Elotuzumab in combination with pomalidomide, bortezomib, and dexamethasone in relapsed and refractory multiple myeloma

There has been a dramatic improvement in outcomes in multiple myeloma in recent years from using immunomodulatory drugs, proteasome inhibitors, monoclonal antibodies, and now therapies targeting newer targets such as B-cell maturation antigen (BCMA) and G protein-coupled receptor class C group 5 member D (GPRC5D). However, multiple myeloma remains incurable, and most patients will relapse and develop increasingly refractory disease over time as they move from 1 line of therapy to the next. Moreover, the depth and durability of response decrease with each successive line of treatment.^1,2 This motivates the development of new treatment strategies for this increasingly challenging to treat patient population.

Several approaches include incorporation of novel agents with established regimens that have different mechanisms of action. Elotuzumab is a humanized immunoglobulin G1 immunostimulatory monoclonal antibody that targets signaling lymphocyte activation molecule F7 (SLAMF7), previously known as CS1.^3,4 SLAMF7 is a glycoprotein highly expressed on the surface of myeloma cells, natural killer (NK) cells, but not on other normal tissues. Elotuzumab improves overall response rate (ORR) and progression-free survival (PFS) in relapsed/refractory disease in multiple combinations including lenalidomide and dexamethasone (ELOQUENT-2)⁵; bortezomib and dexamethasone⁶; and pomalidomide and dexamethasone (ELOQUENT-3).⁷ Moreover, elotuzumab regimens with lenalidomide or pomalidomide also significantly improve overall survival (OS) in relapsed disease.^8,9 Pomalidomide is an immunomodulatory drug that is active in patients with lenalidomide-refractory disease.¹⁰ Furthermore, adding pomalidomide to the bortezomib and dexamethasone showed superior PFS and responses over the doublet in the OPTIMISMM phase 3 trial.¹¹ Motivated by these findings, we evaluated a 4-drug combination of elotuzumab with pomalidomide, bortezomib, and dexamethasone (elo-PVd) in a phase 2 study in relapsed and refractory disease.

We conducted a prospective, multicenter phase 2 trial of patients with relapsed and refractory multiple myeloma across 5 sites in the United States. Patients were eligible if they had received at least 1 prior line of therapy, including at least 2 cycles of lenalidomide and 2 cycles of a proteasome inhibitor (either in separate regimens or as part of the same regimen). All patients had relapsed and refractory disease, defined as disease progression within ≤60 days of completion of last therapy. Key inclusion criteria included adequate organ function and measurable disease (serum monoclonal protein ≥0.5 g/dL, urine monoclonal protein ≥200 mg/d, or serum free light chain ≥100 mg/L with abnormal free light chain criteria). We allowed patients with prior pomalidomide or bortezomib therapy to participate, including disease refractory to these drugs. Prior elotuzumab therapy was an exclusion.

Patients received treatment on a 28-day cycle. Elotuzumab was given 10 mg/kg IV weekly for cycles 1 to 2; 10 mg/kg IV on days 1 and 15 for cycles 3 to 8; and then 20 mg/kg on day 1 for cycles 9 and onward. Pomalidomide was administered 4 mg by mouth daily on days 1 to 21. Bortezomib was given 1.3 mg/m² subcutaneously on days 1, 8, and 15 for cycles 1 to 8. From cycles 9 and onward, bortezomib injections were given on days 1 and 15. On the week of elotuzumab infusion, dexamethasone was given as a premedication; 28 mg by mouth between 3 to 24 hours before the start of the elotuzumab infusion; and then 8 mg 45 to 90 minutes before the start of infusion. On the weeks without elotuzumab, dexamethasone 40 mg was given weekly. The dose of dexamethasone could be split over 2 days. Additional premedication for elotuzumab included diphenhydramine, ranitidine, and acetaminophen. All patients received prophylaxis for thromboembolism with aspirin or equivalent and prophylaxis for varicella-zoster virus reactivation.

The primary end point was the ORR per the International Myeloma Working Group criteria,¹² as determined by the treating investigator. Secondary end points included PFS and safety. Adverse events were graded according to Common Terminology Criteria for Adverse Events (CTCAE) version 4. All patients who received at least 1 cycle of treatment were evaluable for response. All participants who received any study drug were evaluated for toxicity. Low-pass whole-genome sequencing of circulating cell-free DNA was also performed; results of this have been previously published.¹³ 

The study used Simon 2-stage design to allow for early termination of the study if the combination lacked efficacy. In the first stage, 18 patients were to be accrued. If there were ≤7 responses in these 18 patients, the study was to be stopped. Otherwise, 28 additional patients were to be accrued for a total of 46 patients. The treatment was to be considered promising if ≥23 responses were observed in 46 patients. The probability of concluding the treatment is effective is 0.9, assuming a true response rate of 60%, and <0.10, assuming a true response rate of 40%. The probability of stopping early in the first stage if the treatment’s true response rate is 40% is 0.56. With 46 patients, the maximum exact binomial 90% confidence interval (CI) width is 0.26. PFS and OS were estimated by the Kaplan-Meier method.

