CARTITUDE-4 was a randomized, open-label phase III trial assessing treatment with cilta-cel vs the SoC regimens pomalidomide/bortezomib/dexamethasone or daratumumab/pomalidomide/dexamethasone.^40,41 Patients with R/R MM whose disease was lenalidomide refractory after 1-3 previous lines of therapy were enrolled, and patients’ prior therapies must include a PI and an IMiD. Patients who received prior BCMA‑targeting therapy were excluded. In total, 419 patients were randomly assigned to receive a cilta‑cel infusion or SoC administered until disease progression. 

The primary endpoint was PFS. Secondary endpoints included CR or better, overall ORR, MRD negativity, OS, safety, and patient-reported outcomes. 

Patients were stratified by the type of SoC therapy, ISS stage, and the number of previous lines of therapy. The current analysis was completed after a median follow-up of 15.9 months.

The patient groups were fairly balanced between the 2 treatment arms.^40,41 Patients had a reasonably good PS, with 99.5% of patients having a PS of 1 or better in each arm. Approximately 94% of patients had ISS stage I or II disease in each arm. Of interest, approximately one fifth of patients had soft tissue plasmacytomas.

Overall, a high proportion of the trial population had higher-risk disease, suggesting that patients may have progressed after their initial lines of therapy relatively quickly. The median time from diagnosis was approximately 3 years in each arm, and approximately one third of patients had received only 1 prior line of therapy. Approximately 26% of patients were triple-class exposed, 15% were triple-class refractory, and 22% previously had received daratumumab.

Cilta-cel demonstrated a clear improvement in PFS in the intention-to-treat (ITT) population compared with SoC (median PFS: not reached vs 11.8 months; HR: 0.26; 95% CI: 0.18-0.38; P <.0001).^40,41 The median PFS of approximately 12 months reported in the SoC control arm is what one would have anticipated with these 2 triplet regimens in patients who were lenalidomide refractory.  

The 1-year PFS rate was 76% with cilta-cel and 49% with SoC. Of interest, a higher proportion of patients had early PFS events in the cilta‑cel arm during and directly after the bridging phase prior to cilta-cel infusion, which then caught up over time. These data likely reflect the time taken to initiate apheresis and cilta-cel infusion.

Subgroup analyses demonstrated a clear PFS benefit with CAR T-cell therapy regardless of the number of lines of therapy, but patients with 2‑3 prior lines of therapy may have had a slightly greater relative benefit compared with patients who had received only 1 prior line of therapy.^40,41  

The PFS benefit also was seen across ISS stages, but fewer patients had ISS stage III disease. The presence of soft tissue plasmacytomas seems to have somewhat affected outcomes, but these patients still benefited from the use of CAR T‑cell therapy.

Tumor burden and prior exposure to anti-CD38 therapy did not seem to have an impact on the benefit of cilta-cel. The PFS benefit also appeared to be similar regardless of cytogenetic risk, but there may have been a greater benefit in the high‑risk patient population. 

In the ITT analysis, ORR with cilta-cel was 84.6% (58.2% stringent CR [sCR]) vs 67.3% (15.2% sCR) with SoC.^40,41 The median DoR was not reached with cilta-cel and 16.6 months with SoC, whereas the 12-month DoR rates were 84.7% and 63.0%, respectively. Rates of MRD negativity were 60.6% with cilta-cel and 15.6% with SoC (odds ratio: 8.7; P <.0001). 

OS data were immature at the time of analysis; there were 39 deaths in the cilta-cel arm and 47 in the control arm (HR: 0.78; 95% CI: 0.5-1.2; P = .26).

