Subsequent anti-myeloma therapy after idecabtagene vicleucel treatment in patients with relapsed/refractory multiple myeloma: A single center analysis.

Dear Editor,

Due to the advent of novel agents such as proteasome inhibitors (PI), immunomodulatory agents (IMiD), and anti-CD38 monoclonal antibodies (MoAb), the survival of patients with multiple myeloma (MM) has improved and is expected to continue to improve [1, 2]. Nonetheless, relapse is inevitable with subsequent remissions being shorter due to acquired drug resistance and development of refractory disease [3]. Patients that are triple-class refractory (refractory to a PI, IMiD, and anti-CD38 MoAb) have a dismal prognosis with a reported overall response rate (ORR) of 31%, a median progression-free survival (PFS) of 3.4 months, and a median overall survival (OS) of 9.3 months with the subsequent treatment regimens following anti-CD38 MoAb failure [4]. The recently approved B-cell maturation antigen (BCMA)- directed chimeric antigen receptor-modified (CAR) T cell therapy, idecabtagene vicleucel (ide-cel), is indicated for patients who have progressed following four or more lines of therapy [5]. The efficacy of ide-cel therapy was favorable with an ORR of 73%, 26% minimal residual disease (MRD) negativity rate, and an OS of 24.8 months [6]. However, patients continue to relapse even after anti-BCMA CAR-T with little available data on prognosis and management recommendations. Herein, we present the outcomes of 7 patients who received subsequent anti-myeloma therapies (sAMT) following progression on ide-cel.

We reviewed the treatment outcomes of 7 patients who developed progressive disease (PD) and were treated with sAMT following treatment with ide-cel on a clinical trial at Mayo Clinic Florida between December 2019 and November 2020. All patients had received 2–4 prior lines of therapy and prior treatment with at least a PI, and IMiD and an anti-CD38 MoAb. Data were collected retrospectively from hospital records. Response rates were defined based on the International Myeloma Working Group Criteria [7]. PFS and OS were analyzed using Kaplan–Meier analysis and Cox regression. Time-to-event variables were compared using unadjusted Cox proportional hazards regression models.

Patient characteristics are shown in Table 1. The median follow-up time following initiation of first line therapy following CAR-T relapse was 8 months (95% CI 2-NR). One (14.2%) patient was R-ISS I, 5 (71.4%) were R-ISS II, and 1 (14.2%) was R-ISS-III at the time of CAR-T therapy. Two (33%) patients had high risk FISH abnormalities. Seven (100%) patients were triple-class exposed, 7 (100%) were triple-class refractory, and none were pentarefractory at the time of CAR-T therapy. Following CAR-T therapy, 2 (28.5%) patients were MRD positive even at 90 days post CAR-T and 5 (71.4%) patients achieved MRD negativity in the bone marrow by day 90. Two (28.5%) patients had a PET-CT that was negative post CAR-T, 4 (57.1%) patients had a persistently positive PET-CT post-CAR-T, one patient did not have a post-CAR-T PET-CT. The median time to progressive disease following CAR-T therapy was 211 days (range 57–277).

Table 1

Baseline patient characteristics.

Following disease progression, all 7 patients received a subsequent line of therapy. Regimens are shown in Table 1; 2 (28.5%) patients received alkylator-based therapy, 4 (57.1%) patients received carfilzomib-based therapy and 2 (28.5%) received MoAb-based therapy with 1 (14.2%) receiving a quadruplet containing daratumumab and carfilzomib. The median number of treatment cycles was 1 (range 1–4). The ORR to the first sAMT was 28.5% with 1 patient achieving a very good partial response (VGPR) and one patient achieving a partial response (PR). The clinical benefit rate was 57.1% with one patient each achieving a VGPR, PR, minor response (MR), and stable disease (SD). Patients who received a carfilzomib-based regimen (n = 4) were the only responders with one each achieving a VGPR, PR, and MR; 3 of these patients had never been exposed to carfilzomib. The median PFS following 1st post-CAR-T sAMT was 2 months (95% CI:0-NR).

Four (57.1%) patients went on to receive a second and/or third sAMT. Two patients received carfilzomib-based therapy, 1 patient received elotuzumab-based therapy and 1 received high dose cyclophosphamide 1500 mg/m2 for 2 consecutive days. The ORR to second sAMT was 50% with 2 patients achieving a VGPR. The CBR was 75% (n = 3) with 2 patients achieving a VGPR and one achieving an MR. The median number of cycles was 1.5 (range 1–4). The median PFS after second sAMT was 1 month (95% CI 1-NR). Two patients received belantamab mafodotin as a third-line, post-CAR-T sAMT and the ORR was 0% with both patients developing progressive disease as the best response to therapy.

The median OS after initiation of sAMT in the post-CAR-T relapse setting was 5 months (95% CI 4-NR).

