CME
Physicians: Maximum of 1.50 AMA PRA Category 1 Credits™
Released: March 22, 2021
Expiration: March 21, 2022
Shaji Kumar, MD:
Many exciting immunotherapy approaches for patients with R/R MM were presented at ASH 2020. Several trials assessed different CAR T-cell therapies in addition to other immunotherapy approaches, such as bispecific antibodies and antibody–drug conjugates. The CAR T-cell trials discussed here have clearly shown the proof of principle for targeting BCMA in R/R MM, and the other BCMA-targeted immunotherapies are quickly catching up.
Shaji Kumar, MD:
CARTITUDE‑1 is an open-label phase Ib/II trial of ciltacabtagene autoleucel, a BCMA‑targeted CAR T-cell therapy, in 113 patients with R/R MM. At ASH 2020, Madduri and colleagues55,56 presented an update of data from CARTITUDE-1.57 In this trial, patients had received ≥ 3 previous therapies—including PIs, IMiDs, and anti-CD38 antibodies—and could be double refractory to PIs or IMiDs. Patients underwent leukapheresis, followed by lymphodepletion with fludarabine (30 mg/m2) and cyclophosphamide (30 mg/m2 for 3 days), followed by infusion of the manufactured CAR T‑cells (0.5 x 106 to 10 x 106 cells/kg; target 0.75 x 106 cells/kg) and postinfusion assessments. Of the 113 patients enrolled, 97 received a ciltacabtagene autoleucel infusion.
The primary endpoints for phase I were safety and determining the RP2D. The primary endpoint in phase II was efficacy (ORR). Secondary endpoints included PFS, OS, and DoR.
Shaji Kumar, MD:
Patients enrolled in CARTITUDE‑1 were heavily pretreated, with a median of 6 previous lines of therapy at the start of the study. Of the enrolled patients, 83.5% were exposed to 5 of the commonly used drugs, and all were triple-class exposed. Nearly one half of the enrolled patients (42.3%) were penta refractory, and 87.6% were triple-class refractory.
Shaji Kumar, MD:
Hematologic toxicity with ciltacabtagene autoleucel was as anticipated for CAR T-cell therapy. All patients had some degree of hematologic toxicity, a combination of the effect of the lymphodepletion and the CAR T-cell therapy itself. More than one half (57.7%) of these patients experienced ≥ 1 infection (grade ≥ 3/4 in 19.6%). The majority of the patients experienced cytopenias, but these were generally transient. The majority of patients who experienced neutropenia recovered their neutrophil count in approximately 5 weeks, with a median recovery of 2 weeks, and most of the patients with thrombocytopenia recovered their platelet counts approximately 10 weeks after the infusion, with a median recovery of 4 weeks.
Shaji Kumar, MD:
In CARTITUDE-1, nonhematologic AEs were similar to those seen in other CAR T-cell trials. General AEs included fatigue, various gastrointestinal AEs, and hypocalcemia and hypophosphatemia. Specific AEs of interest for all CAR T-cell therapies are cytokine-release syndrome (CRS) and neurotoxicity. Nearly all patients (94.8%) had some degree of CRS, with grade 3/4 CRS seen in only 4.1%. Neurotoxicity was reported in 20% of patients, with grade 3/4 neurotoxicity in 9.3%. CRS is seen across all CAR T-cell trials, and we are becoming increasingly familiar with its management.
Shaji Kumar, MD:
Again, in the CARTITUDE-1 trial, nearly all patients had some degree of CRS (94.8%). Approximately 90% of these patients received supportive care measures (predominantly tocilizumab or corticosteroids). The mean duration of the CRS was 4 days (range: 1-97 days), with a median time of onset of 1 week from the CAR T-cell infusion. CRS resolved in 99% of patients within ≤ 14 days of onset.
Shaji Kumar, MD:
Neurotoxicity was seen in approximately 20% of patients, but grade ≥ 3 events were relatively uncommon (approximately 10% of patients). Unusually, compared with other CAR T-cell trials, a few patients in CARTITUDE-1 had neurotoxicity events that were different from the typical immune effector cell–associated neurotoxicity syndrome (ICANS), including movement and neurocognitive changes, as well as nerve palsy and peripheral motor neuropathy.
Shaji Kumar, MD:
There were 14 deaths during the study; 5 were due to progressive disease. The 6 deaths due to TRAEs were from a combination of infections, CRS, and neurotoxicity.
