The efficacy of teclistamab in patients with multiple myeloma and secondary plasma cell leukemia Open Access

TO THE EDITOR:

Plasma cell leukemia (PCL) is a rare malignancy constituting <5% of malignant plasma cell disorders.^1,2 It is an aggressive form of multiple myeloma (MM) characterized by the presence of ≥5% circulating plasma cells (CPCs).³ PCL occurs in 2 forms: primary PCL, which arises de novo without prior MM diagnosis, and secondary PCL (sPCL), a leukemic transformation of relapsed or refractory multiple myeloma (RRMM). There is an increasing incidence of sPCL attributed to the improved survival rates in patients with MM resulting from advancements in treatment and the gradual emergence of clonal selection over time.⁴ 

The development of sPCL remains associated with a poor prognosis. Analysis of the SEER (Surveillance, Epidemiology, and End Results) database demonstrated that outcomes of sPCL are dismal, with median progression-free survival (mPFS) of 2.5 months and median overall survival (mOS) of 3 to 5 months even with novel therapies such as proteasome inhibitors, immunomodulatory drugs, and CD38-targeting antibodies.^3,5,6 

The advent of T-cell–redirecting therapies has improved response rates in RRMM by targeting novel receptors, including B-cell maturation antigen (BCMA), which has emerged as a promising therapeutic target. Teclistamab, a T-cell–redirecting bispecific antibody, targets both CD3 expressed on T-cell surface and BCMA on surface of myeloma cells. Long-term follow-up results of the MajesTec trial have shown promising efficacy of teclistamab in RRMM with overall response rate (ORR) of 63% and mPFS and mOS of 11.4 and 22.2 months, respectively.⁷ However, patients with sPCL were excluded from these clinical trials, therefore the efficacy and safety of T-cell–redirecting therapies in this patient population remains largely unknown. Only a few case reports are available on BCMA chimeric antigen receptor (CAR) T-cell therapy showing responses in sPCL.^8,9 To our knowledge, till date, our study is the first to specifically investigate the use of bispecific antibody in sPCL.

We conducted a retrospective study of 16 patients with sPCL receiving at least 1 full dose of teclistamab at 3 academic centers in the United States (University of Arkansas for Medical Sciences [UAMS], Medical College of Wisconsin [MCW], and Rutgers Cancer Institute [RCI]). Data including patient demographics, baseline disease characteristics, and treatment details were collected and clinical outcomes, including response rates, PFS, and OS were analyzed. sPCL was defined as CPCs ≥5% of white blood cells in peripheral blood smear at the time of step-up dosing of teclistamab. International Myeloma Working Group response criteria were used to evaluate treatment responses. Cytokine release syndrome (CRS) and immune cell–associated neurotoxicity syndrome were graded according to American Society for Transplantation and Cellular Therapy guidelines. PFS was measured from the start of teclistamab to the date of disease progression, death, or last known follow-up. Logistic regression analysis was performed to assess the association between absolute lymphocyte count (ALC) or CPC and the probability of achieving a response. The relationship between CPC and PFS was assessed with Cox proportional hazards model.

A total of 16 patients with sPCL at the time of teclistamab initiation were included (Table 1). The median percentage of CPCs was 7.45% (range, 5%-82%). Most patients had triple class refractory (93%) MM with 62% being penta-refractory. The median age was 66 years, 50% male; 81% were White and 19% were Black. All patients had high-risk cytogenetics including various translocations t(4;14), t(14;16), deletion 17p, gain 1q21 and deletion 1p. Five out of 16 (31%) patients presented with soft tissue extramedullary disease (EMD) in addition to the sPCL. The median number of prior lines of therapy was 4.5 (2-10), and 87% had received a prior autologous stem cell transplant. Four patients (25%) received BCMA-targeted agents prior to teclistamab (3 received BCMA CAR T cells and 1 received linvoseltamab). CRS grade 1/2 was observed in 56% patients and 55% of those received tocilizumab. No CRS grade 3/4 were observed, whereas 2 patients had immune cell–associated neurotoxicity syndrome (1 with grade 2 and other with grade 3). ORR was 31% (5/16) with 1 person achieving complete response, whereas the best achieved response in the remaining 4 patients was partial response (Figure 1). None of the prior BCMA-exposed patients responded to teclistamab and mPFS was extremely short at 4 days, but 3 of 5 patients with EMD responded. The median time to first response was 2.2 months (range 0.6-4) among responders. The mPFS for the whole patient cohort was only 27 days (95% confidence interval [CI],12-85) which was mainly driven by nonresponders, whose mPFS was 18 days (n = 11; 95% CI, 4-31). Patients who had achieved a response had a mPFS of 4.6 months (n = 5; 95% CI, 0.6 to not evaluable). With a median follow-up of 61 days (95% CI, 12-543), all patients in this study had either progressed or died at the last follow-up. The mOS for nonresponders was 1.13 months (95% CI, 0.43 to not evaluable), whereas the mOS for patients who achieved a response was not reached (median follow-up was 14 months for responders). As the efficacy of teclistamab is T-cell dependent, we further looked at ALC as a predictor for response in sPCL, but did not find a significant difference between responders (median = 0.5 × 10³/μL; range, 0.2 × 10³/μL to 1.3 × 10³/μL) and nonresponders (median = 0.7 × 10³/μL; range, 0.2 × 10³/μL to 6.5 × 10³/μL), odds ratio 0.32 (95% CI, 0.02-3.87; P = .37). Additionally, the percentage of CPC was not significantly associated with clinical response (odds ratio, 1.006; 95% CI, 0.95-1.05; P = .79) or PFS (hazard ratio, 0.98; 95% CI, 0.96-1.01; P = .38).

