Recent updates on CAR T clinical trials for multiple myeloma.

Proteasome inhibitors, immunomodulatory agents and monoclonal antibodies have dramatically changed the natural history of multiple myeloma (MM). However, most patients eventually suffer a relapse and succumb to the disease. Chimeric antigen receptor (CAR) engineered T cells targeting B cell maturation antigen (BCMA), CD138, CS1 glycoprotein antigen (SLAMF7) and light chains are in active development for therapy of refractory /relapsed (RR) MM. CD19- targeted CAR T cells in conjunction with autologous stem cell transplantation also showed activity in RRMM. Dual- target CAR T cells are in clinical trials for RRMM. This review summarized the recent updates of ongoing CAR T clinical trials for multiple myeloma.

Background

Proteasome inhibitors, immunomodulatory agents (IMiDs) and monoclonal antibodies have dramatically changed the natural history of multiple myeloma (MM) [1-7]. However, most patients eventually suffer a relapse and succumb to the disease after multiple lines of therapy [8-11]. Chimeric antigen receptors (CARs) are engineered receptors that can bind to a desired antigen and redirect the effector cells to a defined target [12-17]. Two CD19-engineered CAR T cell products, axicabtagen ciloleucel (yescarta, Kite) and tisagenlecleucel (kymriah, Novartis), have been approved for therapy of advanced B cell malignancies. The major clinical toxicities of CAR T cell therapy are cytokine release syndrome (CRS) and CAR T- related encephalopathy syndrome (CRES) [18-20], which require prompt diagnosis and intervention to prevent fatal complications [21, 22]. The CD19 targeted CAR-T cell therapy has inspired tremendous interests in searching for new targets for MM immunotherapy [23-30]. There are various ongoing clinical trials using CAR T cell technology to target myeloma antigens such as B cell maturation antigen (BCMA), CD138, CS1 glycoprotein antigen (SLAMF7) and immunoglobulin light chains [31-35] (Table 1). This review summarized ongoing CAR T clinical trials for multiple myeloma.

Table 1

BCMA-targeted CAR T clinical trials in multiple myeloma

Registration number (reference)	Phase	Dosage	No. of patients	Responses
NCT02215967 (49)	1	0.3–9.0 × 106 CAR+T cells/kg	12	sCR: 1; VGPR:2; PR:1; SD:8
NCT02658929 (51)	1	50–800 × 106 CAR+ T cells	33	sCR:12; CR:3; VGPR:9; PR:4; SD:4; PD:1
NCT03274219 (52)	1	150 × 106 CAR+ T cells	8	sCR: 1; VGPR: 3; PR: 2 MRD negative: 3
NCT03090659 (55)	1/2	0.07–2.1 × 106 CAR+ T cells/kg	57	CR: 39; VGPR: 3; PR: 8; MRD negative: 36; ORR: 88%
ChiCTR-ONH-17012285 (53)	1	0.21–1.52 × 106 CAR+ T cells/kg	17	sCR: 13; VGPR: 2; ORR: 88.2% NR: 1
NCT03430011 (58)	1/2	50–150 × 106 CAR+ T cells	19	sCR: 2; CR: 1; VGPR: 2; PR: 2; MR: 1
NCT03915184 (59)	NA	0.5–1.8 × 108 CAR+ T cells	16	CR: 2; PR: 4; VGPR: 6 ORR: 100%
NCT03070327 (61)	1	72–818 × 106 CAR+ T cells	11	VGPR: 2; ORR: 64%
ChiCTR1800018137 (64)	1	1.0–6.0 × 106 CAR+ T cells/kg	9	CR:4; VGPR:1; PR: 4; ORR: 100%
NCT02546167 (65)	1	1–50 × 107 CAR+T cells	25	sCR: 1; CR: 1; VGPR: 5; PR: 5
NCT03288493 (67)	1	48–430 × 106 CAR+ T cells	12	sCR: 1; nCR: 1; VGPR: 1; PR: 2
NCT03338972 (69)	1	5–15 × 107 CAR+ T cells	7	ORR:100%

Abbreviations: BCMA B cell maturation antigen, CAR Chimeric antigen receptor, VGPR Very good partial response, SD Stable disease, CR Complete response, PR Partial response, sCR Stringent complete response, PD Progressive disease, ORR Overall response rate, MRD Minimal residual disease, nCR near Complete response, BMPC Bone marrow plasma cells, IHC Immunohistochemistry, FC Flow cytometry, MR Minimal response, NR non-response, RRMM relapsed/refractory Multiple Myeloma, NA not available, E evaluable

BCMA (B cell maturation antigen)

BCMA was discovered initially by several groups [36-39]. BCMA gene was found to be fused to the interleukin-2 gene in the t(4;16) (q26;p13) translocation in a malignant T-cell lymphoma. BCMA gene is localized on chromosome band 16p13.13. The BCMA gene encodes a peptide with 184 amino acid residues and an estimated molecular weight of 20kd [37].