Shaji Kumar, MD:
The ORR with ciltacabtagene autoleucel was quite striking at 96.9%—nearly all patients responded to therapy. The median time to first response was 1 month (range: 0.9-8.5). Approximately two thirds (67%) of the 97 treated patients had sCR, and more than one half (54.6%) achieved MRD negativity. Of the 57 evaluable patients, 93% were MRD negative within a median of 1 month. Nearly three quarters of responses were ongoing at the time of data analysis. This was clearly a very active CAR T-cell product.
Shaji Kumar, MD:
The median PFS has not yet been reached. The 12-month PFS rate was 76.6%, which is approximately the same as many other CAR T-cell studies at the same time point. When assessing PFS by response, the 12-month PFS rate was 84.5% for patients with sCR and 68.0% for patients with VGPR.
Sagar Lonial, MD:
The next CAR T-cell therapy we will mention is bb21217. This is a BCMA-targeted CAR T-cell product with the same CAR molecule as idecabtagene vicleucel (bb2121) but cultured with a PI3K inhibitor ex vivo, with the goal of enriching for T-cells with memory-like phenotype and increasing persistence of the CAR T-cells after infusion.58
In theory, 1 potential reason CAR T-cell therapy has not provided long-term responses or cures in myeloma—as has been seen with R/R acute lymphoblastic leukemia or large B-cell lymphoma—is that myeloma may require better persistence to achieve a long-term benefit.
This concept of skewing CAR T-cells toward a memory phenotype to increase persistence has been demonstrated in numerous animal models, and I am optimistic that it will work for people with myeloma.
Sagar Lonial, MD:
Alsina and colleagues59 presented results from the first-in-human CRB-402 study of the BCMA-targeted CAR T-cell product bb21217.60 CRB-402 is an ongoing, multicenter, open-label, first-in-human, dose-escalation, dose-expansion phase I trial of bb21217 in patients with R/R MM (N = 69).59,60 Eligible patients were previously exposed to ≥ 3 therapies, including IMiD and PI. In the dose-expansion phase, patients also should have received an anti-CD38 antibody and were refractory to their last treatment. In the dose-escalation phase, patients had to have ≥ 50% BCMA expression.
Patients underwent leukapheresis followed by lymphodepletion with fludarabine (30 mg/m2) and cyclophosphamide (30 mg/m2 on Days -5, -4, and -3 before the CAR T-cell infusion). After this, 150-450 x 106 CAR T-cells were infused, followed by postinfusion assessments. The primary endpoints in CRB-402 included AEs and dose-limiting toxicities. Secondary endpoints included PFS, OS, MRD rate, CAR T-cell expansion, and persistence.
Sagar Lonial, MD:
Patients enrolled in this trial had a median age of 62 years and had been exposed to a median of 6 previous lines of therapy (range: 3-17). Most patients (78%) had a previous transplant. Most patients (80%) also were refractory to PIs and IMiDs, and more than one half (64%) were triple refractory to PIs, IMiDs, and an anti-CD38 antibody. The patient population was heavily pretreated across the board. Approximately one third also had high-risk cytogenetics (del17p, t4;14, or t14;16).
Sagar Lonial, MD:
No unexpected AEs were seen, with CRS and neurotoxicity as the most concerning issues, and there were 2 associated deaths. However, we are now better at managing CRS and neurotoxicity, and this has not been an issue with escalated doses in subsequent cohorts.
In addition, pyrexia was seen in more than 40% of patients, and cytopenias were observed in up to 80% of patients, independent of dose, with grade 3/4 infections in 26%.
Sagar Lonial, MD:
For patients who received the highest CAR T-cell dose of 450 x 106, 77% developed CRS and 12% experienced neurotoxicity, but these were mostly grade 1/2. Grade 3 CRS was seen in only 1 patient, and grade 3/4 neurotoxicity was seen in only 3. Unfortunately, 2 patients had grade 5 CRS and died.
However, as was mentioned previously, we are now better at managing CRS and neurotoxicity. The use of tocilizumab and corticosteroids mitigates the progression of a majority of CRS events into more significant issues and allows us to resolve most cases in a reasonable amount of time.
Sagar Lonial, MD:
In the overall patient population, the ORR was 68%, with 73% of patients receiving 450 x 106 CAR T-cells experiencing ORR.