Taken together, our study is the first to investigate the use of teclistamab in sPCL. We show on a patient cohort of 16 patients with sPCL, that teclistamab can offer response in some cases, yet the ORR of 31% seems substantially lower than in previously reported patients without sPCL. Strikingly, patients who did not respond had extremely short PFS of 18 days and OS of a month.

Four of 5 responders achieved only a partial response and most of them progressed within 5 months, indicating that long-term disease control remains challenging in this patient population, even with teclistamab therapy. The reasons for this poor response rate and duration of response are not quite clear, but could be due to de novo high-risk features,¹⁰ alterations in the bone marrow microenvironment,¹¹ or accumulation of genetic alterations.^12,13 Only 1 patient achieved complete response. However, it is important to mention that this patient received an off-label combination of teclistamab with venetoclax, underscoring that the response to single agent teclistamab might have been less effective. This patient eventually progressed after 14 months of therapy.

We further investigated potential clinical parameters as prognostic markers of response. Baseline ALC, as a previously described prognostic factor for BCMA CAR T-cell therapy,¹⁴ did not affect the response rate in this study. Similarly, the amount of CPC was not associated with depth of response, which is of interest as prior reports in MM have indicated that higher CPC levels have an adverse prognostic impact.^15-17 Yet, in PCL it has been shown that levels exceeding 5% are associated with worst outcomes compared to levels below 5%,¹⁸ with no further significant difference in outcomes beyond the 5% CPC threshold,^19,20 a pattern also observed in this study.

In conclusion, the treatment of sPCL remains challenging and the poor prognosis highlights the urgent need for innovative therapeutic strategies along with a better understanding of the disease at molecular level, to improve treatment outcomes. The study was approved by the institutional review board of all 3 academic centers (UAMS, MCW, RCI [approval number 260312]).

Contribution: A.S., M.R.S., M.M., and C.S. designed the study, collected and analyzed the data; A.S., S.T., A.B., B.D., M.Z., S.A.-H., F.v.R., A.D., N.S., M.R.S., M.M., and C.S. provided the patient data; and all authors read and reviewed the manuscript.

Conflict-of-interest disclosure: M.M. has received institutional research funding from Sanofi S.A., Bristol Myers Squibb, Celgene Corporation, and AbbVie; and has served on the advisory boards for Sanofi S.A., Bristol Myers Squibb/Celgene Corporation, Pfizer, Janssen Scientific Affairs LLC, and Legend Biotech. C.S. has served on advisory boards for Janssen, Pfizer, and Arcellx; and has received honoraria from Dava Oncology and Michael J. Hennessy Lifesciences. The remaining authors declare no competing financial interests.

Correspondence: Asis Shrestha, Division of Hematology-Oncology, Department of Internal Medicine, Myeloma Center, University of Arkansas for Medical Sciences, 449 Jack Stephens Drive, Little Rock, AR 72205; email: ashrestha@uams.edu; and Carolina Schinke, Division of Hematology-Oncology, Department of Internal Medicine, Myeloma Center, University of Arkansas for Medical Sciences, 449 Jack Stephens Drive, Little Rock, AR 72205; email: cschinke@uams.edu.