In the as-treated population of patients who received cilta-cel as study treatment, cilta-cel was associated with an ORR of 99.4%, including 68.8%of patients with sCR, and an MRD negativity rate (at a sensitivity of 10^-5) of 72% among 176 assessed patients.^40,41 

At 12 months from the time of apheresis, 90% of patients were alive and progression free. The 12-month PFS rate from the time of cilta-cel infusion was 85%. In comparison, the 12-month PFS rate reported in the CARTITUDE-1 trial in a more heavily pretreated population was 76%, and the rate of MRD negativity in this population was 58%.⁴³ 

Of note, 32 patients included in the ITT population did not receive cilta-cel as study treatment due to disease progression or death during bridging therapy and/or lymphodepletion and were not included in the as-treated analysis, but 20 of these patients received cilta-cel as subsequent therapy.

Rates of any TEAEs and serious TEAEs were similar between arms. Serious TEAEs of any grade occurred in 44.2% of patients in the cilta‑cel arm and 38.9% of patients in the SoC arm.^40,41  

Rates of hematologic toxicities with cilta-cel were as one would anticipate based on prior results. Most patients experienced neutropenia, and rates of grade 3/4 neutropenia were 89.9% with cilta-cel vs 82.2% with SoC. Rates of grade 3/4 thrombocytopenia were 41.3% and 18.8%, respectively, and rates of grade 3/4 lymphopenia were 20.7% and 12.0%, respectively.  

Infections were common in both treatment arms, with grade 3 or 4 infections occurring in 26.9% of patients in the cilta‑cel arm and 24.5% of patients in the SoC arm. In addition, 9 (4.3%) second primary malignancies were reported in the cilta-cel arm and 14 (6.7%) in the SoC arm. This is important to note when we talk about the long‑term follow-up of the CARTITUDE‑1 trial. 

Overall, there were 10 deaths due to TEAEs in the cilta‑cel arm, predominantly from COVID‑19 or other infections. There were 5 deaths due to TEAEs in the SoC arm, again predominantly from infections, but 1 patient died from progressive multifocal leukoencephalopathy and 1 patient from pulmonary embolism.

CRS and neurotoxicity are of particular interest with CAR T‑cell therapy and other immunotherapies. The any-grade CRS rate was 76.1% in CARTITUDE-4, and only 2 patients (1%) developed grade 3/4 CRS, which is substantially lower than what was noted in the initial studies of cilta‑cel.^40,41,43 The median time to onset of CRS was 8 days, and the median duration was 3 days. 

ICANS was observed in 4.5% of patients, with no grade 3 or 4 events reported. This incidence was lower than that reported in the initial phase I/II study of cilta-cel.^40,41,43 The median time to onset of ICANS was 10 days, and the median duration was 2 days. 

Of interest, 9.1% of patients developed cranial nerve palsy in this trial, and this is important to keep in mind for patients with R/R MM who are receiving CAR T‑cell therapy. Peripheral neuropathy and other neurologic toxicities were uncommon, and there were no fatal neurotoxicities. 

Also at ASCO 2023, the end-of-study analysis of the phase Ib/II CARTITUDE-1 trial, which led to the FDA approval of cilta-cel in patients with R/R MM after ≥4 prior therapies including a PI, an IMiD, and an anti-CD38 Ab, was presented.^34,42 This single‑arm, open‑label trial enrolled patients with R/R MM with measurable disease after ≥3 prior therapies including a PI, an IMiD, and an anti-CD38 Ab or who were double refractory to a PI and an IMiD. 

Patients were enrolled, underwent leukapheresis, received bridging chemotherapy as required clinically, and then underwent lymphodepletion with fludarabine/cyclophosphamide followed by infusion of the CAR T-cells. The primary endpoint for CARTITUDE‑1 was safety for phase I and ORR for phase II. Secondary endpoints included PFS, OS, and MRD negativity (at a sensitivity of 10^-5). At 27.7 months of follow-up, cilta-cel in patients who were heavily pretreated demonstrated an ORR of 97.9% with median PFS not yet reached.³⁸ The current analysis reports end-of-study findings at a median follow-up of 33.4 months.

This update confirmed that cilta-cel was associated with an ORR of 97.9% (82.5% sCR).⁴² The median DOR and PFS were quite impressive at nearly 3 years for each (33.9 months and 34.9 months, respectively). The median OS was not reached, but the 3-year OS rate was 62.9%. 

