Talquetamab in heavily pretreated patients with multiple myeloma, including BCMA-refractory Open Access

Relapsed/refractory multiple myeloma (RRMM), especially disease that is refractory to a proteasome inhibitor (PI), immunomodulatory agent, and anti-CD38 monoclonal antibody, poses a unique treatment challenge. Until the recent development of bispecific antibodies (BsAbs) and chimeric antigen receptor (CAR) T-cell (CAR-T) therapies, options were limited, and prognosis was poor.^1,2 The identification of B-cell maturation antigen (BCMA) led to a crucial target for treatment of MM. CAR-Ts and BsAbs have been used to target BCMA with encouraging results. However, heavily pretreated patients tend to eventually relapse, with theoretical reduction in efficacy of subsequent BCMA-targeting agents.^3-5 More recently, the G-protein–coupled receptor class 5 member D (GPRC5D) was identified as a viable target based on its expression in all MM cells, including heavily pretreated MM.^6-9 The receptors are also noted to be highly expressed on keratinized tissue including skin and nails, and at the base of the epithelial columns supporting the filiform papillae.¹⁰ 

Talquetamab is a T-cell engager (TCE), BsAb targeting GPRC5D and CD3, designed to expand treatment options in MM. It received accelerated US Food and Drug Administration approval based on the MonumenTAL-1 trial, which demonstrated its safety and efficacy in patients with MM who had undergone at least 3 prior treatments, including BsAbs and/or CAR-T therapy.¹¹ Overall response rates (ORRs) of talquetamab in the MonumenTAL-1 trial were 70% and 64% with weekly dosing and every-other-week dosing, respectively. The most common side effect was cytokine release syndrome, primarily grade 1, whereas immune effector cell–associated neurotoxicity syndrome (ICANS) occurred in 10% of patients, mostly low grade. In addition, many patients reported skin- and nail-related events or dysgeusia aligning with the known targets on keratinized tissue. Grade 3 and grade 4 adverse events were noted in 90% of patients, mainly due to cytopenias, and infections affected 47% of patients. Importantly, no treatment-related deaths were reported.

Unfortunately, heavily pretreated patients with RRMM often have significant comorbidities, poor functional status, high-risk disease, or residual treatment-related adverse effects limiting trial eligibility.¹¹ Thus, the results of MonumenTAL-1 raise the question of how applicable these data are to real-world populations of heavily pretreated patients with RRMM. As we have acquired comfort with management of older patients who are frail, or those with significant comorbidities, specifically with BsAbs and CAR-Ts, these targeted options have become indispensable in such cohorts. In this multicenter study, we report the real-world efficacy and safety in patients with RRMM, including those who would not have met MonumenTAL-1 inclusion criteria. To our knowledge, this is the first real-world report of safety and efficacy of talquetamab in RRMM.

We performed a retrospective review of electronic medical records of patients with RRMM who received talquetamab monotherapy at 7 US academic medical centers. Patients who received at least 1 full dose of talquetamab before June 2024 were included. The institutional review board at each institution approved the study, including a waiver of informed consent. Furthermore, the study adhered to the ethical principles outlined in the Declaration of Helsinki.¹² 

We reviewed medical records for baseline demographic and descriptive variables, response data, and adverse events of the therapy. Each inclusion and exclusion variable from the MonumenTAL-1 trial was assessed among our patient population, however, only pertinent criteria are mentioned in this article. Responses to therapy were determined using the International Myeloma Work Group criteria.¹³ Adverse events were graded based on Common Terminology Criteria for Adverse Events version 5.0 criteria.¹⁴ The primary end point was determined to be ORR. Secondary end points consisted of overall survival (OS), progression-free survival (PFS), duration of response (DoR), and safety data. Statistical support was provided by the Biostatistics Core at the University of Iowa’s Holden Comprehensive Cancer Center with all authors having access to the data analysis.