BCMA is also known as CD269 and TNF receptor superfamily 17 (TNFRSF17) [40]. BCMA ligands include B cell-activating factor (BAFF, also termed TNFSF13B) and a proliferation- inducing ligand (APRIL, also termed TNFSF13) [41]. BCMA is expressed almost exclusively in B lineage cells including plasmablasts and in particular at the stage from mature B to plasma cell (PC) terminal differentiation. In addition to normal B cells, BCMA is also expressed on MM cells and malignant B cells [31, 42]. BCMA is known to be absent on naïve and most memory B cells. In BCMA knock-out mice it was shown that the mice had normal B cell development and an intact humoral immune system [43]. BCMA expression is upregulated during PC differentiation. Hence, even though BCMA may not be critical for B-cell development, it plays a major role in B-cell maturation and differentiation into plasma cells. BCMA appears to enhance the survival of normal PCs and plasmablasts as well as long-lived PCs in the BM.

BCMA has a soluble form found in the peripheral blood of MM patients [44]. Injection of the soluble BCMA disrupted immune responses, affected splenic architecture and prevented the accumulation of peripheral B cells [45-47]. The soluble BCMA therefore may interfere theoretically with the myeloma-targeting capacities of BCMA-specific immunotherapeutics [48].

BCMA-targeted CAR T cell trials

Early BCMA-targeted CAR T trial

In a study of cell lines and human tissues, BCMA was found to be expressed in plasma cells and myeloma cells, but not in normal tissues and neither in hematopoietic stem cells. The first BCMA CAR contained a CD28 co-stimulation domain [31] (Fig. 1). The first-in-human phase I clinical trial of CAR T cells targeting BCMA was conducted in patients with RRMM (NCT02215967) [49]. Twelve patients were reported in the dose escalation trial. Four dose levels were reported. The four levels were 0.3, 1.0, 3.0, 9.0 × 106/kg. Among the 12 patients, 3 patients entered partial remission (PR), 8 patients had stable disease (SD), and 1 patient achieved stringent complete remission (sCR). Among the 6 patients treated on the 2 lowest dose levels, limited anti-myeloma activity and mild toxicity occurred. On the third dose level, 1 patient obtained a very good PR (VGPR). Two patients were treated on the fourth dose level of 9 × 106 CAR T cells/kg. After treatment, bone marrow plasma cells of the two patients became undetectable by flow cytometry. The first patient entered a sCR that lasted for 17 weeks before relapse, and the serum monoclonal protein of the second patient had decreased by > 95% 28 weeks after infusion of CAR-BCMA T cells. This patient remained in an ongoing VGPR. Both patients treated on the fourth dose level had CRS. The patients who received higher doses of CAR T cells had better responses but also a higher risk for adverse events (AEs), including CRS. This study also noted that soluble BCMA did not interfere with the efficacy of the BCMA-targeted CAR T cells. In addition, decrease of the soluble BCMA in the serum may serve as a biomarker for the efficacy of the anti-BCMA CAR T cells. This study was significant for the proof of concept of BCMA as a unique target for plasma cell malignancies.

Fig. 1

The schematic diagram of representative structures of BCMA-targeted chimeric antigen receptors (CAR). The BCMA CARs contain a single-chain of BCMA antibody variable fragment (ScFv), a transmembrane domain, a hinge region, a co-stimulation domain (4-1BB, CD28 or OX40), and a CD3z domain. Additional sequences (such as PI3K inhibitor) are added to enhance identification of CAR+ T cells. LCAR-B38M CAR contains two epitopes of BCMA ScFv, VHH1 and VHH2. PI3K: phosphoinositol 3 kinase

