrating novel therapeutic mAbs into MM treatments. BCMA, the most specific MM antigen identified to date, has been successfully targeted by novel immunotherapeutics including CAR T, ADC, and BiTE®, which have achieved 70-100% response rates in patients with heavily-pretreated RRMM who have no further treatment options. These unprecedented results further validate BCMA as a novel MM target and provide a very promising and unique platform for future therapeutics. Importantly, CAR-T and BiTE®, offer new approaches to restore anti-MM host immunity. ADC can be used even in patients with RRMM whose immune system is severely compromised. Peptide-based vaccination/adoptive immunotherapy offers the opportunity to achieve memory anti-MM responses [90–93]. Ongoing studies will further enhance selective potency, enhance safety, and prolong responses by promoting persistence of ongoing anti-MM immunity. Importantly, combination of these BCMA targeting immune therapies with each other or other agents, i.e. IMiDs, check point blockade or epigenetic modulators [92], will enhance anti-MM immunity and clinical responses.

Expert Opinion

BCMA-targeted immunotherapies including CAR T, ADC, and BiTE® have achieved remarkable clinical responses in patients with RRMM who have no treatment options. Their relative clinical utility will depend on efficacy, safety, and cost. To date, many clinical trials of BCMA CAR T cell therapy have achieved high rates of deep MRD negative clinical responses in patients with RRMM. However, more broad application of CAR T-cell therapy in MM will require: optimization of persistence and survival of CAR T cells; decreasing the toxicity associated with lymphodepletion conditional chemotherapy and CAR T-cell therapy; strategies to avoid antigen escape and other mechanisms underlying disease progression; and improving cost effectiveness. The requirement for such T cell-based BCMA therapy to be done only at major medical centers also limit its utility. Currently, this therapeutic approach cannot be applied to patients with rapidly progressive disease due to at least 2-3 week-preparation time for autologous BCMA CAR T cells.

Efforts are ongoing to make BCMA CAR T therapy more potent, safe, durable, accessible, and affordable for patients. For example, different CAR designs and transfection methods as well as culture conditions during CAR T cell manufacture are being explored. To reduce inherited immunogenicity by murine scFv and increase persistence of CAR T cells, new CARs utilize humanized scFv. Compared to CD28 costimulatory domain, 4-1BB is favored for the durability of CAR T cells and more persistent anti-tumor activity. At present, the high rate of CRS due to increased inflammatory cytokines upon CAR T treatment has required intensified care including therapies targeting IL-6R/IL-6. Novel CAR design and T cell transfection methods are under evaluation, including transfection with mRNA coding BCMA CAR to T cells [79,94] or novel CAR constructs containing switches to control CAR T expansion in vivo when needed [94], to improve the therapeutic index. New BCMA CAR designs include those containing a suicidal gene inducible caspase 9 and truncated epidermal growth factor receptor, or synthetic Notch (SynNotch) receptors in which binding to tumor and CAR activation requires binding to two antigens [95]. SynNotch receptors control and localize CAR T cell response for precision immunotherapy via combinatorial antigen-sensing circuits. Other approaches can be improved by optimization for spacer length and low antigen-independent (tonic) signaling in BCMA CAR constructs.

As one-third of newly diagnosed MM patients are older than 75 years [96] and total lymphocytes decrease with advancing age, it may be difficult to generate sufficient numbers of autologous CAR T cells. Moreover, >30% elderly patients are frail. Also, serious complications such as renal failure may develop during the typical waiting time of 2-4 weeks required for manufacturing autologous CAR T cells from highly refractory patients. These factors could limit the use of CAR T-cell therapy or incorporation of conditioning chemotherapy to CAR T-cell therapy in these patients. Although lymphodepleting conditioning may increase > grade ¾ hematologic AEs, its application prior to adoptive CAR T cell administration enhances depletion of tumor burden and persistence of infused T cells by decreasing T regulatory cell and increasing cytotoxic T cell function [97]. To make CAR-T therapy more cost-effective and acceptable even in the elderly population, novel allogenic normal donor-derived CAR T cells are being produced using gene editing technologies to generate off-the-shelf universal CAR T products. For example, ALLO-715 is a novel allogeneic BCMA CAR T and may provide a readily available off-the-shelf product for the treatment of RRMM [98]. Cellectis’ UCARTCS1 targeting SLAMF7/CS1 is the first allogeneic CAR-T therapy for MM to enter clinical development this year (https://www.cellectis.com).