Triple-class refractory MM was defined as refractoriness to PIs, immunomodulatory drugs, and anti-CD38 antibodies. Penta-drug refractory was considered refractoriness to 2 PIs, 2 immunomodulatory drugs, and an anti-CD38 antibody, typically lenalidomide, pomalidomide, bortezomib, carfilzomib, and either daratumumab or isatuximab. Extramedullary disease (EMD) in this study was defined as bone-independent tumors of plasma cells growing at anatomic sites outside of the bone marrow before receiving talquetamab. High-risk cytogenetics were defined as the presence of genomic aberrations such as deletion (17p), t(4;14), t(14;16), and/or gain/amplification of 1q on fluorescence in situ hybridization testing at any time before initiation of therapy.

The χ² or Fisher exact tests, as appropriate, were used to compare categorical patient characteristics, and analysis of variance was used to compare numeric patient characteristics between those who responded to treatment and those who did not. OS, PFS, and DoR probabilities were estimated and plotted using the Kaplan-Meier method. Cox regression was used to estimate the effect of baseline patient characteristics on OS, PFS, and DoR. For OS, time was calculated from treatment initiation to death due to any cause. For PFS, time was calculated from treatment initiation to progression or death due to any cause, whichever occurred first. For DoR, time was calculated from first response to progression or death due to any cause, whichever occurred first. Patients were censored at the time of last follow-up. Estimated effects of predictors were reported as hazard ratios (HRs) along with 95% confidence intervals (CIs). As appropriate, a multivariate model was built to include all patient characteristics found to be significant on univariate analysis. Of note, because of the highly collinear nature of the relationship between prior BsAb use and reception of either teclistamab or elrantamab for our whole patient cohort, multivariable analysis was constructed only to include prior BsAb reception because of its stronger relationship with PFS. All statistical testing was 2-sided and assessed for significance at the 5% level using SAS version 9.4 (SAS Institute, Cary, NC).

At the time of data cutoff, June 2024, 68 patients were treated with talquetamab. Patient and disease characteristics are described in Table 1. Notably, 53 (77.9%) patients were White, and 12 (17.6%) were African American, with a median age of 69 years (range, 38-86). The cohort included 45.6% patients with EMD and 44.1% with high-risk cytogenetics. Nearly 40% (38.2%) of our cohort would not have met inclusion criteria for MonumenTAL-1 trial, with most pertinent reasons including cytopenias, poor functional status, renal dysfunction, plasma cell leukemia, or no measurable disease (Table 1).¹¹ The median number of prior lines of therapy received was 7 (range, 4-17), with 88.2% of patients triple-class refractory and 70.6% penta-drug refractory. When data were available, talquetamab was given weekly or every other week, except in 4 patients who received it monthly after the step-up dosing. Treatment interruption was required in 25 patients, mainly because of toxicities (35%), followed by infections and disease progression.

The safety profile among our cohort is detailed in Figure 1 and Table 2. Notably, cytokine release syndrome was observed in 43 patients (63.2%) with nearly all events (42/43) being grade 1 or grade 2. ICANS was observed in 8 patients (11.8%), with 7 experiencing either grade 1 or grade 2 ICANS. Infectious complications were noted in 20 patients (29.4%), with respiratory infections accounting for 70% of these. All grade cytopenias at 30-days after the initial step-up dose of talquetamab for patients with available data consisted of anemia in 38 patients (64.4%), leukopenia in 27 (50%), and thrombocytopenia in 29 (52.7%). Regarding the severity of cytopenias, grade 3 or 4 events accounted for 9 patients with anemia (15.3%), 6 with leukopenia (11.2%), and 13 with thrombocytopenia (23.6%). Types and frequency of supportive treatment for cytopenias and hypogammaglobulinemia are detailed in Table 2. Skin-related events were noted in 34 patients (51.5%), with 24 (38.1%) developing a rash, 1 of which was a grade 3 event (1.6%). Nail-related events were present in 27 patients (39.7%). Dysgeusia was experienced by 53 patients (77.9%) and dry mouth by 28 (41.2%), with 21 patients experiencing resolution of dysgeusia (39.6%) and dry mouth resolved in 10 patients (35.7%). Weight loss of >10% during treatment was observed in 22 patients (32.8%), with 6 patients (27.3%) regaining some of the weight while continuing the drug.