In a follow-up report, 16 patients with RRMM were treated at the highest dose level of 9 × 106 CAR T cells/kg [42]. Among these 16 patients, 81% responded, with 63% VGPR or CR. Median event-free survival (EFS) was 31 weeks. In addition to eradication of bone marrow myeloma cells, soft-tissue plasmacytomas were also eliminated. Negative minimal residual disease (MRD) was achieved in the bone marrow in 11 responders.

bb2121 anti-BCMA CAR T cell trials

Two types of BCMA-targeted CAR T cells from Blue Bio were reported, bb2121 and bb21217. Bb2121 contains a 2nd generation CAR with 4-1BB co-stimulation domain. Based on the bb2121 CAR structure, bb21217 contains an extra domain of bb007 which encodes a PI3K inhibitor [50](Fig. 1). This makes it possible to select out the BCMA CAR engineered T cells.

bb2121 was studied in a phase 1, single infusion dose escalation trial in patients with RRMM (CRB-401, NCT-2658929) [51]. Four dose levels were infused at 50 × 106, 150 × 106, 450 × 106, or 800 × 106 CAR-positive (CAR+) T cells. Two doses at 150 × 106 to 450 × 106 CAR+ T cells were given in the expansion phase. Patients had failed at least 3 lines of therapy. Safety was the primary endpoint. A total of 33 patients were treated with bb2121. The most common severe adverse events (SAE) were cytopenia. 25 out of 33 patients (76%) had CRS grade 1–2, and 2 at grade 3. CRES was seen in 42% of the patients, grade 1–2 in all but one who had grade 4. The ORR was 85%, with 45% CR (n = 15). Six of the CR patients relapsed. The PFS was 11.8 months (6.2–17.7). Sixteen patients were MRD negative (≤10− 4 nucleated cells). The bb2121 CAR T cells remained detectable up to 1 year after infusion. It was noted that CAR T cell expansion correlated with clinical response.

CRB-402 (NCT03274219) is a first-in-human, multi-center phase I dose escalation trial of bb21217 CAR T cells in RRMM patients [52]. The study was designed to assess the safety, efficacy and duration of effect of bb21217. Patients with strong BCMA expression on MM cells (≥ 50% BCMA expression by immunohistochemistry) were enrolled. Four dose levels are planned: 150, 450, 800, and 1200 × 106 CAR+ T cells. Safety was the primary endpoint. In the initial report, 8 patients have received a dose of 150 × 106 CAR T cells, and 7 patients were evaluable for initial (1-month) clinical response. Five of the 8 patients had CRS. As of data cut-off, 6 of 7 patients had demonstrated clinical response per IMWG criteria: 1 sCR, 3 VGPR, 2 PR. MRD negative results were confirmed by next-generation sequencing (NGS) in 3 of 3 evaluable responders. The CAR T cells were detectable at 6 months post-infusion.

LCAR-B38M: biepitopic targeting of BCMA

LCAR-B38M CAR T cells contain a CAR construct with scFv targeting two BCMA epitopes, VHH1 and VHH2 [53-55]. The BCMA-specific LCAR-B38M CAR T cells were initially tested in RRMM patients, some with extramedullary involvement [56]. In addition to objective responses, the subcutaneous mass in 2 patients was significantly shrunken after treatment. The most common adverse effects were CRS. No dose-limiting toxicities (DLTs) or treatment-related deaths occurred in these initial reports.

Results of the LCAR-B38M CAR T trial for RRMM (LEGEND-2, NCT03090659) was updated in 2018 in 57 patients [54, 55]. In this cohort, single-agent cyclophosphamide was used for lymphodepletion. The total dose of LCAR-B38M CAR T cells was divided into 3 infusions: 20, 30, and 50% of the total dose. The median CAR T cell dose was 0.5 × 106 cells/kg [range, 0.07–2.1 × 106]). In this report, 57 patients have been infused with LCAR-B38M CAR T cells. The overall response rate (ORR = PR or better) was 88%. CR rate was achieved in 39 patients (68%), VGPR was achieved in 3 patients, and PR was achieved in 8 patients. MRD was negative in 36 patients. The median time to initial response was 1 month. No clear relationship between LCAR-B38M CAR T cell dose and response was observed. BCMA expression did not correlate with clinical response. CRS was mostly grade 1 (47%) and 2 (35%); 4 patients (7%) had grade 3 CRS. No clear relationship was demonstrated between the CAR T cell dose and CRS.

