User login
In multiple myeloma, survival has been very significantly improved by immunomodulatory drugs, proteasome inhibitors, and CD38-targeting antibodies. Despite these advances, multiple myeloma, which is characterized by malignant proliferation of clonal plasma cells in bone marrow, remains an incurable plasma cell disorder with near-certain relapse after successful treatment. Prognosis for patients who develop triple-class refractory disease is poor, with less than 1-year survival. The substantial unmet therapeutic need extends further to other poor survival multiple myeloma populations that include newly diagnosed patients with high cytogenic risk profiles and those with early relapse after first-line therapy. For all of these, interest in drugs with novel mechanisms of action is naturally high.
More specific, less toxic
Post allogeneic hematopoietic stem-cell transplantation and donor lymphocyte infusion sustained remissions reflect a graft-versus-myeloma effect mediated by donor T cells.1 The substantial morbidity and mortality associated with graft-versus-host disease and opportunistic infections, however, have spurred searches for alternative, more specific, and less toxic T-cell therapies with stronger antitumor activity.
Chimeric antigen receptors (CARs)
In CAR T-cell therapies for multiple myeloma, autologous T cells are harvested from the patient and reprogrammed to target multiple myeloma cells through the introduction of genes that encode CARs, which are fusion proteins coupling an antigen-recognition moiety and a transmembrane-spanning element to a T-cell activation domain (typically CD3 zeta [CD247]). The T cells are then expanded and reinfused to the patient following a lymphodepletion regimen. Five strategies using autologous CAR T cells are currently approved for diffuse large B-cell lymphomas, acute lymphoblastic leukemia, multiple myeloma, and other hematologic malignancies. Notably, in patients with heavily pretreated multiple myeloma, CAR T cells have demonstrated impressive activity.
BCMA-targeting CAR T cells
The B-cell maturation antigen (BCMA; TNFRSF17), which plays an important role in the survival of long-lived plasma cells in bone marrow, is an attractive target for CAR T-cell therapy because it is expressed on normal and malignant plasma cell surfaces and by mature B cells. When ligands (TNFSF 13B/TNFSF13) bind to BCMA expressed on multiple myeloma cell surfaces, survival and proliferation pathways and drug resistance are activated.
High-quality responses have been demonstrated in several trials of anti-BCMA CAR T cells, which kill multiple myeloma cell lines and primary multiple myeloma cells through degranulation of T cells and lysis of tumor cells, even those with low BCMA expression. Based on efficacy in triple-class exposed multiple myeloma that compared favorably to conventional care with improved health-related quality of life, the U.S. Food and Drug Administration gave breakthrough designation to ciltacabtagene autoleucel in December 2019 and approval for idecabtagene vicleucel in March 2021.
Idecabtagene vicleucel
Idecabtagene vicleucel expresses a murine BCMA-targeting single-chain variable fragment with a 4-1BB costimulatory motif. The phase 2 KarMMa study2 evaluated idecabtagene vicleucel (target dose of 450 × 106 CAR T cells; range 150 × 106 to 450 × 106) activity in 128 patients with triple-class exposed multiple myeloma. Partial responses or better were observed in 94 of 128 patients (73%) (95% confidence interval, 66-81); 42 (33%) had a complete response or better (95% CI, 25-41), with a median progression-free survival of 8.8 months (95% CI, 5.6-11.6). Outcomes were improved in the highest fixed-dose group, with partial response or better in 81% (44 of 55), complete response or better in 39% (21), and median overall survival of 12.1 months (95% CI, 8.8-12.3). Patients with high-risk cytogenetic profiles, extramedullary disease, and high tumor burden also had deep and durable responses. Outcomes were less favorable in patients with revised International Staging System stage 3 disease.
Ciltacabtagene autoleucel
Ciltacabtagene autoleucel, a 4-1BB–based CAR T-cell therapy with two BCMA-targeting domains, confers high-avidity binding. In the phase 1b/2 CARTITUDE-1 study, conducted in the United States and Europe, preliminary results in 97 patients showed a 97% response rate with ciltacabtagene autoleucel (target dose 0.75 × 106 CAR T cells per kg), and in 65 patients, a complete response (67%). Progression-free survival at 12 months was 77% (95% CI, 66-84) and overall survival was 89% (95% CI, 80-94).3
In the phase 1 LEGEND-2 study4 that was conducted at four sites in China among less heavily pretreated multiple myeloma patients, while all used the same CAR construct, sites used variable conditioning regimens (split versus single). In the site using cyclophosphamide as the lymphodepletion therapy and three split CAR T-cell infusions, partial response or better was achieved in 50 patients (88%) with a median of three prior therapy lines. The complete response rate was high (74%) and minimal residual disease negativity was reached in 39 patients (68%). Median progression-free survival was 19.9 months (95% CI, 9.6-31.0), but 28.2 months among those with complete responses (95% CI, 19.9-not estimable). Median overall survival was also favorable at 36.1 months (95% CI, 26.4-not estimable); it was 35.0 months-not estimable among patients with complete responses. Results from the other three sites were comparable.