The best ORR was 70.6% with 48 of 68 patients experiencing a response to therapy (Figure 2), 3 achieving stringent complete response (CR; 4.4%), 14 CR (20.6%), 18 very good partial response (VGPR; 26.5%), and 13 PR (19.1%) as illustrated in Figure 2. Significantly lower responses were observed in patients with EMD (51.6% vs 86.5%; P < .01) or prior bispecific TCE (59.2% vs 100%; P < .01), especially patients with prior teclistamab or elrantamab exposure (48.6% vs 93.3%; P < .01; Figure 3). Subgroup analysis for ORR is illustrated in Figure 3 and supplemental Table 1. Of 26 patients who would have been ineligible for MonumenTAL-1, ORR was 65.4%. The median DoR for all responders was 8.5 months, with an estimated 6-month DoR of 63% (95% CI, 42-77; Figure 4). None of the variables assessed had statistically significant association with the DoR on univariate analysis (Figure 5).

With a median follow-up of 3 months (range, 0.5-13.2), median PFS was 4.9 months; estimated 6-month PFS was 45% (95% CI, 30-58) and 12-month PFS was 24% (95% CI, 8-44; Figure 4). On univariate analysis (Figure 5), EMD (HR, 2.58; P < .01), prior TCE BsAb (HR, 3.52; P = .02), and prior use of teclistamab or elrantamab (HR, 2.27; P = .04) were associated with significantly shorter PFS. Prior use of both CAR-Ts and TCE BsAb (HR, 2.00; P = .05) also correlated with worse PFS, but this was borderline statistically significant. On multivariate analysis, a significant association of prior TCE BsAb exposure (HR, 3.87; P = .01) and EMD (HR, 2.81; P < .01) with PFS persisted.

The median OS was not reached at the time of data analysis. At median follow-up of 4.6 months, estimated 6-month OS was 74% (95% CI, 59-85) and 12-month OS was 67% (95% CI, 46-81; Figure 4). On univariate analysis (Figure 5), inability to meet MonumenTAL-1 eligibility (HR, 4.51; P = .01) had an increased risk of death, so did EMD (HR, 3.20; P = .05) but it was borderline statistically significant. High-risk cytogenetics, prior TCE exposure, and prior CAR-T exposure illustrated a trend toward worse OS, although this did not meet statistical significance. There were not enough events to perform a multivariate analysis of OS.

Within the cohort of 68 patients, 63 (92.6%) were noted to have been exposed to a prior BCMA-directed therapy (BDT). BDT distribution included CAR-Ts in 43 patients (68.3%), TCE in 49 patients (77.8%), and antibody-drug conjugate (ADC) in 6 patients (9.5%; supplemental Table 1). About half of the patients (n = 30, 47.6%) received ≥2 BDT, with 29 of 30 patients (93.5%) receiving both CAR-Ts and TCE. Overall, 52 patients (82.5%) were refractory to BDT. BDT was the immediate prior therapy before talquetamab for 25 patients (39.7%) and 37 patients (58.7%) received BDT within 6 months of starting talquetamab.

Regarding disease responsiveness, 43 patients (68.3%) who received prior BDT achieved a response to talquetamab. Best responses were observed as stringent CR in 3 patients (4.8%), CR in 11 (17.5%), VGPR in 16 (25.4%), and PR in 13 (20.6%). Thus, a response of VGPR or better was achieved in 30 patients (47.6%), and CR or better in 14 (22.2%) for those who had received prior BDT. ORR was lowest in the patients who had received ADC, although this was not found to be a significant difference of ORR as compared with those who had not received prior ADC (50.0% vs 70.2%; P = .37). Conversely, there was a significant reduction in ORR in patients who had prior treatment with TCE (59.2% vs 100%; P < .01). Patients who received both TCE and CAR-Ts illustrated a response rate of 58.6% (17/29). ORR was negatively associated with exposure to prior BDT within 6 months of talquetamab as compared with >6 months (56.8% vs 84.6%; P = .02) and if BDT was the immediate prior treatment before talquetamab (48% vs 80.6%; P < .01) (Figure 3). Median DoR was 7.6 months within BDT-exposed cohort, with an estimated 6-month DoR of 61% (95% CI, 39-77).