The median time to first response of ≥ PR was 1-2 months, which is typical of CAR T-cell therapies. The median DoR has not yet been reached in the 2 higher-dose groups (450 or 300 x 106 cells) but was 11.5 months for the 10 patients who received 150 x 106 cells.
Sagar Lonial, MD:
Changes in the manufacturing process were made to enhance the safety and ease of manufacturing the CAR T-cell product during this trial. The data suggest that this has improved the ORR (from 57% to 84%).
It is difficult to know if these data are due to the small sample size, but certainly the changes make bb21217 an easier product to give and may make it more reliable overall.
Sagar Lonial, MD:
Patients who had a more durable response had significantly (P < .0001) better CAR T-cell persistence in this trial, which was one of the main goals of the CRB-402 study. Patients with a greater percentage of CAR T-cells with a memory‑like phenotype had a significantly (P ≤ .0024) higher peak expansion of CD4+ and CD8+ central memory T-cells. In addition, patients with memory-like T-cell phenotype (determined by naive vs effector gene signature) were more likely to have a sustained response.
These data fit the paradigm proposed for this study and speaks to the idea that once a CAR T-cell product is working, we can continue to tweak the treatment approaches to improve efficacy for patients. This study and subsequent studies that we will mention are all attempts to update the ideas of existing CAR T-cell products to ultimately improve the durability of responses for patients with refractory MM.
Shaji Kumar, MD:
The majority of the CAR T-cell products that have been investigated in clinical trials have been created using a lentiviral vector. By contrast, the PRIME study used the CAR T-cells created using a transposon as the vector for the genetic engineering for the CAR.
PRIME is an ongoing multicenter, open-label, nonrandomized, multicohort phase I/II trial in patients with R/R MM evaluating different regimens before infusion of the CAR T-cell product P-BCMA-101.61,62 To date, the study has enrolled 55 of a planned 220 patients. Participants in PRIME must have ≥ 3 previous lines of therapy (including PI and IMiD) or ≥ 2 previous lines and be refractory to PI and IMiD. Patients were grouped into 6 cohorts, and all underwent initial apheresis (followed by either rituximab or lenalidomide in 3 cohorts) before 3-day lymphodepletion (fludarabine 30 mg/m2/day plus cyclophosphamide 300 mg/m2/day) and infusion of P-BCMA-101. P-BCMA-101 was initially given at escalating single-infusion doses of 0.75-15 x 106 cells/kg and subsequently at a median dose of 0.75 x 106 cells/kg in biweekly cycles. This study also looked at different systemic therapy after CAR T-cell therapy: 1 cohort received lenalidomide after infusion, 1 received rituximab, and 2 received additional infusions of P-BCMA-101. A single cohort received only 1 infusion of P-BCMA-101.
The primary endpoints for phase I of the study were safety and determining the maximum tolerated dose. For phase II, primary outcomes were safety, ORR, and DoR. Key secondary endpoints were ORR, PFS, OS, and DoR.
Shaji Kumar, MD:
The majority of patients in this trial were heavily pretreated, as in the other CAR T-cell therapy trials. Patients had previously been exposed to a median of 8 previous lines of therapy (range: 2-18). In total, 60% of patients were refractory to PI, IMiD, and anti-CD38 antibodies, and 7% were refractory to anti-BCMA therapies.
Shaji Kumar, MD:
The majority of patients experienced ≥ 1 AE. CRS was noted in only 17% of patients, and none was grade ≥ 3. CRS was not seen in patients who received rituximab or lenalidomide. Two patients experienced grade 3 neurotoxicity. The rate of hematologic toxicities was comparable with other trials of CAR T-cells, including 77% of patients experiencing neutropenia and 42% experiencing thrombocytopenia. A total of 45% of patients experienced infections, but only approximately 20% were grade ≥ 3.
Shaji Kumar, MD:
The ORR shows that, especially at higher dose levels, the majority of evaluable patients responded (ranging from 44% to 75%), including some who achieved ≥ VGPR. For patients receiving P-BCMA-101 with nanoplasmid, the ORR was 66.7% vs 50.0% for those receiving P-BCMA-101 with the standard plasmid.
Shaji Kumar, MD:
In the phase I/II PRIME trial, P-BCMA-101 was safe and effective at doses up to approximately 1200 x 106 CAR T-cells in patients with R/R MM. Although CRS was seen in 17%, it was low grade and easily managed with tocilizumab and steroids.