In a separate report, the LCAR-B38M CAR T cells were given to 17 RRMM patients after lymphodepletion with cyclophosphamide and fludarabine [53]. In this report, two infusion schedules were compared, three infusions (n = 8) versus one infusion (n = 9) of the total CAR T dose. No differences in response were observed among the two delivery subgroups. Toxicities between the two subgroups were quite similar. Therefore, the two groups of patients (n = 17) were analyzed together. The median follow-up was 417 days. The ORR was 88.2%, including 13 sCR, 2 VGPR, 1 non-responder and 1 toxic death. Eight patients remained progression-free, 6 relapsed, and 1 progressed after VGPR. CAR T cell level correlated with disease status, with low level being associated with relapse /progression. It was also noted that anti-CAR antibody was responsible for high risk of relapse /progression. Consistent with previous observations, prior ASCT correlated with more durable response.

The outcome from the trial of 74 patients after LCAR-B38M CAR T therapy suggests that the biepitopic CAR T cells against BCMA can be an important type of immunotherapeutics for RRMM. Development of anti-CAR antibody requires attention in future trials. The second-generation human BCMA-targeted CAR may avoid this drawback.

JCARH125

JCARH125 is a CAR T cell product containing a lentiviral BCMA- targeted CAR construct with an optimized spacer and a 4-1BB co-stimulatory domain [57]. This CAR T cell product is being evaluated in a multi-center phase I/II clinical trial, the EVOLVE (NCT03430011) trial in RRMM patients who have exhausted therapies [58]. A single dose of JCARH125 is scheduled on day 1 in each cohort. Dose escalation was done with the first 2 dose levels at 50 and 150 × 106 CAR+ T cells. In this early report, 19 patients have been enrolled and leukapheresis completed. Among these, 13 patients received JCARH125 cells. Eight patients were evaluable for toxicities, 6 of which had grade I/II CRS. Three of the 8 patients had CRES. Objective response was seen in all 8 patients. However, the follow-up was very short, longer follow-up and more patients are needed. The study remains active.

CT053

CT053 are genetically modified T cells with a human BCMA scFv and 4-1BB costimulatory motif. CT053 CAR T cells were studied in a multi-center investigator-initiated clinical trial in patients with RRMM (NCT03915184). BCMA expression was seen on MM cells in all patients. A single dose of CAR T cells is planed and a second dose is allowed as clinically indicated. In a recent report, 16 patients received CT053 cells [59]. CRS was seen in 3 patients, no CRES nor DLT was observed. With median follow-up time of 8 (4 to 36) weeks, ORR was 100% in the 13 evaluable patients. The CT053 CAR-T cells were detectable up to 4–6 months in 11/13 patients. CT053 BCMA-targeted CAR T cells appeared to have the potential for further development for RRMM.

MCARH171

MCARH171 has a second generation, human derived BCMA targeted CAR which contains a 4-1BB domain and a truncated epidermal growth factor receptor safety system [60]. MCARH171 CAR T cells were studied in a phase I clinical trial and the final result of the phase I trial was updated at the 2018 ASH annual meeting [61]. The trial followed a standard 3 + 3 design. The 11 enrolled RRMM patients received one of the following doses per cohort: (1) 72 × 106, (2) 137 × 106, (3) 475 × 106, (4) 818 × 106 viable CAR+ T cells. Safety and efficacy as well as the persistence of CAR T cells were evaluated. The ORR was 64% and the median duration of response was 106 days (range: 17 to 235 days). Results indicated the peak amplification and persistence of MCARH171. Durable clinical responses were dose dependent. Patients treated on the first two cohorts (≤150 × 106 CAR T cells) had a lower peak peripheral blood expansion (mean 14,098 copies/μL; N = 6), compared with patients in the cohort 3 and 4 (≥450 × 106 CAR T cells; N = 5), where the average peak expansion was 90,208 copies/μL (p < 0.05). The response rate was 100% in 5 patients receiving higher doses (450 × 106). The duration of response was also related to the cell dose, with 3 of 5 patients (60%) treated in the cohorts ≥450 × 106 had a clinical response lasting > 6 months, compared with only 1 of 6 (16.7%) patients who received lower doses. Two patients continued to respond with VGPR during follow-up of 7.5 and more than 10 months respectively. As shown in this study, MCARH171 has acceptable safety and no DLT has been reported. The dose-response relationship with toxicity was not clearly observed. However, a dose-response relationship was observed with promising clinical efficacy at dose levels of ≥450 × 106 CAR T cells.