Within the BCMA-exposed cohort, median PFS was 4.8 months, with 6-month PFS of 42% (95% CI, 27-56). On univariate analysis, PFS illustrated an increased risk of progression with EMD (HR, 2.60; P < .01), immediate prior use of BDT (HR, 2.10; P = .04), and prior BDT treatment <6 months before talquetamab (HR, 2.32; P = .03). On multivariate analysis, BDT timing and immediate prior use of BDT were not significantly associated with PFS after accounting for EMD.

At a median follow-up of 4.6 months, median OS was not reached. Estimated 6-month OS was 75% (95% CI, 59-85) and estimated 12-month OS was 66% (95% CI, 43-82). On univariate analysis, BDT <6 months before treatment with talquetamab (HR, 4.82; P = .04) and not meeting MonumenTAL-1 eligibility criteria (HR, 4.51; P = .01) were associated with a worse OS. There were not enough events to perform a multivariate analysis of OS.

To our knowledge, this report is the first to illustrate the outcomes and safety profile of talquetamab in a real-world population of patients with RRMM. Overall, the efficacy of talquetamab in our cohort was comparable with the results of the MonumenTAL-1 trial, with an ORR of ∼71%. In addition, side effect profiles were similar, with cytopenias being the most frequent and severe adverse effect. Skin-related events, and nail changes were reported at a similar rate as MonumenTAL-1; however, dysgeusia was more common in our population (78%) compared with MonumenTAL-1 weekly/biweekly cohorts (57%/63%).¹¹ Notably, side effects such as dysgeusia, dry mouth, and weight loss recovered over time for only a small number of patients, in the order of ≤40% (Table 2). Without efficacious supportive treatment for these adverse events, it makes the continuous use of this drug quite challenging for patients. This was demonstrated by 43% of patients needing treatment interruption, mainly (35%) because of toxicities exclusive of infections (Table 1). Infection rate was lower (29%) than that seen in MonumenTAL-1 cohorts (47%/34%), which may be related to the lower rates of COVID-19 infection worldwide when our patients received talquetamab than in the MonumenTAL-1 population (enrolled between January 2018 and November 2021).¹¹ The infection rate is also significantly lower than with BCMA-directed BsAbs, 76% and 70% reported with teclistamab and elrantamab, respectively,^15,16 which is likely related to the target antigen (GPRC5D vs BCMA) as MonumenTAL-1 also reported lower infection rates.

With the availability of multiple treatment options for RRMM after 4 lines of therapy, now including talquetamab, optimal sequencing of therapies is a challenging task. Additionally, there is no determined standard of care, leading to a wide variety of options beside talquetamab,⁴ including but not limited to BDT, alkylating chemotherapy, selinexor, and elotuzumab. Moreover, there are limited data regarding treatment options after relapse or refractoriness to BDT. Further complicating the decision, myeloma that progresses through 1 type of BDT may be refractory to another group of BDT; for instance, BCMA TCE BsAb has shown to be less responsive to subsequent T-cell–directed therapy, such as BCMA CAR-Ts or belantamab.^3,17-21 

Our data show that talquetamab conferred decent responses in patients previously treated with BDT, thus illustrating the option to sequence talquetamab after BDT. This included patients who received prior TCE BsAb; however, the responses were lower than those who had no prior TCE BsAb exposure (Figure 3) and may not be sustained, as illustrated by worse PFS on multivariable analysis. Furthermore, our results illustrate worse response rate and PFS (on univariate analysis) based on the timing or immediate prior use of BDT relative to talquetamab (Figures 3 and 5). The finding of inferior responses within patients exposed to previous TCE BsAb may be explained by the hypothesis of T-cell exhaustion with the use of prior BCMA-bispecific TCE (which are currently dosed continuously).²² Our study indicates that one must exercise caution with immediate prior use of BDT, particularly within 6 months of using talquetamab; however, this suggestion is limited by the lack of significant ORR and survival differences on multivariate analysis, because of small cohort size. We noticed that prior exposure to CAR-Ts did not significantly affect response rate, PFS, or OS (Figures 3 and 5).