In our small cohort of 7 patients who progressed following anti-BCMA CAR-T cell therapy with ide-cel, outcomes were dismal with short-lived ORR and PFS and a median OS of 5 months following sAMT. Carfilzomib-based therapy appears to be transiently active in both the first and second-line of therapy following progression on anti-BCMA CAR-T, however, responders were mainly carfilzomib-naïve and responses were short-lived. While previous reports [8, 9] have shown that anti-BCMA agents can be sequenced, 2 patients in our cohort who received the anti-BCMA antibody-drug conjugate belantamab mafodotin as third line sAMT following anti-BCMA CAR-T progression did not respond.

The ideal treatment strategy for patients who progress following anti-BCMA CAR-T cell therapy is currently unknown. Given that most patients are at least triple-class refractory, if not pentarefractory by the time they receive anti-BCMA CAR-T cell therapy, using standard FDA approved anti-myeloma agents in the PI, IMiD, or anti-CD38 MoAb class upon relapse may not be possible. Recent data has shown that in 68 patients who progressed following ide-cel in the phase I KarMMa I study [NCT02658929] and received sAMT, the most frequent subsequent anti-myeloma drug classes were corticosteroids (n = 58) and PIs (n = 47), while the most frequent agents were dexamethasone (n = 56) and carfilzomib (n = 32). Anti-BCMA agents included antibody-drug conjugate belantamab mafodotin (n = 10). ORR to sAMT were not reported. In patients who received sAMT, the median duration of first sAMT was 48 days and PFS was 15.5 months [10]. A possible discrepancy between the KarMMA I study and our study in terms of PFS to first sAMT may be that in KArMMa 1, the median time between MM diagnosis and CAR-T was 60 months, whereas it was 37 months (range:19–95) in our cohort of patients [11]. Therefore, the patients in our study may have had more biologically aggressive disease.

In another retrospective study of 31 patients who developed PD after anti-BCMA CAR-T cell therapy on a clinical trial, patients received a median of 2 additional treatment lines (range 0–8). The most common initial treatment after CAR-T relapse was chemotherapy with (V)DCEP or VD-PACE (10/28, 36%). Five patients (18%) were treated with a bispecific antibody immediately after CAR-T and 12 (43%) received bispecific antibody therapy at any point after CAR-T. Selinexor-based regimens were used in 3 (11%) patients immediately after and in 10 (36%) at any point after CAR-T relapse. The ORR to the subsequent therapies in this cohort of patient was 46%. Median time to progression was 105 days (95% CI: 78–204) for the initial treatment after CAR-T [12]. In a retrospective study of 11 MM patients who relapsed after prior anti-BCMA directed therapy, use of selinexor-containing regimens resulted in an ORR of 54.5%, a 6 months PFS probability of 68.6%, and a median OS of 10.5 months [13].

It remains to be determined whether treating with another BCMA-targeted agent or switching to a sAMT that targets a different antigen or cellular pathway is the ideal treatment strategy following progression on anti-BCMA CAR-T. A better understanding of the mechanisms of resistance to anti-BCMA CAR-T will be key to determining if anti-BCMA agents can be sequenced. It is currently postulated that antigen loss, BCMA shedding from plasma cell surface, secretion of anti-drug antibodies, T-cell exhaustion, and the emergence of a nonpermissive microenvironment are potential mechanisms of resistance to BCMA-directed therapies, and understanding which resistance mechanism occurs in an individual patient will be key to helping tailor sAMT [9].

Anti-myeloma agents that are not IMiDs, PIs, anti-CD38 MoAb, alkylators, corticosteroids, or anti-BCMA directed therapies have been evaluated in the post-BCMA directed therapy setting and have shown efficacy. In a phase I study evaluating talquetamab, a G-protein-coupled receptor family C group 5 member D x CD3-targeting bispecific antibody (NCT03399799), 31 evaluable patients achieved an ORR of around 70% and 30% of those patients had received prior anti-BCMA directed therapy [14]. In a phase I trial of cevostamab (NCT03275103), an Fc receptor-homolog 5 × CD3 bispecific antibody, out of 158 evaluable patients, 33.8% of patients had received ≥1 prior anti-BCMA targeting agent. At target dose levels >90 mg, ORRs in pts with prior exposure to CAR-Ts, bispecific antibodies, antibody-drug conjugates, and anti-BCMA targeting agents were 44.4% (4/9 pts), 33.3% (3/9), 50.0% (7/14), and 36.4% (8/22), respectively [15].

In conclusion, the prognosis of MM patients following progression on anti-BCMA CAR-T cell therapy in the relapsed/refractory setting is dismal, and novel therapeutic strategies are necessary to improve the outcomes of these patients. As patients are likely to have exhausted all FDA-approved therapeutic options when they progress following anti-BCMA CAR-T, novel agents that target novel plasma cell antigens or novel cellular pathways are the likely therapeutic strategies necessary to improve outcomes in this patient population.