CT103A

Another BCMA – targeted CAR T cell product contained a lentiviral CAR with a murine anti-BCMA scFv and CD28z domain (BRD015) [62]. In the phase I trial, the target dose levels of anti-BCMA CAR T cells ranged between 5.4 ~ 25.0 × 106 cells /kg. Among 28 evaluable patients, 26 achieved remission. Twenty-two patients had strong BCMA expression (> = 50% expression rate) on MM cells, whereas 6 cases had weaker BCMA expression. ORR was 87% for strong expressors, 100% for weak expressors. The OS for the strong expressors was not yet reached at the time of report, whereas the OS for the weak expressors was 206 days (p = 0.0468). It was observed that higher peak level of CAR T cells in the blood was associated with better responses. A patient with POEMS syndrome responded to the BCMA-targeted CAR T therapy [62, 63].

This BRD015 product has the murine BCMA epitope. Human blocking antibodies against murine antigen makes it ineffective for the CAR T cell re-infusion. To conquer this issue, a novel BCMA-targeted CAR-T, CT103A, was engineered, which contains a lentiviral vector with a fully human BCMA scFv and a 4-1BB co-stimulatory domain. In the latest update, a single-center, dose-escalating phase I trial of CT103A in patients with RRMM enrolled 9 patients, including 3 patients who have relapsed after the murine BCMA-targeted CART, BRD015, therapy (clinical trial registration: ChiCTR1800018137) [64]. The CT103A CAR T cells were administered following a standard 3 + 3 dose-escalation design, with three doses at 1, 3, and 6 × 106 cells/kg. The conditioning chemotherapy regimen was cyclophosphamide and fludarabine. The median prior lines of therapy of the enrolled patients were 4 (range 3–5). The ORR was 100%. The response was observed within 14 days. In the first two dose levels, the CRS was mild. It is intriguing that among the three patients who relapsed after the murine BCMA CAR T therapy, two patients achieved CR and one patient achieved VGPR after CT103A therapy. Further studies are ongoing for this promising humanized BCMA- targeted CAR T therapy in RRMM.

Cart-BCMA

Another autologous T cell product, CART-BCMA, containing a fully human, BCMA-specific CAR with CD3ζ and 4-1BB signaling domains was studied in a phase I clinical trial (NCT02546167) for RRMM patients (Table 1). This was a single-center standard 3 + 3 dose-escalation study with 3 cohorts, with doses ranging between 1.0 to 50 × 107 total CART-BCMA cells. The cohort I group had no lymphodepletion therapy, whereas cohort 2 and 3 had cyclophosphamide for lymphodepletion. In the updated reports, 25 subjects were enrolled [65]. Among these, 8 had severe CRS and 3 had severe CRES respectively. One patient died with candidemia, severe CRS and encephalopathy. Responses were seen in all cohorts, including 5 PR, 5 VGPR, and 2 CR. There was correlation of responses and CART-BCMA expansion with CD4/CD8 T cell ratio and frequency of CD45RO−CD27+CD8+ T cells in the initial leukapheresis product. In this study, higher dose of CART-BCMA cells without lymphodepleting chemotherapy were tested in cohort 1. The CAR T cells were shown to be clinical active in heavily pretreated patients with MM in all cohort groups, with or without lymphodepletion chemotherapy. More subjects are being enrolled in expansion cohorts.

P-BCMA-101

P-BCMA-101 encodes a CARTyrin that can target BCMA. CARTyrin is a non-immunoglobulin- based scaffold Centyrin molecule produced with a novel non-viral piggyBac transposon-based delivery system [66]. The Centryins are fully human with high binding affinities. These Centyrins are therefore less immunogenic. P-BCMA-101 is hence a novel CAR T product that can target BCMA. Since this is not based on a viral vector system, it is less costly. This approach also allowed manufacture of CAR T cells with predominantly more favorable stem cell memory T phenotype (TSCM). P-BCMA-101 is being studied in a phase 1, 3 + 3, single-administration dose escalation trial in patients with RRMM (NCT03288493) (Table 1). In the update at the 2018 ASH meeting, 12 high-risk heavily pre-treated patients with RRMM have been treated with P-BCMA-101 CAR-T cells in 3 cohorts [67]. Grade 2 CRS was observed in one patient. The treatment was well tolerated with no death and no CRES. DLT was not yet observed. These were consistent with the improved therapeutic index for these Tscm CAR T cells. The ORR was 83% in the evaluable patients.