Sustained MRD negativity for ≥12 months was attained in more than 50% of evaluable patients (26 of 49); 18 of those 26 patients remained MRD negative with a CR or better at 24 months post infusion.

The median PFS with cilta-cel was 34.9 months in the overall population, with a 3-year PFS rate of 47.5%. Among patients who achieved a CR or better at any time during the study, the median PFS increased to 38.2 months, with a 3-year PFS rate of 59.8%. Among patients with MRD negativity sustained for ≥12 months, the median PFS was not reached, and the 30-month PFS rate was 74.9%.⁴² Clearly, patients who had a deep or sustained response had a significant PFS benefit.

Regarding safety in CARTITUDE‑1, a few additional pieces of information have become evident with longer‑term follow-up.⁴² At a median follow-up of 27.2 months, 35 deaths have occurred among the 97 enrolled patients. Approximately one half of these deaths (18 of 35) were due to causes other than MM disease progression, and 17 were due to progressive disease. 

The most common cause of death other than progressive disease was infections, including pneumonia (n = 2 unrelated to treatment), respiratory failure (n = 3 unrelated to treatment and n = 1 related to treatment), and sepsis (n = 2 unrelated to treatment and n = 2 related to treatment). There were 3 deaths due to AML and 1 death due to MDS. One patient with AML had MDS and cytogenetic abnormalities prior to their cilta‑cel infusion. The cytogenetic profile details are not available for the others. There were 5 additional deaths since the last analysis, but these all were considered unrelated to cilta-cel. 

It is important to note the spectrum of AEs that have occurred among patients receiving cilta-cel, particularly the impact of infections in this patient population and the potential signal of second primary malignancies. It is unknown if the emergence of these second primary malignancies is related to the fact that these patients have lived longer than what they otherwise would have, representing a natural history of the disease, or whether these malignancies are related to the perturbation of their immune system or immune surveillance, allowing cancers that might have been present even before to progress more rapidly. Certainly, we have a lot to learn about those aspects.

In addition to CAR T-cell therapy, bispecific antibodies were another type of immunotherapy in MM for which we saw important updates at ASCO 2023. These agents are engineered artificial antibodies that bind to the CD3 subunit within the T-cell receptor with one arm and to a tumor-associated antigen with the other.⁴⁴ 

Teclistamab is an off-the-shelf anti-BCMA x CD3 bispecific antibody that is approved for patients with R/R MM who have received ≥4 prior lines of therapy including a PI, an IMiD, and anti-CD38 Ab.³⁵ This approval was based on results from the phase I/II MajesTEC-1 trial, which evaluated the safety and efficacy of teclistamab in patients with triple class‒exposed R/R MM with no prior BCMA therapy.⁴⁵ The trial population had similar characteristics as the population from the CARTITUDE‑1 study. 

At ASCO 2023, van de Donk and colleagues⁴⁶ presented an update from MajesTEC-1 after a median follow-up of 23 months. The phase I portion included IV and SC dose escalation to find the randomized phase II dosing. Then, in the phase II dose-expansion cohort, patients received teclistamab at 1.5 mg/kg SC weekly, which is the approved dose for clinical use. Responding patients could switch to every-2-week or every-4-week dosing after 6 months of therapy. The primary endpoint for MajesTEC‑1 was ORR. Key secondary endpoints included pharmacokinetics/pharmacodynamics, DoR, PFS, OS, uMRD, safety, and health-related quality of life. After a median follow-up of 14.1 months, an ORR of 63% was reported with teclistamab.⁴⁵

The baseline characteristics of patients enrolled on this trial were as expected.⁴⁶ The median age of enrolled patients was 64 years (range: 33-84), 58.2% of patients were male, 81.2% of patients were White, and 12.7% of patients were Black. In this population, 17% of patients had extramedullary disease, 25.7% had high-risk cytogenetics, and 87.7% had ISS stage I-II disease. The median time since diagnosis was 6 years, with a median of 5 previous lines of therapy; 70.3% of patients were penta exposed, and 30.3% were penta refractory. After a median follow-up of 23 months, 47 of 165 patients remained on treatment, with 42 patients receiving an every-2-week regimen and 9 patients receiving an every-4-week regimen.