Thus far, the PFS post BCMA CAR T therapy is about 12 months overall and 18 months in patients achieving MRD negativity. It is therefore critical to define mechanisms underlying relapse. Newer CARs are under investigation to target other MM antigens, i.e., CD38, SLAMF7, CD138, or to target combinations of dual MM antigens to make CAR more efficacious. Combining BCMA CAR with CAR targeting another MM antigen may reduce the risk of relapse due to tumor antigen escape. For example, combined infusion of CD19 and BCMA-specific CART cells after autologous transplantation (SZ-MM-CART02 study, ) is under evaluation as consolidation treatment to abrogate relapse in patients with high risk MM.

Among 4 BCMA CAR T products under FDA’s rapid support for approval, P-BCMA-101 also represents a significant improvement over first generation, immunoglobulin-based, and virally-transduced CAR-T products. Since this product does not depend on viral transfection, it is more cost effective and quickly available for infusion. Moreover, predominant T stem memory (Tscm) cells in this product may extend the duration of response and reduce toxicity compared to other CAR-T approaches.

In patients with high MM burden, using other anti-MM agents or BCMA ADCs to achieve response and then use CAR T cell therapy would likely improve both efficacy and safety. In addition, repeated infusions with lower number of CAR T cells or transient CAR T cells generated via mRNA but not viruses, represents another strategy to reduce risk of CRS and still achieve efficacy. Other approaches include combining CAR T cells with IMiDs, checkpoint blockade, or epigenetic inhibitors such as histone deacetylase inhibitors, to enhance T cell persistence and prolong response. Finally, BCMA peptide vaccination or BCMA-dendritic cells (DC) post BCMA CAR T cell therapy may prolong BCMA specific memory immunity and clinical response. Futhermore, the use of BCMA-CAR NK cells represents another potential treatment for MM. NK cell-based immunotherapy may result in less severe side effects than genetically modified T cell-based immunotherapy, since NK cells generally have shorter life span than cytotoxic T cells (CTLs). However, a direct comparison of side effects between CAR-modified T cells and NK cells is not currently available. In the future, a closed, automated workflow system is critical to decrease the cost of generating CAR-modified T, NK cells.

BCMA BiTE and BCMA CAR T-cell therapy achieve significant responses and may allow for treatment-free intervals, when patients can enjoy a high quality of life without therapy-related toxicities. Although the first MM BiTE® AMG 420 has a relatively short serum half-life and may not stimulate persistent MM cell killing, it is safe, off the shelf, and can be re-administered if needed. To overcome the obstacle of maintaining serum levels with bolus or intermittent infusion, the next generation long half-life BCMA BiTE® AMG 701 molecule shows enhanced MM cell lysis, especially when combined with IMiDs [84]. CRS and NTX can be observed after BiTE treatment, requiring close monitoring and management; however, these agents may be useful even in elderly or frail patients. Finally, a novel BCMA x CD3 TNB-383B which induces significantly reduced cytokine secretion by T-cells without appreciable reduction of efficacy in vivo or ex vivo [85] will soon enter clinical study in RRMM.