BCMA CAR T cells with defined formulation (CD4+:CD8+ =1:1)

It has been reported that CAR T cells with defined proportion of CD4 and CD8 cells may have advantages [68]. A first-in-human phase I trial of BCMA CAR T cells with defined formulation was done in RRMM patients (NCT03338972) [69]. In this study, CD8+ and CD4+ T cells were isolated, enriched and cryopreserved separately. These T cells were transfected with a fully human BCMA scFv -containing CAR via a lentiviral vector. The cells were expanded in the culture and the cell product for infusion was formulated to contain equal number of CD4+ and CD8+ BCMA CAR T cells. To facilitate tracking of the BCMA-targeting CAR T cells, a truncated non-functional human epidermal growth factor receptor (EGFRt) was inserted into the CAR cassette. The first dose level was 5 × 107 EGFRt+ T cells (cohort A, n = 5). At the time of this report, 2 patients have been enrolled in the cohort B at the dose of 15 × 107 EGFRt+ T cells. These seven patients have received a median of 8 prior regimens (range 6 to 11), including autologous stem cell transplant (SCT) (71%) and allogeneic SCT (43%). All seven patients had a response at 28 days. The EGFRt+ CAR T cells were still detectable 90 days after infusion. The median survival was 16 wks (range 2 to 26 wks) with all patients alive. One relapsed patient was found to become BCMA negative in the myeloma cells. At the time of the update, DLT and CRES have not been observed. Only grade 2 or lower CRS was reported. In summary, BCMA-targeted CAR T cells with a 1:1 ratio of CD4+:CD8+ were well tolerated and were shown to be effective at total cell doses as low as 5 × 107. This approach deserves further investigation. Longer follow-up is needed.

CD19- targeted CAR-T cell trials for MM

Tisagenlecleucel has been approved for advanced B cell acute lymphoid leukemia (ALL) and diffuse large cell lymphoma [24–28, 70–77]. CD19 expression is lost in plasma cells [78]. However, minor subsets of myeloma cells with unique propagating properties were found to express low levels of CD19. CR was reported in a case of RRMM patient after treatment with tisagenlecleucel cells (CTL019) following high dose melphalan (140 mg /m2) and autologous stem-cell transplantation (ASCT) [79]. The dramatic response was surprising and intriguing since there lacked CD19 expression in 99.95% of myeloma cells in this patient. The durable response continued even after disappearance of CTL019 cells in the blood (day 2 to day 47 post CTL019 infusion), suggesting that the sustained response did not require the persistent presence of the CAR T cells. This therapeutic approach of RRMM is being further investigated in a clinical trial (NCT02135406). A recent update detailed treatment of 10 patients with RRMM with CTL019 following high-dose melphalan and ASCT [32]. Eleven patients were enrolled at the time of the report, though T cells from one patient failed to proliferate to the required number. These 10 patients had previously undergone ASCT but progressed within 1 year. The treatment with ASCT + CTL019 was safe and feasible. There was no severe CRS, correlating with low concentration of B cells in the peripheral blood of these patients. The most toxicity observed was attributed to ASCT. In this study, 4 patients achieved objective response (sCR, n = 1, VGPR, n = 1, PR, n = 2), and other 6 objects remained progression free. Durable progression- free survival (PFS) after ASCT + CTL019 was reported in 2 of 10 subjects. Peak frequency of CTL019 cells in bone marrow and emergence of immune responses against the stem-cell antigen Sox2 correlated with the favorable clinical outcome. The two patients with the best responses had significant elevation of anti-Sox2 antibodies. Sox2 expression has been shown to correlate with the myeloma-propagating capability in myeloma cell lines [80-82]. It appears that CTL019 cells in this setting induced the immune responses against Sox2. This intriguing phenomenon may be related to epitope- spreading since CD19 and Sox2 are co-expressed in myeloma-propagating cells (MPC) which are targeted by CTL019 cells [32]. This possible MPC targeting mechanism was further investigated ex vivo. It was demonstrated that MPCs as colony forming cells were reliably targeted by the combination of CTL019 and anti-BCMA CAR T cells, since CTL019 and anti-BCMA CAR T cells each can target a subset of MPCs. These observations suggest that the surface immunophenotypes of MPCs are heterogeneous, some resembling CD19+ B cells, some resembling BCMA+ plasma cells.

