Dear Editor,

With the advent of immunomodulatory agents (IMiD), proteasome inhibitors (PI), and anti-CD38 monoclonal antibodies (MoAb), the survival outcomes of patients with multiple myeloma (MM) has improved and is expected to continue to improve [1]. Elotuzumab is a MoAb that binds to signaling lymphocytic activation molecule F7 (SLAMF7) on the surface of myeloma cells and induces natural killer cell-mediated antibody-dependent cellular cytotoxicity of SLAMF7-expressing myeloma cells [2]. In the phase III ELOQUENT 2 trial, the addition of elotuzumab to lenalidomide and dexamethasone (ERd) led to the superior overall response rates (ORR), progression free survival (PFS) and overall survival (OS) compared to lenalidomide and dexamethasone in patients who had received 1–3 prior lines of therapy [LOT] [3, 4]. In the randomized phase II ELOQUENT 3 trial, the addition of elotuzumab to pomalidomide and dexamethasone (EPd) led to superior ORR, PFS and OS compared to pomalidomide and dexamethasone in patients with RRMM, about 60% of which had only received ≤3 prior LOT [5, 6]. A minority of patients derive long term disease control with elotuzumab-based regimens with 20% of patients treated with ERd in ELOQUENT 2 and about 4% of patients treated with EPd in ELOQUENT 3 remaining on treatment and being free of progression at 4 years [3, 6]. CD38 is highly expressed on NK-cells, and rapid NK-cell depletion in the peripheral blood and bone marrow has been seen in patients treated with daratumumab with NK cell numbers recovering 3–6 months after daratumumab has been discontinued [7]. Hence it is plausible that the efficacy of elotuzumab-based regimens may be diminished if used immediately after daratumumab-based therapy as elotuzumab depends on NK-cell function for efficacy. In ELOQUENT 2, no patients were previously treated with daratumumab and in ELOQUENT 3, <5% of patients were previously treated with daratumumab. Retrospective data suggest that the efficacy of elotuzumab is diminished with prior daratumumab exposure [8]. Data from the retrospective MAMMOTH study revealed that the use of elotuzumab + IMiD-based treatment in 19 (18 of which had triple class refractory [TCR] myeloma; myeloma that is refractory to a PI, IMiD and anti-CD38 MoAb) anti-CD38 MoAb refractory patients resulted in a median PFS of 2.6 months and median OS of 8.3 months [9]. The efficacy of elotuzumab-based regimens in the post-daratumumab progression and in the TCR setting have not been evaluated in any prospective clinical trials. Herein, we present the results of a prospective phase II study evaluating the efficacy of EPd in patients who have progressed after a prior daratumumab-containing regimen (NCT03713294).

This study was approved by the Institutional Review Board at Mayo Clinic Florida. Key inclusion criteria included measurable disease according to the International Myeloma Working Group (IMWG) criteria [10], RRMM after receiving at least 1 prior LOT, and disease refractory to daratumumab defined as no response (<partial response [PR]) or disease progression while on or ≤60 days of completing treatment with a daratumumab-containing regimen as part of any prior line of therapy. Key exclusion criteria were prior exposure to elotuzumab and refractoriness to pomalidomide. The primary objective was to estimate the ORR of utilizing EPd in patients with daratumumab-refractory, RRMM. Secondary objectives included evaluation of the complete response (CR) rate, PFS, OS and the safety profile of EPd in the daratumumab-refractory setting. EPd was administered as per the ELOQUENT 3 study [5] and patients were treated for up to 24 cycles. Adverse events (AEs) were graded according to the NCI CTCAE v5.0. Response rates were assessed as per the modified IMWG response criteria [10]. Clinical benefit rate (CBR) was defined as patients who achieved ≥ stable disease (SD). A modified two-stage Simon optimum design (accrual was not halted after first stage) was utilized and required 37 evaluable patients to test the null hypothesis that the true ORR in this patient population is at most 20% vs. the alternate hypothesis that the ORR was 40%.