Noteworthy among other BCMA-targeting CAR T-cell products in earlier stages of clinical development is orvacabtagene autoleucel, which has a fully human BCMA-specific binding domain. At higher doses (300 × 106 to 600 × 106 CAR T cells) among 62 patients with triple-class–exposed multiple myeloma in the EVOLVE trial, 92% had a partial or better response, with complete responses or better in 36%, all with an encouraging safety profile.
BCMA-targeting CAR T cell toxicity
While van de Donk, Usmani, and Yong, in their review1 note a lack of evidence of off-target toxicity with BCMA-targeting CAR T-cell therapy in clinical studies so far, they do point to several clinical syndromes (cytokine release syndrome, infections, respiratory failure, neurotoxicity, pulmonary aspergillosis, gastrointestinal hemorrhage) caused by cytokines produced during CAR T-cell expansion and to cytopenias and infections arising from prior treatment, bridging therapy, and lymphodepleting conditioning. Deaths attributed to treatment in the above-mentioned trials underscore the need for careful monitoring and early intervention.
Cytokine release syndrome
In the BCMA-targeting CAR T-cell therapy studies, the frequency of cytokine release syndrome varies widely from 17% to 95% but is generally attributed to CAR T-cell activation and is associated with increased serum ferritin concentrations, high c-reactive proteins, and proinflammatory cytokines. High tumor load, in multiple myeloma patients receiving CD19-targeting CAR T cells, was associated with a higher incidence of severe cytokine release syndrome. In a small number of patients, macrophage activation syndrome and hemophagocytic lymphohistiocytosis, the most aggressive variants of cytokine release syndrome, are caused by severe immune activation and lead to multiorgan dysfunction.
Neurotoxicity
Immune effector cell–-associated neurotoxicity syndrome (ICANS) symptoms, in multiple myeloma patients treated with BCMA-targeting CAR T cells, may include delirium, transient confusion, aphasia, lethargy, tremor, dysgraphia, seizures, cerebral edema, and rarely, posterior reversible encephalopathy syndrome.1 While the pathophysiology of CAR T cell–related neurotoxicity is not well understood, high tumor load, higher peak concentrations of CAR T cells, and more severe cytokine release syndrome are more common in patients with severe neurotoxicity. “The frequency of neurotoxicities,” Dr. Yong noted in an interview, “has been reduced by steps taken to mitigate these risk factors.”
High interest in phase I study
A phase I study presented in Blood has attracted interest because the novel BCMA-targeting CAR agent (CT103A) being tested is fully human.4 In an accompanying editorial, Lee and Yong note that doubt for any real potential for durable CAR T therapy responses in multiple myeloma is raised by the poor persistence of multiple myeloma CAR T cells in multiple myeloma patients.3
In the earliest trials of BCMA CARs, while reported rates of objective antimyeloma responses were in the approximately 33%-88% range among patients with relapsed/refractory multiple myeloma (RRMM), persistence was typically 6 months or less. Lee and Yong point out, however, that while correlation between persistence and duration of response (DOR) has been variable, median persistence was 308 days in the phase I study. Wang and colleagues, the phase I study authors, state that levels of CAR T-cell proliferation and duration of cellular persistence may be determinants of DOR in CAR T therapy for multiple myeloma. They observe that the multiple mechanisms potentially responsible for the inability of some CAR T cells to survive in vivo, may include antigen escape, T-cell intrinsic mechanisms, tumor microenvironment–mediated suppression, and host anti-CAR immunity. CARs with humanized or fully human single-chain variable fragments (scFvs), prior studies suggest, may retain antitumor activity through bypassing potential host anti-CAR immunogenicity.
In the study, CT103A, a fully human scFv, was tested in an open-label, single-arm design for safety and preliminary efficacy in 18 patients (8 female; median age 53.5 years) with RRMM (at least three lines of prior therapies including a proteasome inhibitor and an immunomodulatory agent) who had undergone leukapheresis and had received lymphodepletion chemotherapy with fludarabine and cyclophosphamide. Four patients (22.2%) had been treated previously with murine anti-BCMA CAR T cells. Safety and tolerability (including dose-limiting toxicity) were the primary endpoints, with efficacy and pharmacokinetics secondary.
Rapid responses
Two weeks after infusion, the overall response rate (ORR) was 77.8% (14 of 18) and by 1 month it was 88.9% (16 of 18). Eventually, all responded and 72.2% (13 of 18) achieved a complete response (CR) or stringent complete response (sCR). All 17 patients evaluated for minimum residual disease (MRD) in bone marrow were MRD-negative at 10-4 nucleated cells by flow cytometry within 1 month. Median DOR was 325 days (range, 7-573 days) for all patients and 412 days (range, 213-573 days) for the 13 with CR/sCR. CAR transgenes were detectable at the cutoff date in 77.8% of patients, with a median CAR transgene persistence of 307.5 days.
During follow-up, four deaths were reported, including one patient with persistent sCR (sudden severe infection). Progression-free survival (PFS) and overall survival (OS) rates at 1 year were 58.3% and 75%, respectively. Extramedullary myeloma was associated with a shortened PFS (79.1% versus 20.0%, P = .015), but not OS (79.1% versus 60.0%, P = .328) at 1 year.