Since the majority of RRMM patients have defective host immune function, potent BCMA ADCs may have an advantage over CAR T cells. ADCs are off the shelf, do not depend on host immune function, and can be administered without the delays inherent to CAR T cell production. The first BCMA ADC GSK2857916 (Belantamab mafodotin) delivering MMAF warhead has shown efficacy, without the CRS and NTX attendant to BCMA CAR T therapy. In addition, MEDI2228 ADC depletes MM cells and MPCs by PBD-induced irreparable DNA damage and is under evaluation in patients with RRMM. In the future, these BCMA ADCs could be used to quickly reduce tumor burden, and then CAR T therapy or BiTE could be administered at lower and safer doses to prolong responses with fewer side effects. Other novel promising payloads are under development, including α-amanitin, tubulysins, hizoxin, or spliceostatins [50]. To improve penetration, novel ADC formats such as non-IgG scaffolds or non-internalizing mAb scaffolds, may be applied to anti-BCMA ADCs. Besides modification of ADC structure, combinations of ADC with other anti-tumor agents with different mechanisms of action are also under clinical investigation, i.e., , , . Given that immune checkpoint inhibitors have clinical efficacy in several cancers, a study evaluating the clinical efficacy of combining immune checkpoint inhibitor Pembrolizumab with ADC GSK2857916 has recently started ().

For those MM patients with low BCMA expression, BCMA-targeted immunotherapies may be improved by increasing binding affinity to membrane BCMA. For example, MEDI2228 ADC has improved membrane BCMA binding and shows very potent anti-MM effects in preclinical studies [59]. Alternatively, reduced γ-secretase activity can enhance membrane BCMA while minimize sBCMA, suggesting the utility of combining γ-secretase blockade with current BCMA targeting therapies [99]. Indeed, BCMA CAR T cells combined with a γ-secretase inhibitor (JSMD194) is under evaluation in a clinical trial in patients with RRMM (). Finally, ongoing studies are determining the optimal strategies to assure that sBCMA does not interfere with clinical efficacy of BCMA targeting therapies.

Although it currently remains unknown whether any of these methods or strategies targeting BCMA may preclude another and how they will be utilized best, it is likely that optimal clinical benefit may be achieved by their use in combination or in sequence, as well as treatment in earlier stages. For example, naked anti-BCMA mAb or BCMA peptide vaccination/adoptive immunotherapy may be useful to treat precursor disease states MGUS and SMM and delay or prevent progression to active disease. BCMA-based immunotherapies will continue to transform the treatment landscape and patient outcome in MM for years to come.

Article highlights.

• B cell maturation antigen (BCMA) promotes MM pathogenesis in the BM microenvironment and is a very specific MM target antigen.

• Immunologically-based therapies targeting BCMA demonstrate promise independent of the genetic heterogeneity and genetic risk even in MM patients with no other treatment options.

• Immunotherapies targeting BCMA including antibody-drug conjugates (ADCs), autologous chimeric antigen receptor engineered T cells (CAR-T), and bispecific T cell engager (BiTE®) have achieved responses even in RRMM.

• The US FDA has granted Breakthrough Therapy Designation in 2017 to GSK for anti-BCMA ADC (GSK2857916/Belantamab mafodotin) auristatin immunotoxin and to Celgene/Bluebird Bio for anti-BCMA CAR-T therapy (bb2121/Idecabtagene vicleucel); as well as in 2018 to Poseida Therapeutics/Regenerative Medicine Advanced Therapy (RMAT) for P-BCMA-101 CAR-T cells, and to Amgen for anti-BCMA BiTE® (AMG 420).

• Although BCMA-targeted CAR-T cell therapy for MM has achieved remarkable responses even in advanced refractory disease, ongoing research is directed to increasing efficacy and prolonging responses, as well as improving safety and abrogating cytokine release syndrome (CRS)-related toxicity. Ocular toxicity and frequent infection are the major adverse events (AEs) reported in patients receiving GSK2857916 immunotoxin and AMG 420 BiTE®, respectively.

• Potentially more durable off-shelf BCMA-targeted immunotherapies include allogenic CAR T cells and BCMA peptide vaccines/adoptive immunotherapy. Ultimately combination immune approaches will be needed to achieve optimal clinical benefit.