Thirty-seven patients were enrolled and 36 received at least one full cycle of treatment. Baseline characteristics are shown in Table 1. Patients received a median of 4 (2–7) prior lines of therapy; 27 (73%) patients got daratumumab-based therapy as the immediate LOT prior to EPd (after a median of 4 prior lines [range 2–6]) whereas 10 (27%) patients got daratumumab-based therapy ≥1 line prior to EPd (after a median of 4.5 prior lines [range 2–7]). The ORR in all patients was 35% (n = 13), with no CRs, 1 (3%) VGPR, and 12 (32%) PR. Four patients (11%) achieved an MR, 15 (41%) achieved SD. The median follow-up time for patients still alive was 45.6 (95% CI 39.9–60.6) months. The median time to best response was 3 (95% CI 2.8–NA) months. The median PFS for the entire population was 3.7 (95% CI 2.9–11.1) months and the median OS was 56.7 (95% CI 27.2–NA) months (Fig. 1A). The study met its primary endpoint. The median time to next treatment was 7.6 (95% CI 6.2–20.4) months. Patients exposed to pomalidomide had received a median of 6 (4–7) prior lines of therapy compared to 4 (2–6) in patients not exposed to pomalidomide, p = 0.0013. The ORR in patients exposed to pomalidomide was 0% (n = 10), compared to 48% (n = 27) for patients not exposed to pomalidomide, p = 0.0067. However, the CRB for pomalidomide-exposed patients was 80% (8/10 patients achieved MR or SD). The median PFS for patients exposed to pomalidomide was 3.3 (95% CI 1.8–NA) months compared to 4.3 (95% CI 2.9–15.7) months for patients not exposed to pomalidomide, p = 0.0628. The median OS for patients exposed to pomalidomide was 45.5 (95% CI 13.9–NA) months compared to 56.7 (95% CI 27.1–NA) months for patients not exposed to pomalidomide, p = 0.6063. Patients that were TCR had received a median of 4 (2–7) prior LOT compared to 3.5 (2–6) for patients that were not TCR, p = 0.7338. The ORR in TCR patients was 37% (n = 27) compared to 30% (n = 10) in non-TCR patients, p = 1.0000. The median PFS in TCR patients was 4.03 (95% CI 2.9–14.0) months compared to 3.1 (95% CI 1.1–NA) months in non-TCR patients, p = 0.4931. The median OS in TCR patients was 56.7 (30.3–NA) months compared to NA (95% CI 13.2–NA) months in non-TCR patients, p = 0.9035 (Fig. 1B). The median prior LOT for patients who received daratumumab as the most immediate prior line of therapy was 4 (2–6) compared to 4.5 (2–7) for patients who received daratumumab ≥1 line prior to EPd, p = 0.1452. The ORR in patients with daratumumab as the immediate prior line of therapy was 40.7% (n = 27) compared to 20.0% (n = 10) in patients who had received daratumumab-based therapy ≥1 line prior to EPd, p = 0.4395. The median PFS of patients with daratumumab as the immediate prior line of therapy was 4.4 (95% CI 3.0–15.7) months compared to 2.8 (95% CI 2.0–NA) months in patients who had received daratumumab-based therapy ≥1 line prior to EPd (p = 0.0804). The median OS of patients with daratumumab as the immediate prior line of therapy was 56.7 (95% CI 45.5–NA) months compared to 20.5 (95% CI 5.3–NA) months in patients who had received daratumumab-based therapy ≥1 line prior to EPd (p = 0.0579). Patients who started EPd <3 months after the last dose of daratumumab (n = 24) achieved a PFS of 4.1 months and patients who started EPd >3 months after the last dose of daratumumab (n = 13) achieved a PFS of 3.6 months (p = 0.9965). The OS for patients who started EPd <3 months from the last dose of daratumumab (n = 24) was NR (30.3–NR) and the OS for patients who started EPD > 3 months from the last dose of daratumumab (n = 13) was 43.9 months (p = 0.3431). Patients who started EPd <6 months from the last dose of daratumumab (n = 27) achieved a PFS of 4.4 months and patients who started EPd >6 months from the last dose of daratumumab (n = 10) achieved a PFS of 3.3 months (p = 0.8062). The OS for patients who started EPd <6 months from the last dose of daratumumab (n = 27) was NR (30.3–NR) and the OS for patients who started EPd >6 months from the last dose of daratumumab (n = 10) was 41.9 months (p = 0.4831). Adverse events are summarized in Supplementary Table 1. The median PFS for patients who progressed after EPd and received subsequent treatment was 5.1 (95% CI 3.7–30.2) months and the ORR was 35.7% (n = 28). Patients were treated with regimens as shown in Table 1. The most common post-EPd progression regimens were BCMA-directed therapies which resulted in an ORR, median PFS and median OS of 40%, 7.6 (95% CI 5.1–NA) months and 39.0 (95% CI 30.6–NA) months and carfilzomib-based therapies which resulted in an ORR, median PFS, and median OS of 27.3%, 5.0 (95% CI 4.3–NA) months and 31.3 (95% CI 27.1–NA) months.