Cocktail strategy of BCMA- and CD19- targeted CAR T trials for MM

To combat antigen loss and resistance of CAR T therapies, combining CAR T cells with different target in a cocktail infusion has been reported [83-85]. One report evaluated the safety and efficacy of combined infusion of CD19- and BCMA- targeted CAR T Cells for RRMM (NCT 03196414) [86]. The CARs contained anti- BCMA or anti-CD19 single chain variable fragment (scFv), the cytoplasmic portion of the OX40 and CD28 costimulatory moiety, and the CD3z T-cell activation domain. These are third generation CARs. CAR T-19 cells were infused on day 0 at the dose of 1 × 107/kg, and CAR T-BCMA cells were given as split-dose infusions (40% on day 1 and 60% on day 2). Two of the 8 patients received haplo-identical BCMA targeted CAR T cells. All 8 patients had CRS. Due to the several compounding factors (such as haplo-identical T cells) in this trial, it is difficult to evaluate the outcome of this small combined CAR T clinical trial in RRMM.

Using the same strategy of cocktail CAR T therapies, the same group reported infusion of BCMA- and CD19-targeted CAR T cells into RRMM patients on day 14 to day 20 following autologous stem cell transplantation (SZ-MM-CART02 study, NCT 03455972) [87]. The CAR T cells were thereby used as post-transplant consolidation therapy. The median follow-up was only 3 months (2–11 months) at the time of report with 9 patients enrolled in cohort 1. All patients were positive for BCMA and negative for CD19 expression on MM cells. The ORR was 100%, with 3 CR, 6 VGPR after CAR T therapy. MRD negativity in BM was 37.5% after transplantation which increased to 66.7% after CAR T therapy. All patients experienced mild grade 1–2 CRS. There was more than 1000-fold expansion observed at peak level. A 100- fold increase was reported in a similar cohort. This phenomenon is intriguing, since it has been postulated that CD19 CAR may enhance T cell expansion in vivo at the expenses of B cell hypoplasia. Presence of the CD19- targeted CAR T cells in the cocktail may be responsible for the significant expansion of the CAR T cells. As discussed above, CTL019 infusion after ASCT can induce CR. It is therefore not clear what proportion of the response in the cocktail CAR T therapy after ASCT in this study was attributable to the cocktail vs to the CD19-targeted CAR T. Theoretically, the strategy of cocktail infusion of CAR T cells may inherently lead to higher toxicities due to simultaneous targeting of two or more antigens /epitopes and therefore enhanced T cell activation and tumor lysis. From these early studies of cocktail CAR T cell therapy, it is still difficult to make conclusions. More studies with larger sample sizes are needed. CRS as the main AE is better recognized and managed [18-20]. Recent recognition of roles of macrophage and monocyte in CRS and CRES may lead to earlier and better therapy as well as possible prophylaxis.

SLAMF7-targeted CAR T trials

SLAMF7 (also known as CS1) is an antigen abundantly expressed on the surface of normal and neoplastic plasma cells and NK cells as well as a small subset of lymphocytes [88-94]. Elotuzumab (huLuc63) is a monoclonal antibody against SLAMF7. In clinical trials, elotuzumab in combination with IMiDs and proteasome inhibitors has been shown to be highly effective in treating RRMM [95]. Elotuzumab has gained FDA approval for the treatment of multiple myeloma (MM).

In a preclinical study, a retroviral construct of a SLAMF7-specific CAR was engineered and inserted into primary human T cells [96]. The SLAMF7–CAR T cells were tested on human MM tumor cells in vitro, ex vivo, and in orthotopic MM xenograft mouse models. The CAR T cells showed enhanced cytotoxicity against MM cell lines as well as primary MM cells. The SLAMF7 CAR T cells significantly prolonged survival of the mice xenografted with human MM.1S and IM9 myeloma cells. Another study reported construction of SLAMF7 CAR from the huLuc63 monoclonal antibody (elotuzumab) scFv [33]. The SLAMF7 CAR T cells induced rapid cytolysis of primary myeloma cells from patients with untreated and RRMM. The SLAMF7-CAR T cells were also shown to be effective in elimination of medullary and extramedullary myeloma in a murine xenograft model. These preclinical studies have paved the way for clinical translation. Two clinical phase I trials of SLAMF7-targeted CAR T cells are ongoing (NCT03710421 and 03778346).