All patients experienced grade 3 or higher adverse events, most of which were expected hematologic effects of lymphodepleting chemotherapy and CT103A infusions. Grade 1 and 2 cytokine release syndromes occurred in 70.6% patients (17 of 18), with 1 grade 4 event (5.9%). The patients receiving a dose of up to 3.0 × 106 CAR+ T cells/kg required less treatment of cytokine release syndrome than the patients who received a dose of 6.0 × 106 CAR+ T cells/kg. No immune effector cell–associated neurotoxicity syndrome was observed. Antidrug antibody positivity occurred in only 1 patient.
Two characteristics of CT103A may contribute to its long persistence, stated study senior author Jianfeng Zhou, MD, PhD, chairman and professor of the department of hematology, Tongji Hospital in Wuhan, China. “One is the reduced immunogenicity achieved by the fully human construct; another is the relatively low binding affinity of the CAR binder. Notably, four patients who previously received murine BCMA CAR were included and still benefit from CT103A. It demonstrates the possibility of retreatment with a different CAR.” Dr. Zhou also emphasized that the lack of ICANS in the entire cohort reflects the excellent safety profile of CT103A.
The editorial commentary in Blood by Lydia Sarah Hui Lee, MD, and Kwee L. Yong, PhD, underscored impressive responses to CT103A, specifically to the median time to response of 15 days, the 100% ORR, and the not reached median progression-free survival at 394 days).5 The best results in other published nonhuman BCMA CAR T-cell trials, they note, were about 1 month (time to response), approximately 33%-88% (ORR), and median progression-free survival of 7-15 months.
Immune responses, Dr. Yong said in an interview, can guide subsequent treatment. “For example, if a patient previously exposed to BCMA CAR T cells in which the construct is either chimeric or humanized, but retains some murine elements, and had detectable antimurine antibodies, we may aim for a fully human one if we are considering treating with a different BCMA CAR T-cell product.” She added, “On the other hand, a similar patient whose serum did not contain such antibodies may be a candidate for a humanized product that retained some murine elements.”
Wang and colleagues concluded, “Altogether, CT103A is safe and highly active in patients with relapsed/refractory multiple myeloma and can be developed as a promising therapy for relapsed/refractory multiple myeloma.”4 An ongoing multicenter phase II trial with single-arm design is recruiting 100 patients. The infusion dosage, suggested by the phase I trial, is 1 × 106 cells/kg. Endpoints include efficacy and safety.
Improving CAR T
Optimizing CAR design and adapting manufacturing processes to generate cell products enriched for T-cell subsets, such as early memory cells, are among strategies being explored to improve CAR T effectiveness.1 Also, dual-antigen targeting to interdict antigen escape and rational combination treatments to enhance persistence are under investigation, along with efforts to improve CAR T-cell therapy safety (for example, incorporation of a suicide gene safety system). They note further that several groups are researching use of induced pluripotent stem cells to generate large quantities of off-the-shelf CAR T-cell immunotherapies that would circumvent the complex, costly, and time-consuming process of manufacturing patient-specific autologous CAR T cells.
References
1. van de Donk N et al. Lancet Haematol. 2021 June;8(6):e446-61.
2. Munshi NC et al. N Engl J Med 2021; 384:705-716.
3. Berdeja JG et al. The Lancet. 2021 July; 398:314-24.
4. Wang D et al. Blood. 2021 May;137(21):2890-901.
5. Lee L and Yong K. Blood. 2021 May;137(21):2859-60.
In multiple myeloma, survival has been very significantly improved by immunomodulatory drugs, proteasome inhibitors, and CD38-targeting antibodies. Despite these advances, multiple myeloma, which is characterized by malignant proliferation of clonal plasma cells in bone marrow, remains an incurable plasma cell disorder with near-certain relapse after successful treatment. Prognosis for patients who develop triple-class refractory disease is poor, with less than 1-year survival. The substantial unmet therapeutic need extends further to other poor survival multiple myeloma populations that include newly diagnosed patients with high cytogenic risk profiles and those with early relapse after first-line therapy. For all of these, interest in drugs with novel mechanisms of action is naturally high.
More specific, less toxic
Post allogeneic hematopoietic stem-cell transplantation and donor lymphocyte infusion sustained remissions reflect a graft-versus-myeloma effect mediated by donor T cells.1 The substantial morbidity and mortality associated with graft-versus-host disease and opportunistic infections, however, have spurred searches for alternative, more specific, and less toxic T-cell therapies with stronger antitumor activity.
Chimeric antigen receptors (CARs)
In CAR T-cell therapies for multiple myeloma, autologous T cells are harvested from the patient and reprogrammed to target multiple myeloma cells through the introduction of genes that encode CARs, which are fusion proteins coupling an antigen-recognition moiety and a transmembrane-spanning element to a T-cell activation domain (typically CD3 zeta [CD247]). The T cells are then expanded and reinfused to the patient following a lymphodepletion regimen. Five strategies using autologous CAR T cells are currently approved for diffuse large B-cell lymphomas, acute lymphoblastic leukemia, multiple myeloma, and other hematologic malignancies. Notably, in patients with heavily pretreated multiple myeloma, CAR T cells have demonstrated impressive activity.