The use of transposons and nanoplasmids as an alternate approach to creating CAR T-cells holds promise and resulted in a 66.7% response rate. Because the lentiviral vector approach used in many other CAR T-cell therapies requires a long manufacturing time and is costly, different and more efficient approaches for creating CAR T-cells may allow for shorter manufacturing times (thus a shorter waiting period for patients) and some cost savings. Less wait time before starting new treatment and lower costs would definitely be an advantage for patients requiring therapy with CAR T-cell products.
Sagar Lonial, MD:
Another innovative approach for CAR T-cell therapy in MM is the use of allogeneic “off-the-shelf” CAR T-cell products. Normally, CAR T-cells are autologous—meaning that T-cells are taken from each patient individually and altered ex vivo—with weeks required to manufacture the patient-specific drug, but using universal donor cells rather than collecting and modifying autologous T-cells should allow for more rapid treatment. In addition, this approach may remove the question of donor T-cell health for the durability and efficacy of CAR T-cell therapies.
Sagar Lonial, MD:
UNIVERSAL is an ongoing, multicenter, open-label, dose-escalation, first-in-human phase I study using an allogenic CAR T-cell targeting BCMA (ALLO-715).63,64The 35 patients with R/R MM enrolled in this study were all heavily pretreated (≥ 3 previous therapies). Patients received lymphodepletion using cyclophosphamide plus an anti-CD52 antibody (ALLO-647), with or without fludarabine, and then were infused with a single dose of anti-BCMA ALLO-715 (40, 160, 320, or 480 x 106 CAR T-cells). Primary endpoints were safety and tolerability, and key secondary endpoints were ORR, PFS, DoR, and MRD, plus cellular kinetics and pharmacokinetics.
Sagar Lonial, MD:
Of the 35 enrolled patients, 31 were evaluated for safety. These patients had a median of 5 previous lines of therapy (range: 3-11), and 94% were penta exposed. In total, 48% had high-risk cytogenetics, and 23% had extramedullary disease.
Sagar Lonial, MD:
Overall, a manageable safety profile was associated with allogeneic BCMA-targeted CAR T-cells, with no graft-vs-host disease or immune effector cell–associated neurotoxicity syndrome. However, 45% of patients in this study had grade 1/2 CRS. Infections occurred in 42% of patients, but this was not a surprise given how heavily pretreated these patients were and the standard lymphodepletion used. Five infections (16%) were grade ≥ 3.
Sagar Lonial, MD:
The ORR was 50% to 75%, depending on the cohort. Of note, 60% of patients in the cohort receiving lymphodepletion with fludarabine, cyclophosphamide, and ALLO-647 followed by ALLO-715 at a dose of 320 x 106 cells responded to treatment, and 40% achieved ≥ VGPR. This suggests that ALLO-715 could be a viable treatment approach and that many of these patients had durable responses. Of the 40% of patients with VGPR, 5 of 6 had negative MRD status.
Sagar Lonial, MD:
The median time to response was relatively quick (16 days). Of more importance, several patients are continuing to respond months after their initial infusion, including 6 of 9 patients who received 320 x 106 or 480 x 106 cell dosing. The response rates appear to be dose dependent.
Sagar Lonial, MD:
One of the concerns regarding allogeneic CAR T-cells was based on the assumption that, because persistence of allogeneic stem cells after transplantation has been an issue in myeloma, this may also be true for allogeneic CAR T-cells. In general, cells from an allogeneic donor often have shorter persistence than from an autologous donor (ie, the patient themself).
In UNIVERSAL, it was intriguing that numerous patients—particularly at the highest dose (320 x 106 cells)—had persistence of the allogeneic CAR T-cells 4 months after infusion. This suggests that the rejection concern for allogenic CAR T-cells may be more limited than previously thought, potentially due to the other immunosuppression that these heavily pretreated patients have undergone.
This study makes allogenic CAR T-cells a more viable concept in myeloma: It works, it’s safe, and significant persistence can be achieved. Moreover, UNIVERSAL’s novel approach for delivering CAR T-cell therapy may be more efficient than some current approaches.
Shaji Kumar, MD:
In the first-in-human phase I UNIVERSAL study, allogeneic CAR T-cell therapy with the anti-BCMA product ALLO-715 preceded by an anti-CD52 antibody was well tolerated and showed dose-dependent activity in heavily pretreated patients with R/R MM.