With longer-term follow-up, response rates have remained quite consistent, with an ORR of 63.0%, including 45.5% of patients achieving a CR or better.⁴⁶ The ORR was higher among patients who had received ≤3 prior lines of therapy compared with those who had received >3 prior lines of therapy (74.4% vs 59.0%, respectively). The ORR in patients with high‑risk cytogenetics and/or extramedullary disease was slightly lower at 53.3%. 

Among patients who responded to teclistamab therapy, the responses were durable, with a median DoR of 21.6 months in all responding patients and 26.7 months among patients with a CR or better. The median time to achieve a CR or better was 4.6 months (range: 1.6-18.5).

The median PFS was 11 months for the overall population and 27 months for the 45.5% of patients with a CR or better. The median OS was 22 months for the overall population and was not reached for patients with a CR or better.  

Among 42 patients evaluable for MRD at Day 100, the rate of uMRD at a 10^-5 sensitivity was 81%. Among 54 patients evaluable at any point, the rate of uMRD was 81.5%.⁴⁶ These results suggest that teclistamab is inducing deep responses in these patients with advanced R/R MM.

In this longer follow-up, 1 AE led to a dose reduction and 8 AEs led to discontinuations (5 due to infection).⁴⁶ However, there were 7 treatment‑related deaths, including 4 due to COVID‑19, reflecting the period of enrollment during the pandemic and highlighting the increased susceptibility to infection in general. 

The most frequent hematologic AEs were neutropenia (71.5%; 65.5% grade 3/4), anemia (54.5%; 37.6% grade 3/4), thrombocytopenia (42.4%; 22.4% grade 3/4), and lymphopenia (36.4%; 34.5% grade 3/4). Although rates of neutropenia may have been a bit lower compared with rates reported with CAR T-cell therapy, neutropenia remains a predominant AE for these BCMA-targeted immunotherapies and is often grade 3 or 4. 

The most frequent nonhematologic AE was infection, occurring in 80% (55.2% grade 3/4) of patients. CRS was reported in 72% of patients, but only 1 (0.6%) patient experienced grade 3/4 CRS. 

ICANS occurred in 3% of patients with no grade 3/4 events. All cases of ICANS resolved. ICANS appears not to be as much of a concern here as we have seen with CAR T-cell therapy.  

Other common AEs included diarrhea (33.9%), pyrexia (31.5%), fatigue (29.1%), COVID-19 (29.1%), nausea (27.3%), cough (26.7%), injection-site erythema (26.1%), arthralgia (25.5%), headache (24.2%), constipation (21.8%), and hypogammaglobulinemia (20.6%).

The rate of infections associated with BCMA-targeted CAR T-cell therapy and bispecific antibodies has been high, and current studies are assessing the various infection profiles with these classes of agents, including the type and kinetics of infections. 

There have been quite a few opportunistic infections, including Pneumocystis jirovecii pneumonia and others associated with teclistamab therapy. In the first year after starting treatment with teclistamab, approximately 20% of patients developed a new-onset grade ≥3 infection every quarter. Beyond 1 year, the proportion of new‑onset grade ≥3 infection seems to decrease over time. Factors that likely contribute to this decline include decreased dosing frequency, better management of infections and prophylaxis, better disease control long term, and patient selection. This is important information as we try to reduce the risk of infections in these patients.

With the availability of multiple modalities to target BCMA for patients with MM, including the CAR T-cell therapies ide-cel and cilta-cel and the bispecific antibodies teclistamab and elranatamab, one of the big questions that arises is how to optimally sequence these different BCMA‑targeted therapies. 