BCMA-targeting CAR T cells
The B-cell maturation antigen (BCMA; TNFRSF17), which plays an important role in the survival of long-lived plasma cells in bone marrow, is an attractive target for CAR T-cell therapy because it is expressed on normal and malignant plasma cell surfaces and by mature B cells. When ligands (TNFSF 13B/TNFSF13) bind to BCMA expressed on multiple myeloma cell surfaces, survival and proliferation pathways and drug resistance are activated.
High-quality responses have been demonstrated in several trials of anti-BCMA CAR T cells, which kill multiple myeloma cell lines and primary multiple myeloma cells through degranulation of T cells and lysis of tumor cells, even those with low BCMA expression. Based on efficacy in triple-class exposed multiple myeloma that compared favorably to conventional care with improved health-related quality of life, the U.S. Food and Drug Administration gave breakthrough designation to ciltacabtagene autoleucel in December 2019 and approval for idecabtagene vicleucel in March 2021.
Idecabtagene vicleucel
Idecabtagene vicleucel expresses a murine BCMA-targeting single-chain variable fragment with a 4-1BB costimulatory motif. The phase 2 KarMMa study2 evaluated idecabtagene vicleucel (target dose of 450 × 106 CAR T cells; range 150 × 106 to 450 × 106) activity in 128 patients with triple-class exposed multiple myeloma. Partial responses or better were observed in 94 of 128 patients (73%) (95% confidence interval, 66-81); 42 (33%) had a complete response or better (95% CI, 25-41), with a median progression-free survival of 8.8 months (95% CI, 5.6-11.6). Outcomes were improved in the highest fixed-dose group, with partial response or better in 81% (44 of 55), complete response or better in 39% (21), and median overall survival of 12.1 months (95% CI, 8.8-12.3). Patients with high-risk cytogenetic profiles, extramedullary disease, and high tumor burden also had deep and durable responses. Outcomes were less favorable in patients with revised International Staging System stage 3 disease.
Ciltacabtagene autoleucel
Ciltacabtagene autoleucel, a 4-1BB–based CAR T-cell therapy with two BCMA-targeting domains, confers high-avidity binding. In the phase 1b/2 CARTITUDE-1 study, conducted in the United States and Europe, preliminary results in 97 patients showed a 97% response rate with ciltacabtagene autoleucel (target dose 0.75 × 106 CAR T cells per kg), and in 65 patients, a complete response (67%). Progression-free survival at 12 months was 77% (95% CI, 66-84) and overall survival was 89% (95% CI, 80-94).3
In the phase 1 LEGEND-2 study4 that was conducted at four sites in China among less heavily pretreated multiple myeloma patients, while all used the same CAR construct, sites used variable conditioning regimens (split versus single). In the site using cyclophosphamide as the lymphodepletion therapy and three split CAR T-cell infusions, partial response or better was achieved in 50 patients (88%) with a median of three prior therapy lines. The complete response rate was high (74%) and minimal residual disease negativity was reached in 39 patients (68%). Median progression-free survival was 19.9 months (95% CI, 9.6-31.0), but 28.2 months among those with complete responses (95% CI, 19.9-not estimable). Median overall survival was also favorable at 36.1 months (95% CI, 26.4-not estimable); it was 35.0 months-not estimable among patients with complete responses. Results from the other three sites were comparable.
Noteworthy among other BCMA-targeting CAR T-cell products in earlier stages of clinical development is orvacabtagene autoleucel, which has a fully human BCMA-specific binding domain. At higher doses (300 × 106 to 600 × 106 CAR T cells) among 62 patients with triple-class–exposed multiple myeloma in the EVOLVE trial, 92% had a partial or better response, with complete responses or better in 36%, all with an encouraging safety profile.
BCMA-targeting CAR T cell toxicity
While van de Donk, Usmani, and Yong, in their review1 note a lack of evidence of off-target toxicity with BCMA-targeting CAR T-cell therapy in clinical studies so far, they do point to several clinical syndromes (cytokine release syndrome, infections, respiratory failure, neurotoxicity, pulmonary aspergillosis, gastrointestinal hemorrhage) caused by cytokines produced during CAR T-cell expansion and to cytopenias and infections arising from prior treatment, bridging therapy, and lymphodepleting conditioning. Deaths attributed to treatment in the above-mentioned trials underscore the need for careful monitoring and early intervention.
Cytokine release syndrome
In the BCMA-targeting CAR T-cell therapy studies, the frequency of cytokine release syndrome varies widely from 17% to 95% but is generally attributed to CAR T-cell activation and is associated with increased serum ferritin concentrations, high c-reactive proteins, and proinflammatory cytokines. High tumor load, in multiple myeloma patients receiving CD19-targeting CAR T cells, was associated with a higher incidence of severe cytokine release syndrome. In a small number of patients, macrophage activation syndrome and hemophagocytic lymphohistiocytosis, the most aggressive variants of cytokine release syndrome, are caused by severe immune activation and lead to multiorgan dysfunction.