Importantly, nearly half of the patients within our cohort would have been ineligible for MonumenTAL-1. This subset of patients also exhibited no significant differences in response rates and PFS as compared with patients within our study that would have been eligible for MonumenTAL-1 or as compared with the cohort studied in MonumenTAL-1 itself. Thus, suboptimal functional status, comorbidities, and/or older age should not be a limitation to using talquetamab. It is noted that inability to meet MonumenTAL-1 criteria was associated with worse OS, which may potentially be because of the underlying severity of disease and/or significant comorbidities rather than suboptimal treatment. Our data highlight that the widespread use of these treatments in trial ineligible population is safe for most patients.

One known factor of high-risk disease, independent of cytogenetic risk, is extramedullary myeloma/plasmacytoma.^23-26 We noted in our study that EMD was associated with worse ORR and PFS after talquetamab, indicating that this group of patients is resistant to even the best possible therapies, and newer targets of treatment are needed. The underlying etiology is unclear and likely multifactorial. EMD is thought to be an immunological sanctuary with more suppressive tumor microenvironment associated with large and highly active tumors.^26-28 An effort to combine different targets of BsAbs, that is anti-GPRC5D and anti-BCMA, resulted in excellent responses in patients with EMD, however at the expense of more toxicity.²⁹ 

This study is limited by its retrospective nature with a relatively small cohort of patients among several institutions. Notably, details and management of side effects that require subjective assessment were not fully available, specifically nonhematological adverse events such as dysgeusia. Additionally, follow-up periods at different institutions may have varied, reducing the data for objective variables, such as cytopenias. Additionally, practice patterns regarding infection prophylaxis, frequency of IV immunoglobulin, or other supportive measures were difficult to obtain, which makes the most precise reason for our comparatively lower infection rate difficult to ascertain. Finally, because of a relatively small number of patients and need for longer follow-up, our ability to perform multivariate analyses and to comment on the survival outcomes was limited.

In conclusion, in this real-world experience, the efficacy and adverse effect profile of talquetamab in patients with RRMM are comparable with that of the Monumental-1 study, although nearly half of our patients would have been ineligible for the trial. Talquetamab appears to be a safe and effective option for heavily pretreated patients even for those who are BCMA refractory.

Our study provides the impetus for the need of larger studies with a longer follow-up period to evaluate the use of talquetamab in groups not included in clinical trials. Although talquetamab has set a new standard for efficacy in patients refractory to BDT, the unique mucocutaneous toxicities it comes with complemented by cytopenias limit its continuous use as demonstrated by 43% patients who had treatment interrupted because of toxicities; hence, time-limited therapy and/or infrequent dosing with such therapies need to be investigated.²² This is substantiated by comparable safety and efficacy of the weekly and biweekly dosing cohorts of talquetamab in MonumenTAL-1.¹¹ T-cell immunophenotyping may help guide the interval between talquetamab and prior TCE therapies by helping assess T-cell reserve/exhaustion.

Contribution: H.S. was the primary investigator of this study; B.T.L. provided the statistical analysis; and all authors contributed to data acquisition, data analysis, and editing of the final manuscript.

Conflict-of-interest disclosure: P.M. reports advisory board roles with Pfizer and Johnson & Johnson. C.S. reports research support through Poseida Therapeutics, Takeda, Janssen, and Bristol Myers Squibb (BMS). Y.K.R.T. reports advisory board roles with Novartis and Acrotech; and honoraria from Plexus Communications. N.A. reports advisory board roles with BMS, Kite/Gilead, Invivyd, and Legend Biotech; and consultancy for BMS and Kite/Gilead. A.M.K. reports research support from BMS and Sanofi; speakers bureau participation for Amgen and Sanofi; and reports travel support from BMS. H.H. reports advisory board role with Janssen and speakers bureau participation for Janssen and Karyopharm. B.P. reports advisory board roles for Johnson & Johnson and AbbVie and research support from BMS. S.A. reports advisory board roles for BMS and Johnson & Johnson; consultancy for Karopharm, Sanofi, and Pfizer; and grant support from GlaxoSmithKline and Amgen. The remaining authors declare no competing financial interests.

Correspondence: Hira Shaikh, University of Iowa Healthcare, 200 Hawkins Dr, C32 GH, Iowa City, IA 52242; email: [email protected].