The time to CAR T-cell infusion was more rapid than with autologous CAR T-cells, with 90% of patients receiving their infusion within 5 days of enrollment. This approach appears safe, with no graft-vs-host disease or immune effector cell–associated neurotoxicity syndrome and only low-grade CRS. Of note, efficacy seems dose dependent, with greater doses achieving higher responses.
Shaji Kumar, MD:
Overall, the results from the CARTITUDE-1 study of ciltacabtagene autoleucel, the CRB-402 study of bb21217, the PRIME trial of P-BCMA-101, and the UNIVERSAL study of ALLO-715 suggest that anti-BCMA CAR T-cells are effective and provoke an immune response in patients with MM. This has been a real concern, as patients with MM are exposed to so many drugs throughout the course of their disease. This year, the CAR T-cells we mentioned also highlight some of the nuances that healthcare professionals and researchers are considering to improve on the current paradigm for CAR T-cell therapy. Skewing CAR T-cells to a memory phenotype, as was done with bb21217, may improve their persistence. Use of transposons and nanoplasmids as vectors can decrease both manufacturing time and cost. Allogeneic CAR T-cells offer the ability to use an off-the-shelf product. The speed to treatment is especially important in the R/R setting, where patients cannot wait very long to get their own T-cells manufactured and brought back.
Sagar Lonial, MD:
I am impressed with the evolution from the first CAR T-cells into modern CAR T-cell therapies with improved efficacy, durability, and safety.
Many of us were initially disappointed with the long-term efficacy of CAR T-cells in myeloma, but these studies give us hope that there will be subtle nuances that help improve the overall efficacy of this approach.
Shaji Kumar, MD:
I agree. Another key for improving efficacy of the CAR T-cell therapies in MM is going to be using them in earlier settings. I hope there will be more durable responses as these trials progress.
Shaji Kumar, MD:
Bispecific antibodies represent another new approach for targeting BCMA that has shown promise in MM. The premise is similar to that of CAR T-cell therapy, in that these molecules harness the power of the body’s own T-cells to develop an anti-MM response that can lead to disease control. Unlike with CAR T-cell therapy, which requires apheresis and manufacturing, bispecific antibodies are designed to bring normal T-cells close to myeloma cells in the body, resulting in an immune synapse formation and cell killing. Bispecific antibodies are recombinant proteins made up of 2 single-chain variable fragments or antibodies linked together: 1 targeting a T-cell surface molecule, such as CD3, and 1 targeting an antigen common to a specific type of cancer cell, such as BCMA for MM. Of importance, bispecific antibodies are available “off the shelf” for patients but do generally require repeated dosing.
Several bispecific antibodies for MM are currently in clinical trials. One of the first bispecific antibodies to be tested in the clinical trial setting was the BCMA-/CD3-targeted bispecific antibody pavurutamab (AMG 701) and its precursor molecule, AMG 420. Although the first-generation bispecific antibodies had to be infused frequently, subsequent generations have modifications and can be given weekly or even less often.
Shaji Kumar, MD:
At ASH 2020, Harrison and colleagues65 presented results from an ongoing, first-in-human, dose-escalation phase I study of pavurutamab in patients with R/R MM after ≥ 3 previous therapies including PI, IMiD, and an anti-CD38 antibody.66 None of the patients in this study had a history of anti-BCMA treatment. In total, 85 patients were enrolled in this study, and they received weekly IV infusions with AMG 701 at various doses after being premedicated with 8 mg (or equivalent) of dexamethasone given in the first 2 cycles. Patients were treated on 4-week cycles until progressive disease, toxicity, death, or withdrawal. The primary endpoint of this study was safety/tolerability and estimated active dose, and secondary endpoints include pharmacokinetics and ORR.
Shaji Kumar, MD:
Patients in this study had a median of 6 previous lines of therapy (range: 2-25). Again, similar to the CAR T-cell trials, 93% were triple exposed (PI, IMiD, and anti-CD38 antibody), and 62% were triple refractory. Of note, 82% had received a previous stem cell transplant (80% allogeneic).
Shaji Kumar, MD:
The dose-limiting toxicities associated with pavurutamab primarily involved CRS, as expected, which was well managed with tocilizumab and corticosteroids.