Neurotoxicity
Immune effector cell–-associated neurotoxicity syndrome (ICANS) symptoms, in multiple myeloma patients treated with BCMA-targeting CAR T cells, may include delirium, transient confusion, aphasia, lethargy, tremor, dysgraphia, seizures, cerebral edema, and rarely, posterior reversible encephalopathy syndrome.1 While the pathophysiology of CAR T cell–related neurotoxicity is not well understood, high tumor load, higher peak concentrations of CAR T cells, and more severe cytokine release syndrome are more common in patients with severe neurotoxicity. “The frequency of neurotoxicities,” Dr. Yong noted in an interview, “has been reduced by steps taken to mitigate these risk factors.”
High interest in phase I study
A phase I study presented in Blood has attracted interest because the novel BCMA-targeting CAR agent (CT103A) being tested is fully human.4 In an accompanying editorial, Lee and Yong note that doubt for any real potential for durable CAR T therapy responses in multiple myeloma is raised by the poor persistence of multiple myeloma CAR T cells in multiple myeloma patients.3
In the earliest trials of BCMA CARs, while reported rates of objective antimyeloma responses were in the approximately 33%-88% range among patients with relapsed/refractory multiple myeloma (RRMM), persistence was typically 6 months or less. Lee and Yong point out, however, that while correlation between persistence and duration of response (DOR) has been variable, median persistence was 308 days in the phase I study. Wang and colleagues, the phase I study authors, state that levels of CAR T-cell proliferation and duration of cellular persistence may be determinants of DOR in CAR T therapy for multiple myeloma. They observe that the multiple mechanisms potentially responsible for the inability of some CAR T cells to survive in vivo, may include antigen escape, T-cell intrinsic mechanisms, tumor microenvironment–mediated suppression, and host anti-CAR immunity. CARs with humanized or fully human single-chain variable fragments (scFvs), prior studies suggest, may retain antitumor activity through bypassing potential host anti-CAR immunogenicity.
In the study, CT103A, a fully human scFv, was tested in an open-label, single-arm design for safety and preliminary efficacy in 18 patients (8 female; median age 53.5 years) with RRMM (at least three lines of prior therapies including a proteasome inhibitor and an immunomodulatory agent) who had undergone leukapheresis and had received lymphodepletion chemotherapy with fludarabine and cyclophosphamide. Four patients (22.2%) had been treated previously with murine anti-BCMA CAR T cells. Safety and tolerability (including dose-limiting toxicity) were the primary endpoints, with efficacy and pharmacokinetics secondary.
Rapid responses
Two weeks after infusion, the overall response rate (ORR) was 77.8% (14 of 18) and by 1 month it was 88.9% (16 of 18). Eventually, all responded and 72.2% (13 of 18) achieved a complete response (CR) or stringent complete response (sCR). All 17 patients evaluated for minimum residual disease (MRD) in bone marrow were MRD-negative at 10-4 nucleated cells by flow cytometry within 1 month. Median DOR was 325 days (range, 7-573 days) for all patients and 412 days (range, 213-573 days) for the 13 with CR/sCR. CAR transgenes were detectable at the cutoff date in 77.8% of patients, with a median CAR transgene persistence of 307.5 days.
During follow-up, four deaths were reported, including one patient with persistent sCR (sudden severe infection). Progression-free survival (PFS) and overall survival (OS) rates at 1 year were 58.3% and 75%, respectively. Extramedullary myeloma was associated with a shortened PFS (79.1% versus 20.0%, P = .015), but not OS (79.1% versus 60.0%, P = .328) at 1 year.
All patients experienced grade 3 or higher adverse events, most of which were expected hematologic effects of lymphodepleting chemotherapy and CT103A infusions. Grade 1 and 2 cytokine release syndromes occurred in 70.6% patients (17 of 18), with 1 grade 4 event (5.9%). The patients receiving a dose of up to 3.0 × 106 CAR+ T cells/kg required less treatment of cytokine release syndrome than the patients who received a dose of 6.0 × 106 CAR+ T cells/kg. No immune effector cell–associated neurotoxicity syndrome was observed. Antidrug antibody positivity occurred in only 1 patient.
Two characteristics of CT103A may contribute to its long persistence, stated study senior author Jianfeng Zhou, MD, PhD, chairman and professor of the department of hematology, Tongji Hospital in Wuhan, China. “One is the reduced immunogenicity achieved by the fully human construct; another is the relatively low binding affinity of the CAR binder. Notably, four patients who previously received murine BCMA CAR were included and still benefit from CT103A. It demonstrates the possibility of retreatment with a different CAR.” Dr. Zhou also emphasized that the lack of ICANS in the entire cohort reflects the excellent safety profile of CT103A.
The editorial commentary in Blood by Lydia Sarah Hui Lee, MD, and Kwee L. Yong, PhD, underscored impressive responses to CT103A, specifically to the median time to response of 15 days, the 100% ORR, and the not reached median progression-free survival at 394 days).5 The best results in other published nonhuman BCMA CAR T-cell trials, they note, were about 1 month (time to response), approximately 33%-88% (ORR), and median progression-free survival of 7-15 months.