An universal “off-the-shelf” allogeneic SLAMF7-specific CAR T cell product, UCARTCS1, was developed using TALEN-targeted gene editing [97, 98]. UCARTCS1 was tested in vitro and in mouse models against MM cell lines and primary human myeloma cells [99]. This study clearly showed that the UCARTCS1 cells could specifically target SLAMF7 and lyse MM cells both in vitro and in vivo. UCARTCS1 has been cleared by FDA for phase I study in MM patients.

A dual-target CAR construct targeting both BCMA and SLAMF7 (CS1) was also studied in vitro and in mouse models [29]. The compound CAR T cells (cCAR) contain two complete and independent CARs. the cCAR T cells were shown to have sustained in vivo activity against the MM1S cell line and induced superior survival in a mixed cell mouse model. Clinical trials are needed for these dual-target CAR T cells.

CD138-targeted CAR T cell trial

CD138 is a surface molecule highly expressed on MM cells [100-103]. CD138 plays a significant role in the development and/or proliferation of plasma cells. CD138- targeted therapy with radioimmunoconjugate appears to be a novel approach for MM [104, 105]. A preclinical study evaluated CAR T cells targeting CD138 in vitro and in an animal model [106]. The study showed that the CAR T cells had no off-tumor toxicities. A CD138- directed CAR has been constructed with a 41BB domain [34]. The CD138- targeted CAR T cells (CART-138) were studied in five patients with RRMM in a dose-escalation phase I study (NCT01886976) (Table 2). The CD138-targeted CAR T cells were well tolerated. Four of the five patients achieved stable disease for 3 to 7 months and the CAR-T cells remained detectable by flow cytometry for more than 3 months.

Table 2

Non-BCMA-targeted CAR T clinical trials in multiple myeloma

NCT number (reference)	Target	Phase	Dosage	Number of patients	Responses
NCT02135406 (32)	CD19	1	1.1–6.0 × 108 CAR+T cells	10	VGPR: 6; PR: 2; PD: 2
NCT01886976 (34)	CD138	1/2	0.44–1.51 × 107 CAR+ T cells/kg	5	SD: 4; PD: 1
NCT00881920 (35)	κ light chain	1	0.2–2.0 × 108 CAR+ T cells/m2	16 (7 MM)	4 SD of 7 MM

Abbreviations: BCMA B cell maturation antigen, CAR Chimeric antigen receptor, PD Progressive disease, PR Partial response, VGPR Very good partial response, SD Stable disease, MM Multiple myeloma

Light chain-targeted CAR T trial

Malignant B cells are frequently light chain-restricted cells. Light chain-specific CAR (κ.CAR and lambda CAR) T cells have been engineered [35, 107]. Sixteen patients with relapsed or refractory κ + non-Hodgkin lymphoma/chronic lymphocytic leukemia (NHL/CLL) or multiple myeloma (MM) were enrolled in a phase I clinical trial of autologous κ.CAR T cells (κ.CARTs) (NCT00881920). Other treatments were discontinued in 11 of the 16 patients at least 4 weeks prior to CAR T cell infusion. The κ.CART infusion doses ranged from 0.2 to 2 × 108 κ.CARTs /m2. κ.CART expansion was observed. Two of the 9 patients with relapsed NHL or CLL achieved CR and 1 had a PR. Four of 7 patients with RRMM remained stable for 2–17 months. The κ.CARTs were well tolerated [35].

Conclusions and future perspectives

BCMA- targeted CAR T cells have shown high response rate in patients with RRMM with even high risk features [108]. A variety of BCMA targeted CAR T products are in active clinical development. BCMA- targeted CAR T cell product is expected to be approved for clinical therapy of RRMM soon. CAR T cells targeting CD138, CS1 glycoprotein antigen (SLAMF7) and light chains appear to be promising. CD19- targeted CAR T cells in conjunction with autologous stem cell transplantation also showed activity in RRMM. Dual- target CAR T cells are in clinical trials for RRMM. Advances in cellular immunotherapy will likely lead to significant improvement in MM therapy [30, 109–114]. In addition, most of these antigen targets are also being studied for construction of antibody-drug conjugates and bispecific antibodies [115-118]. It may be possible to combine these either concurrently or sequentially to enhance clinical efficacies.