Of the 65 patients with CRS, 9 had a grade 3 event. The median duration of CRS was short (2 days), and CRS generally resolved with tocilizumab and/or steroid treatment. The other AEs were as anticipated with these treatments: hematologic toxicity (neutropenia in 25% and thrombocytopenia in 21%) and gastrointestinal toxicity. Non-CRS dose-limiting toxicities included grade 3 atrial fibrillation and acidosis and grade 4 thrombocytopenia (n = 1 each).
Shaji Kumar, MD:
Across the study, the ORR was 26%, but this varied substantially. The majority of patients (n = 55) were in the 3-18 mg cohort (n = 55), which had a 36% ORR. Of note, the ORR was just 4% in the 27 patients who received 0.015-1.6 mg of pavurutamab. By contrast, the ORR was 83% in the most recently evaluated cohort (n = 6).
When assessing patients with ≥ VGPR, 6 of 7 patients (86%) tested for MRD were negative by next-generation sequencing (≤ 10-5 per International Myeloma Working Group) or flow cytometry (≤ 3 x 10-5).
Shaji Kumar, MD:
One interesting aspect of this study is that many of these responses were durable, with some patients still receiving treatment after 2 years on therapy (maximum: 26 months). The median DoR was not reached in this study at a median follow-up of 6.5 months.
Obviously, we need more longer-term data. Data are especially needed at higher levels of the RP2D for this bispecific antibody.
Shaji Kumar, MD:
Pharmacokinetic studies indicate a clear dose-dependent response, with higher responses at the higher doses. These results support once-weekly dosing. Of note, baseline BCMA levels may affect pavurutamab free-drug exposure.
Shaji Kumar, MD:
These results from an ongoing phase I study suggest that a BCMA-targeted bispecific antibody approach with pavurutamab can be quite effective in a heavily pretreated group of patients with R/R MM. In this trial, pavurutamab had a manageable safety profile, including transient, reversible CRS (median duration: 2 days). With the caveat of a small patient group, the 83% ORR is promising, as is the high rate of MRD negativity (86%), and the pharmacokinetics profile supports once-weekly dosing.
Sagar Lonial, MD:
Another BCMA-/CD3-targeted bispecific antibody in early clinical trials for R/R MM is teclistamab, which has subtle structural differences from pavurutamab. Pavurutamab uses fusion proteins from the variable regions of the antibody-heavy and light-chain regions to create a bispecific T-cell engager molecule, but teclistamab comprises the heavy- and light-chain homodimers from 2 separate monoclonal antibodies to form a single heterodimeric antibody structure.
Sagar Lonial, MD:
At ASH 2020, Garfall and colleagues67 presented updated results from the dose-escalation, dose-expansion phase I study of teclistamab in149 patients who had R/R MM or who were intolerant to established treatments.68 Teclistamab was started at 1-3 step-up doses, then given as full doses weekly after cycle 1. Patients were split into 2 step-up dosing cohorts: IV (n = 84) with doses from 0.3-720 µg/kg and SC (n = 65) with doses from 80-3000 µg/kg.
The RP2D based on this study is 1500 µg/kg SC. The key endpoints for part 1 of this study were safety and tolerability, and determining the RP2D, and key endpoints for part 2 were antitumor activity, safety, and pharmacokinetic and pharmacodynamic data.
Sagar Lonial, MD:
Patients in this study were exposed to a median of 6 previous lines of therapy (median of 5 lines in the 33 patients treated at the RP2D of 1500 µg/kg SC). Most patients (96%) were triple exposed, and 69% of patients in the total group were penta exposed (2 PIs, 2 IMiDs, 1 anti-CD38 antibody). In total, 39% were penta refractory, and in the RP2D group, 36% were penta refractory. Approximately 90% were refractory to their last line of therapy.
Sagar Lonial, MD:
Nonhematologic grade ≥ 3 AEs were minimal in this early report. There was no grade ≥ 3 CRS, and 55% of patients experienced lower-grade CRS. Hematologic toxicities were also noted, but that was no surprise. Grade ≥ 3 hematologic AEs included neutropenia (46%), anemia (32%), thrombocytopenia (40%), and leukopenia (14%).
Infections were observed in 52% of all patients and 27% of patients who received the RP2D, and 15% of all infections were grade ≥ 3.
No true maximum tolerated dose was identified, as there were no dose-limiting toxicities at the RP2D. But from a dose-saturation standpoint, patients probably received the maximum amount of antibody that was necessary to target BCMA.