Immune responses, Dr. Yong said in an interview, can guide subsequent treatment. “For example, if a patient previously exposed to BCMA CAR T cells in which the construct is either chimeric or humanized, but retains some murine elements, and had detectable antimurine antibodies, we may aim for a fully human one if we are considering treating with a different BCMA CAR T-cell product.” She added, “On the other hand, a similar patient whose serum did not contain such antibodies may be a candidate for a humanized product that retained some murine elements.”
Wang and colleagues concluded, “Altogether, CT103A is safe and highly active in patients with relapsed/refractory multiple myeloma and can be developed as a promising therapy for relapsed/refractory multiple myeloma.”4 An ongoing multicenter phase II trial with single-arm design is recruiting 100 patients. The infusion dosage, suggested by the phase I trial, is 1 × 106 cells/kg. Endpoints include efficacy and safety.
Improving CAR T
Optimizing CAR design and adapting manufacturing processes to generate cell products enriched for T-cell subsets, such as early memory cells, are among strategies being explored to improve CAR T effectiveness.1 Also, dual-antigen targeting to interdict antigen escape and rational combination treatments to enhance persistence are under investigation, along with efforts to improve CAR T-cell therapy safety (for example, incorporation of a suicide gene safety system). They note further that several groups are researching use of induced pluripotent stem cells to generate large quantities of off-the-shelf CAR T-cell immunotherapies that would circumvent the complex, costly, and time-consuming process of manufacturing patient-specific autologous CAR T cells.
References
1. van de Donk N et al. Lancet Haematol. 2021 June;8(6):e446-61.
2. Munshi NC et al. N Engl J Med 2021; 384:705-716.
3. Berdeja JG et al. The Lancet. 2021 July; 398:314-24.
4. Wang D et al. Blood. 2021 May;137(21):2890-901.
5. Lee L and Yong K. Blood. 2021 May;137(21):2859-60.
In multiple myeloma, survival has been very significantly improved by immunomodulatory drugs, proteasome inhibitors, and CD38-targeting antibodies. Despite these advances, multiple myeloma, which is characterized by malignant proliferation of clonal plasma cells in bone marrow, remains an incurable plasma cell disorder with near-certain relapse after successful treatment. Prognosis for patients who develop triple-class refractory disease is poor, with less than 1-year survival. The substantial unmet therapeutic need extends further to other poor survival multiple myeloma populations that include newly diagnosed patients with high cytogenic risk profiles and those with early relapse after first-line therapy. For all of these, interest in drugs with novel mechanisms of action is naturally high.
More specific, less toxic
Post allogeneic hematopoietic stem-cell transplantation and donor lymphocyte infusion sustained remissions reflect a graft-versus-myeloma effect mediated by donor T cells.1 The substantial morbidity and mortality associated with graft-versus-host disease and opportunistic infections, however, have spurred searches for alternative, more specific, and less toxic T-cell therapies with stronger antitumor activity.
Chimeric antigen receptors (CARs)
In CAR T-cell therapies for multiple myeloma, autologous T cells are harvested from the patient and reprogrammed to target multiple myeloma cells through the introduction of genes that encode CARs, which are fusion proteins coupling an antigen-recognition moiety and a transmembrane-spanning element to a T-cell activation domain (typically CD3 zeta [CD247]). The T cells are then expanded and reinfused to the patient following a lymphodepletion regimen. Five strategies using autologous CAR T cells are currently approved for diffuse large B-cell lymphomas, acute lymphoblastic leukemia, multiple myeloma, and other hematologic malignancies. Notably, in patients with heavily pretreated multiple myeloma, CAR T cells have demonstrated impressive activity.
BCMA-targeting CAR T cells
The B-cell maturation antigen (BCMA; TNFRSF17), which plays an important role in the survival of long-lived plasma cells in bone marrow, is an attractive target for CAR T-cell therapy because it is expressed on normal and malignant plasma cell surfaces and by mature B cells. When ligands (TNFSF 13B/TNFSF13) bind to BCMA expressed on multiple myeloma cell surfaces, survival and proliferation pathways and drug resistance are activated.
High-quality responses have been demonstrated in several trials of anti-BCMA CAR T cells, which kill multiple myeloma cell lines and primary multiple myeloma cells through degranulation of T cells and lysis of tumor cells, even those with low BCMA expression. Based on efficacy in triple-class exposed multiple myeloma that compared favorably to conventional care with improved health-related quality of life, the U.S. Food and Drug Administration gave breakthrough designation to ciltacabtagene autoleucel in December 2019 and approval for idecabtagene vicleucel in March 2021.
Idecabtagene vicleucel
Idecabtagene vicleucel expresses a murine BCMA-targeting single-chain variable fragment with a 4-1BB costimulatory motif. The phase 2 KarMMa study2 evaluated idecabtagene vicleucel (target dose of 450 × 106 CAR T cells; range 150 × 106 to 450 × 106) activity in 128 patients with triple-class exposed multiple myeloma. Partial responses or better were observed in 94 of 128 patients (73%) (95% confidence interval, 66-81); 42 (33%) had a complete response or better (95% CI, 25-41), with a median progression-free survival of 8.8 months (95% CI, 5.6-11.6). Outcomes were improved in the highest fixed-dose group, with partial response or better in 81% (44 of 55), complete response or better in 39% (21), and median overall survival of 12.1 months (95% CI, 8.8-12.3). Patients with high-risk cytogenetic profiles, extramedullary disease, and high tumor burden also had deep and durable responses. Outcomes were less favorable in patients with revised International Staging System stage 3 disease.