Elranatamab is a novel humanized BCMA x CD3 bispecific antibody that was granted accelerated approved by the FDA in August 2023 for patients with R/R MM after ≥4 previous lines of therapy including a PI, an IMiD, and an anti-CD38 Ab. Results from cohort A of the phase II MagnetisMM-3 trial of elranatamab in patients with R/R MM who were BCMA therapy naive demonstrated an ORR of 61% at a median of 14.7 months of follow-up.⁴⁷ At ASCO 2023, Nooka and colleagues⁴⁸ presented a pooled analysis of 4 different MagnetisMM studies assessing the efficacy and safety of single-agent elranatamab in patients with R/R MM who have received ≥1 PI, ≥1 IMiD, and ≥1 anti-CD38 Ab and have been previously exposed to other BCMA-targeted therapy. The analysis included 87 patients with a median follow-up of 11.3 months.

The median age in the cohort of patients who had received any prior BCMA therapy was 66 years, with 14.9% aged 75 years and older.⁴⁸ Patients who received previous CAR T-cell therapy were slightly younger than those who received a BCMA-targeted ADC. 

High-risk cytogenetics were present in 24.1% of patients overall, 22% of patients who had received an ADC, and 30.6% of patients who had received CAR T-cell therapy. Approximately 23% of patients had stage III disease by the revised ISS criteria (27% with previous ADC therapy and 14% with previous CAR T-cell therapy), and 54% had extramedullary disease (56% with previous ADC and 47% with previous CAR T-cell therapy).

Patients had received a median of 7 prior lines of therapy (range: 3-19).⁴⁸ Prior BCMA therapy included an ADC for 59 patients (67.7%), CAR T-cell therapy for 36 patients (41.3%), and both therapies for 8 patients (9.2%). 

Most patients (85.1%) were penta exposed, and 55.2% were penta refractory. The proportion of patients with penta-refractory disease was 63.9% among those who had received CAR T-cell therapy and 54.2% among those who had received an ADC. The investigators defined penta refractory as prior exposure to 2 PIs, 2 IMiDs, and 1 anti-CD38 Ab.

The ORR for elranatamab in patients with any previous BCMA-targeted therapy was 46% (42.4% among patients with prior ADC therapy and 52.8% among patients with prior CAR T-cell therapy).⁴⁸ The median DoR was reasonable at 17.1 months (13.6 months with previous ADC therapy and NE with previous CAR T-cell therapy), but the median PFS and OS were short. The median PFS was 5.5 months (3.9 months with previous ADC therapy and 10.0 months with previous CAR T-cell therapy), and the median OS was 12.1 months in the overall patient population and with previous ADC or CAR T-cell therapy.  

There seemed to be a numerical difference in these data depending on whether patients had previously received an ADC vs CAR T-cell therapy, but given the small number of patients in this pooled analysis, these data are difficult to interpret. 

Overall, the ORR is lower than the 61% ORR reported in the MagnetisMM-3 trial in patients with R/R MM who were BCMA therapy naive, highlighting a drop in the response rate among patients who have previously received BCMA‑targeted therapy.⁴⁷ However, it is unclear how much of this reduction in efficacy is due to the more advanced disease state in these patients as we go through 2 different BCMA‑targeted therapies vs a clear loss of efficacy due to common mechanisms.

The safety profile with elranatamab after prior BCMA therapy was consistent to what we have observed with elranatamab in general.⁴⁸ We see hematologic toxicity including anemia (58.6%; 46.0% grade 3/4), neutropenia (44.8%; 41.4% grade 3/4), and thrombocytopenia (40.2%; 28.7% grade 3/4). In addition, CRS was reported in 65.5% of patients, but this was predominantly grade 1 or 2, with 2.3% developing grade 3/4 events. 

Infections were reported in 73.6% of patients, including 26.4% grade 3/4 events and 9.2% grade 5 events. The rates of infection with elranatamab after prior BCMA therapy observed in this analysis generally were comparable to those observed with other trials of BCMA‑targeted bispecific antibodies. 

There were 23 deaths related to TEAEs; these included disease progression (n = 11), COVID-19 (n = 5), septic shock or sepsis (n = 3), and other individual events. We hope that this rate of fatalities due to COVID-19 will be reduced moving forward, but it will be important to continue to monitor the risk of infection with this agent and BCMA-targeting therapies in general.