Table 1 Patient characteristics (N = 37).
Full size table
Fig. 1: Survival outcomes.
figure 1

A PFS and OS of study population. B PFS and OS of study population based on triple-class refractory status.

Full size image

The results of our study show the clinical efficacy of EPd in a predominantly TCR and heavily pre-treated, RRMM patient population. Compared to the ELOQUENT 3 study population treated with EPd (median prior LOT of 3, 0% TCR, and only 1 patient previously treated with daratumumab), the patients on our study received a median of 4 prior LOT, 75% were TCR and 100% were refractory to daratumumab which likely explains the lower ORR and PFS compared to ELOQUENT 3 [5]. In this study, the efficacy of EPd appeared to be as effective in TCR compared to non-TCR patients with similar PFS, ORR, and OS in both groups. The PFS of EPd-treated patients on our study coincide with the median PFS of 2.6 months noted in TCR patients treated with elotuzumab + IMiD-based regimens from the MAMMOTH study [9]. However, the OS of EPd treated patients on our study of 45.5 months was much higher than that reported in the MAMMOTH study (8.6 months) which is most likely due to the advent of BCMA-directed therapies such as CAR-T, bispecific antibodies, and antibody drug conjugates which have improved the traditionally dismal survival outcomes of TCR MM patients [11]. In this study, the median OS for patients treated with BCMA-directed therapies in the post-EPd progression setting was an impressive 36.3 months. While the ORR was lower for pomalidomide exposed vs. pomalidomide-naïve RRMM patients, both groups have similar survival outcomes when treated with EPd as no significant PFS or OS differences were noted between the two groups. A plausible explanation for this may be that the CBR in pomalidomide exposed patients was 80% with 8/10 pomalidomide-exposed patients achieving MR or SD and this stabilization of disease may have allowed patients to achieve a PFS similar to pomalidomide-naïve patients as the PFS for both groups was relatively short at 3–4 months. Interestingly, administration of EPd immediately after a daratumumab-based regimen does not appear to effect the efficacy of EPd as there were no significant differences in ORR and PFS between patients who received daratumumab-based therapy immediately prior to EPd or ≥1 line prior to EPd and in fact, patients who received a daratumumab-based regimen immediately prior to EPd had a superior OS possibly due to the fact that they were less heavily pretreated. In addition, it is known that daratumumab can deplete NK cells and NK cells recover 3–6 months after the last dose of daratumumab [7], yet we found that whether a patient started EPd <3 vs. >3 months after the last dose of daratumumab or <6 vs. >6 months after the last dose of daratumumab, there were no differences in survival outcomes (PFS and OS) suggesting that the timing of prior daratumumab does not impair the efficacy of elotuzumab-based regimens as previously reported [12]. Our trial enrolled a considerable amount of African American patients at 22% which is in line with NCI/FDA goals of increasing accrual of minority patient populations onto clinical trials as African Americans represent 20% of the US myeloma population but have traditionally represented <5% of enrolled patients in US myeloma clinical trials [13]. While there are more efficacious therapies for TCR RRMM such as anti-BCMA CAR-T and bispecific antibodies, talquetamab, carfilzomib-based, selinexor-based or venetoclax-based (for t(11;14) MM) combinations [14], EPd is nonetheless an active therapeutic regimen in this patient population as it shows efficacy in TCR MM patients, including those who have received >1–3 prior LOT. Emerging data is showing that the optimal sequence of CAR-T and bispecific antibodies in TCR MM patients appears to be CAR-T first followed by bispecific antibodies in the post-CAR-T relapse setting [15, 16] hence EPd can be an effective bridging therapy for TCR RRMM patients that are also carfilzomib refractory and are awaiting CAR-T manufacturing.