Ciltacabtagene autoleucel
Ciltacabtagene autoleucel, a 4-1BB–based CAR T-cell therapy with two BCMA-targeting domains, confers high-avidity binding. In the phase 1b/2 CARTITUDE-1 study, conducted in the United States and Europe, preliminary results in 97 patients showed a 97% response rate with ciltacabtagene autoleucel (target dose 0.75 × 106 CAR T cells per kg), and in 65 patients, a complete response (67%). Progression-free survival at 12 months was 77% (95% CI, 66-84) and overall survival was 89% (95% CI, 80-94).3
In the phase 1 LEGEND-2 study4 that was conducted at four sites in China among less heavily pretreated multiple myeloma patients, while all used the same CAR construct, sites used variable conditioning regimens (split versus single). In the site using cyclophosphamide as the lymphodepletion therapy and three split CAR T-cell infusions, partial response or better was achieved in 50 patients (88%) with a median of three prior therapy lines. The complete response rate was high (74%) and minimal residual disease negativity was reached in 39 patients (68%). Median progression-free survival was 19.9 months (95% CI, 9.6-31.0), but 28.2 months among those with complete responses (95% CI, 19.9-not estimable). Median overall survival was also favorable at 36.1 months (95% CI, 26.4-not estimable); it was 35.0 months-not estimable among patients with complete responses. Results from the other three sites were comparable.
Noteworthy among other BCMA-targeting CAR T-cell products in earlier stages of clinical development is orvacabtagene autoleucel, which has a fully human BCMA-specific binding domain. At higher doses (300 × 106 to 600 × 106 CAR T cells) among 62 patients with triple-class–exposed multiple myeloma in the EVOLVE trial, 92% had a partial or better response, with complete responses or better in 36%, all with an encouraging safety profile.
BCMA-targeting CAR T cell toxicity
While van de Donk, Usmani, and Yong, in their review1 note a lack of evidence of off-target toxicity with BCMA-targeting CAR T-cell therapy in clinical studies so far, they do point to several clinical syndromes (cytokine release syndrome, infections, respiratory failure, neurotoxicity, pulmonary aspergillosis, gastrointestinal hemorrhage) caused by cytokines produced during CAR T-cell expansion and to cytopenias and infections arising from prior treatment, bridging therapy, and lymphodepleting conditioning. Deaths attributed to treatment in the above-mentioned trials underscore the need for careful monitoring and early intervention.
Cytokine release syndrome
In the BCMA-targeting CAR T-cell therapy studies, the frequency of cytokine release syndrome varies widely from 17% to 95% but is generally attributed to CAR T-cell activation and is associated with increased serum ferritin concentrations, high c-reactive proteins, and proinflammatory cytokines. High tumor load, in multiple myeloma patients receiving CD19-targeting CAR T cells, was associated with a higher incidence of severe cytokine release syndrome. In a small number of patients, macrophage activation syndrome and hemophagocytic lymphohistiocytosis, the most aggressive variants of cytokine release syndrome, are caused by severe immune activation and lead to multiorgan dysfunction.
Neurotoxicity
Immune effector cell–-associated neurotoxicity syndrome (ICANS) symptoms, in multiple myeloma patients treated with BCMA-targeting CAR T cells, may include delirium, transient confusion, aphasia, lethargy, tremor, dysgraphia, seizures, cerebral edema, and rarely, posterior reversible encephalopathy syndrome.1 While the pathophysiology of CAR T cell–related neurotoxicity is not well understood, high tumor load, higher peak concentrations of CAR T cells, and more severe cytokine release syndrome are more common in patients with severe neurotoxicity. “The frequency of neurotoxicities,” Dr. Yong noted in an interview, “has been reduced by steps taken to mitigate these risk factors.”
High interest in phase I study
A phase I study presented in Blood has attracted interest because the novel BCMA-targeting CAR agent (CT103A) being tested is fully human.4 In an accompanying editorial, Lee and Yong note that doubt for any real potential for durable CAR T therapy responses in multiple myeloma is raised by the poor persistence of multiple myeloma CAR T cells in multiple myeloma patients.3
In the earliest trials of BCMA CARs, while reported rates of objective antimyeloma responses were in the approximately 33%-88% range among patients with relapsed/refractory multiple myeloma (RRMM), persistence was typically 6 months or less. Lee and Yong point out, however, that while correlation between persistence and duration of response (DOR) has been variable, median persistence was 308 days in the phase I study. Wang and colleagues, the phase I study authors, state that levels of CAR T-cell proliferation and duration of cellular persistence may be determinants of DOR in CAR T therapy for multiple myeloma. They observe that the multiple mechanisms potentially responsible for the inability of some CAR T cells to survive in vivo, may include antigen escape, T-cell intrinsic mechanisms, tumor microenvironment–mediated suppression, and host anti-CAR immunity. CARs with humanized or fully human single-chain variable fragments (scFvs), prior studies suggest, may retain antitumor activity through bypassing potential host anti-CAR immunogenicity.