CRS developed in 65.5% of patients but was primarily grade 1 or 2, with only 2.3% of patients developing grade 3 CRS and no grade 4 events. The median time to onset of CRS was 2 days (range: 1-4), and the median time to resolution was also 2 days (range: 1-10). 

ICANS developed in 5 patients (5.7%), including 3 grade 2 events and 2 grade 3 events. The median time to onset of ICANS was 2 days (range: 1-3), as was the median time to ICANS resolution (range: 1-18).

Currently available CAR T-cell therapies are engineered by modifying a patient’s individual T-cells to express the CAR for personalized therapy.⁵² Because of this process, a challenge with CAR T-cell therapies is the extended manufacturing times required, which leads to reduced availability and delays in patient care. PHE885 is an investigational BCMA-directed CAR T-cell therapy that has a manufacturing time of <2 days and aims for a door-to-door time of <10 days in the United States.^53,54 

At ASCO 2023, investigators presented updated results from a phase I study evaluating the safety and efficacy of PHE885 in patients with R/R MM following ≥2 prior lines of therapy including a PI, an IMiD, and an anti-CD38 Ab.⁵⁵ The study enrolled 50 patients who underwent leukapheresis, lymphodepleting chemotherapy, and then a single infusion of PHE885. Escalating doses of PHE885 were studied, ranging from 2.5 x 10⁶ cells to 20 x 10⁶ cells. The primary endpoints were dose-limiting toxicities (DLTs) and safety. Secondary endpoints included the manufacturing success rate, ORR, and DoR. Results were presented after a median follow-up of 6.7 months.

The median age of enrolled patients was 65 years, (range: 45-81).⁵⁵ Patients had received a median of 4 prior lines of therapy (range: 2-10). Extramedullary disease was present in 33% of patients, and 36% of patients had high-risk cytogenetic abnormalities. As reflected in the enrollment criteria, 94% of patients were triple refractory, and 62% were penta refractory. 

Bridging therapy was administered to 34% of patients in this clinical trial protocol.

DLTs developed in 8 of 50 patients and included neutropenia, neurotoxicity, transient ejection fraction decrease, atrial fibrillation, transaminitis, and increases in lipase and serum amylase.⁵⁵ The ejection fraction decrease and atrial fibrillation were both in the context of CRS. 

There were 15 deaths on the study, including 13 due to disease progression, 1 due to hemorrhagic shock unrelated to treatment, and 1 due to lung infection in the context of PHE885-related myelosuppression.

Overall, 98% of patients had a TRAE of any grade, with 88% experiencing a TRAE of grade ≥3. The most frequent hematologic grade ≥3 TRAEs were neutropenia (80%), anemia (56%), thrombocytopenia (50%), and leukopenia (22%). Other nonhematologic grade ≥3 TRAEs included increased aspartate aminotransferase (16%) and decreased blood fibrinogen (12%).⁵⁵

CRS of any grade occurred in 96% of patients receiving PHE885, and 10% of patients developed grade 3 CRS. The median time to onset of CRS was 8 days (range: 2-16), and the median duration of CRS was 4 days (range: 1-19). 

ICANS developed in 20% of patients, and 6% developed grade 3 ICANS. The median time to ICANS onset was 10.5 days (range: 6-16). All ICANS events occurred concomitant to CRS and resolved. There were no reports of parkinsonism or other delayed neurotoxicity.⁵⁵

In this preliminary analysis, PHE885 was associated with an ORR of 98%, including 80% with a very good PR or better.⁵⁵ When broken down by individual dose levels, the ORR was 100% at all but the lowest dose.

The median time to first response was approximately 1 month, and the median time to best response was 2.8 months. Investigators noted that many responses deepened over time, with conversions to CR or sCR occurring up to 18 months after the PHE885 infusion.⁵⁵  

Across doses, MRD negativity was attained in 14 of 19 patients (74%) by Month 6.