Sagar Lonial, MD:
The rate of CRS with SC dosing was almost identical to what we saw with IV dosing. Again, most of this was grade 1 (15% and 16% for IV and SC, respectively) and grade 2 (38% and 42% for IV and SC), but no patient experienced worse than grade 2 CRS across the board.
There was some neurotoxicity seen in 7 patients at the RP2D, but it resolved with appropriate supportive care, which is similar to how it is managed in patients receiving CAR T-cells.
Sagar Lonial, MD:
Efficacy data for teclistamab at the RP2D of 1500 µg/kg SC showed a rate of ≥ VGPR of 55%, with an ORR of 73%.
The ORR was 67% and 71% at the most active doses of 270-720 µg/kg IV and 720-3000 µg/kg SC, respectively. Most patients who were triple-class refractory (70%) or penta refractory (75%) responded. There were 8 MRD-negative CRs at 10-6 in the 11 patients who had MRD assessment performed.
Sagar Lonial, MD:
In summary, teclistamab is clearly another active BCMA-targeted bispecific antibody. It can be given SC, which is a huge convenience factor for patients across the board.
Soluble BCMA does not appear to impact teclistamab, and it is very active. Patients had very little severe CRS and very little severe neurotoxicity, most of which can be managed similarly to what we do for CAR T-cells.
Shaji Kumar, MD:
Data for several other bispecific molecules were presented during the meeting, including REGN5458. This is a humanized BCMA x CD3 bispecific antibody construct that is designed to mimic natural human antibodies.
In this open-label, dose-escalation phase I trial assessing REGN5458, 49 patients with R/R MM who had received ≥ 3 lines of previous therapy (including PI, IMiD, and anti-CD38 monoclonal antibodies) received a weekly infusion and then biweekly infusions of REGN5458.69,70 The primary objective was to evaluate safety and tolerability and to understand the RP2D. The secondary objectives were ORR, DoR, PFS, MRD status, and OS.
Shaji Kumar, MD:
Again, 49 heavily pretreated patients were enrolled in this study. Patients were exposed to a median of 5 previous lines of therapy (range: 2-17). All patients were triple refractory, and 57% were penta refractory. All patients were refractory to anti-CD38 antibodies, 92% were refractory to pomalidomide, and 80% were refractory to carfilzomib.
Shaji Kumar, MD:
The AEs associated with REGN5458 were similar to other anti-BCMA bispecific antibodies, and CRS was seen in approximately 40% of these patients.
Hematologic toxicity was reported, as well, with 37% experiencing anemia (22% grade ≥ 3) and < 20% of patients experiencing lymphopenia, thrombocytopenia, or neutropenia.
Some patients experienced upper respiratory tract infections (12%) and pneumonias (12%). There were not opportunistic infections, but there were 2 dose-limiting toxicities: acute kidney injury and hepatotoxicity. No significant neurotoxicities were seen in this study.
Shaji Kumar, MD:
At the highest dose level studied, the ORR was 62.5%. Of the patients responding, 42% achieved CR or sCR, and 95% achieved ≥ VGPR.
Based on these data, we can see that even patients with highly refractory disease are able to achieve a response with REGN5458. Among patients who achieved CR or sCR and had MRD testing, 57% (4/7) were MRD negative (at 10-5).
Shaji Kumar, MD:
Responses with REGN5458 were durable. The median DoR was 6 months, but several patients remain in response, with a treatment duration of almost a year or beyond. At this time, 74% of patients remain on REGN5458 therapy. As we go to the higher dose levels, we hope to continue to see these durable responses.
Shaji Kumar, MD:
This trial also looked at quality-of-life metrics. It is difficult to interpret this data in the context of a single-arm dose-escalation study, but there did not seem to be a significant negative impact on quality of life with the use of this bispecific antibody. None of the acute toxicities that occurred soon after the infusion seemed to have an impact on quality of life.
Sagar Lonial, MD:
In considering these trials of anti-BCMA bispecific antibodies and T-cell engagers, we are currently where we were with anti-BCMA CAR T-cell trials approximately a year ago. Each of these agents has significant activity—more than I expected from this class of therapy, given that the T-cell health of patients with refractory myeloma has historically been poor. Now that we see these feasibility studies, we need data with larger patient numbers and longer follow-up to start to understand whether the subtle structural differences in each bispecific antibody translate into clinical differences, as well.