In the study, CT103A, a fully human scFv, was tested in an open-label, single-arm design for safety and preliminary efficacy in 18 patients (8 female; median age 53.5 years) with RRMM (at least three lines of prior therapies including a proteasome inhibitor and an immunomodulatory agent) who had undergone leukapheresis and had received lymphodepletion chemotherapy with fludarabine and cyclophosphamide. Four patients (22.2%) had been treated previously with murine anti-BCMA CAR T cells. Safety and tolerability (including dose-limiting toxicity) were the primary endpoints, with efficacy and pharmacokinetics secondary.
Rapid responses
Two weeks after infusion, the overall response rate (ORR) was 77.8% (14 of 18) and by 1 month it was 88.9% (16 of 18). Eventually, all responded and 72.2% (13 of 18) achieved a complete response (CR) or stringent complete response (sCR). All 17 patients evaluated for minimum residual disease (MRD) in bone marrow were MRD-negative at 10-4 nucleated cells by flow cytometry within 1 month. Median DOR was 325 days (range, 7-573 days) for all patients and 412 days (range, 213-573 days) for the 13 with CR/sCR. CAR transgenes were detectable at the cutoff date in 77.8% of patients, with a median CAR transgene persistence of 307.5 days.
During follow-up, four deaths were reported, including one patient with persistent sCR (sudden severe infection). Progression-free survival (PFS) and overall survival (OS) rates at 1 year were 58.3% and 75%, respectively. Extramedullary myeloma was associated with a shortened PFS (79.1% versus 20.0%, P = .015), but not OS (79.1% versus 60.0%, P = .328) at 1 year.
All patients experienced grade 3 or higher adverse events, most of which were expected hematologic effects of lymphodepleting chemotherapy and CT103A infusions. Grade 1 and 2 cytokine release syndromes occurred in 70.6% patients (17 of 18), with 1 grade 4 event (5.9%). The patients receiving a dose of up to 3.0 × 106 CAR+ T cells/kg required less treatment of cytokine release syndrome than the patients who received a dose of 6.0 × 106 CAR+ T cells/kg. No immune effector cell–associated neurotoxicity syndrome was observed. Antidrug antibody positivity occurred in only 1 patient.
Two characteristics of CT103A may contribute to its long persistence, stated study senior author Jianfeng Zhou, MD, PhD, chairman and professor of the department of hematology, Tongji Hospital in Wuhan, China. “One is the reduced immunogenicity achieved by the fully human construct; another is the relatively low binding affinity of the CAR binder. Notably, four patients who previously received murine BCMA CAR were included and still benefit from CT103A. It demonstrates the possibility of retreatment with a different CAR.” Dr. Zhou also emphasized that the lack of ICANS in the entire cohort reflects the excellent safety profile of CT103A.
The editorial commentary in Blood by Lydia Sarah Hui Lee, MD, and Kwee L. Yong, PhD, underscored impressive responses to CT103A, specifically to the median time to response of 15 days, the 100% ORR, and the not reached median progression-free survival at 394 days).5 The best results in other published nonhuman BCMA CAR T-cell trials, they note, were about 1 month (time to response), approximately 33%-88% (ORR), and median progression-free survival of 7-15 months.
Immune responses, Dr. Yong said in an interview, can guide subsequent treatment. “For example, if a patient previously exposed to BCMA CAR T cells in which the construct is either chimeric or humanized, but retains some murine elements, and had detectable antimurine antibodies, we may aim for a fully human one if we are considering treating with a different BCMA CAR T-cell product.” She added, “On the other hand, a similar patient whose serum did not contain such antibodies may be a candidate for a humanized product that retained some murine elements.”
Wang and colleagues concluded, “Altogether, CT103A is safe and highly active in patients with relapsed/refractory multiple myeloma and can be developed as a promising therapy for relapsed/refractory multiple myeloma.”4 An ongoing multicenter phase II trial with single-arm design is recruiting 100 patients. The infusion dosage, suggested by the phase I trial, is 1 × 106 cells/kg. Endpoints include efficacy and safety.
Improving CAR T
Optimizing CAR design and adapting manufacturing processes to generate cell products enriched for T-cell subsets, such as early memory cells, are among strategies being explored to improve CAR T effectiveness.1 Also, dual-antigen targeting to interdict antigen escape and rational combination treatments to enhance persistence are under investigation, along with efforts to improve CAR T-cell therapy safety (for example, incorporation of a suicide gene safety system). They note further that several groups are researching use of induced pluripotent stem cells to generate large quantities of off-the-shelf CAR T-cell immunotherapies that would circumvent the complex, costly, and time-consuming process of manufacturing patient-specific autologous CAR T cells.
References
1. van de Donk N et al. Lancet Haematol. 2021 June;8(6):e446-61.
2. Munshi NC et al. N Engl J Med 2021; 384:705-716.
3. Berdeja JG et al. The Lancet. 2021 July; 398:314-24.
4. Wang D et al. Blood. 2021 May;137(21):2890-901.
5. Lee L and Yong K. Blood. 2021 May;137(21):2859-60.