Shaji Kumar, MD:
The bispecific antibodies open the myeloma field for immunotherapies. One of the disadvantages with the CAR T-cell, in addition to the waiting time with manufacturing, is the fact that only a limited number of institutions have the needed infrastructure for administration of CAR T-cell therapy and management of patients after they receive the therapy.
By contrast, the use of bispecific antibodies will allow many more institutions and hospitals to provide this type of therapy for patients, and it can be given in a timely fashion. Of most importance, these novel treatments can be combined with other anti-MM therapies, which makes them attractive in earlier lines of therapy, including the frontline setting, where we have many other effective drugs. We hope that using these anti-BCMA therapies in combination with various backbones will allow us to get these patients into deep responses.
Shaji Kumar, MD:
BCMA is an excellent target for developing therapeutics for myeloma. This has been taken advantage of, as we have seen with both the CAR T-cell trials and bispecific antibodies. When considering antibody–drug conjugates, we have already seen the activity of the anti-BCMA belantamab mafodotin, which is now FDA approved for patients with R/R MM who have received ≥ 4 previous therapies, including an anti-CD38 antibody, an IMiD, and a PI.
Shaji Kumar, MD:
MEDI2228 is a monoclonal antibody–targeting BCMA that is linked to a toxin called PBD (pyrrolobenzodiazepine). At ASH 2020, data were presented from the first-in-human, open-label, dose escalation phase I study with MEDI2228.71,72 Overall, 82 patients with R/R MM who had progressed after treatment with 3 drug classes (PI, IMiD, and monoclonal antibodies) were enrolled in this study. The dose-escalation phase included 41 patients, and the dose-expansion phase included the remaining 41 patients. The primary endpoints were safety and tolerability, and the secondary endpoints were efficacy, pharmacokinetics, and immunogenicity.
Shaji Kumar, MD:
In this study, we have a relatively heavily pretreated group of patients. All patients were previously treated with PIs, 98.8% were previously treated with an IMiD, and 97.6% were previously treated with antibodies (anti-CD38 or anti-SLAMF7). In the total patient population, 57.3% were triple refractory, and in the dose-expansion phase, 56.1% were triple refractory.
Shaji Kumar, MD:
The type of AEs reported in this trial were consistent with what was expected given the PBD cytotoxic payload. Patients experienced photophobia, thrombocytopenia, pleural effusion, abnormal liver function test results, and rash.
In general, TRAEs increased with dose, and most events were grade 1 or 2. As the dose increased, there were a higher number of dose delays and dose omissions due to TRAEs.
Shaji Kumar, MD:
The eye toxicity encountered in this study was unusual. In total, 24 patients developed ocular toxicity, with a median time to onset of approximately 2-3 cycles, which is approximately 2 months. Patients presented predominantly with photophobia, without any clear findings on the physical examination and no keratopathy reported. The photophobia increased over time for 9 of 24 patients and completely resolved for 4 patients.
Many of these patients had early discontinuation, so we don’t have a full picture of the rate at which these AEs have resolved in the study.
Shaji Kumar, MD:
When you look at the 0.14 mg/kg cohort, which is the RP2D, almost 66% of patients had a response to therapy. Many of these responses have been durable, and some patients had deep responses, including CR and VGPR. These responses have been lasting, with some going beyond a year at this point.
Shaji Kumar, MD:
The pharmacokinetic studies show that there is an increased serum concentration with increased dose. No significant free toxin (tesirine) could be detected in the serum, suggesting no significant dissociation before the antibody–drug conjugate was internalized by the tumor cell.
Shaji Kumar, MD:
Overall, the MEDI2228 data were promising, but MEDI2228 has some ocular toxicity reminiscent of what has been seen with belantamab mafodotin. However, the mechanisms for ocular toxicity are completely different in this study, given the lack of keratopathy that has been seen in studies with belantamab mafodotin. Now, alternate dosing strategies are being explored for MEDI228 to see if there is a dosing schedule that will retain the efficacy with minimal toxicity.
Sagar Lonial, MD:
More therapies coming to the party in myeloma is a good thing. So, the idea of an additional antibody–drug conjugate targeting BCMA with a slightly different mechanism of action will help us overcome drug resistance. All of these approaches with CAR T-cells, bispecific antibodies, and antibody–drug conjugates are welcomed to see whether we can make a difference overall.
Back