The study was registered at ClinicalTrials.gov (identifier: NCT02718833). The protocol was approved by institutional review boards of all participating institutions. All patients provided written informed consent. The study was performed in accordance with the Declaration of Helsinki and the International Conference on Harmonisation for Good Clinical Practice. The data were collected, analyzed, and interpreted by the investigators. All authors had full access to the data in the study, wrote and revised the manuscript, and vouch for the accuracy and the completeness of the data.

We enrolled 48 patients who went on to receive treatment. Enrollment occurred across 5 sites from June 2016 to September 2018. In May 2017, we modified the eligibility criteria from 2 to ≥1 prior lines of treatment. Four patients were screened and did not start treatment. Table 1 summarizes the baseline demographics and characteristics. All patients were refractory to their last line of therapy and had a median of 3 prior lines of therapy (range, 1-9). All patients had prior lenalidomide and proteasome inhibitor, and this included 16 (33%) with prior pomalidomide and 14 (29%) with prior carfilzomib treatment. Bortezomib and pomalidomide were in the last line of therapy before enrolling in the trial for 24 and 6 patients, respectively. Fourteen (29%) had prior anti-CD38 monoclonal antibody and were all refractory to anti-CD38 monoclonal antibody. Seven patients had an anti-CD38 antibody as the last line of therapy; for the remaining 7, there was a median of 8 months from prior anti-CD38 antibody (range, 3-12). A quarter of patients had high-risk fluorescence in situ hybridization, defined as t(4;14), t(14;16), or del 17p. At the time of data cutoff, 5 patients continue on study treatment; 38 patients discontinued for progressive disease; 3 patients discontinued due to adverse events (sepsis, pulmonary embolism leading to death, or stroke); 1 patient underwent autologous stem cell transplantation; and 1 patient was lost to follow-up.

We observed an ORR of 56.3% (95% CI, 42.3-69.3) for the overall population (Table 2), with higher responses seen in patients with 1 prior line of therapy (73.7%; 95% CI, 51.2-88.2) than those with ≥2 prior lines of therapy (44.8%; 95% CI, 28.4-62.3). This includes responses in patients with prior anti-CD38 antibody therapy, for whom the ORR was 35.7% (95% CI, 16.3-61.3). At a median follow-up of 36.8 months, the median PFS was 10 months (95% CI, 7.75-20.1; Figure 1A). For patients with 1 prior line of therapy (n = 19; all lenalidomide refractory; bortezomib, n = 18; carfilzomib, n = 1), the median PFS was 23.4 months (95% CI, 10 to not reached); for ≥2 prior lines of therapy, the median PFS was 7.75 months (95% CI, 6.54-13.1; supplemental Figure 1). In patients with prior anti-CD38 antibody, the median PFS was 5.82 months (95% CI, 2.8-27.9), and when it was the last line of therapy, the median PFS was 7.69 months (95% CI, 2.83 to not reached). In patients with high-risk fluorescence in situ hybridization, the median PFS was 8.89 months (95% CI, 6.54 to not reached). The median OS for all treated patients was 25.2 months (95% CI, 18.4 to not reached; Figure 1B).

The most common hematologic and nonhematologic adverse events are shown in Table 3. The most common grade ≥3 adverse events were neutropenia (33%); infections, any (33%); lung infection (27%); hypophosphatemia (19%); and thrombocytopenia (15%). Two patients had febrile neutropenia. There were 2 deaths related to infection at the time of disease progression, due to Escherichia coli bacteremia and pneumonia. There were no infusion-related reactions from elotuzumab. There were 3 grade 3 cases of rash attributed to pomalidomide. In all cases, the rash resolved with supportive measures, and the patients were able to resume treatment on study with pomalidomide. Three patients discontinued treatment due to adverse events: stroke, pulmonary embolism leading to death, and infection.

Dose reductions occurred in the majority of patients. Pomalidomide was reduced to 3 mg in 25 patients, 2 mg in 5 patients, and 1 mg in 2 patients. Bortezomib was decreased to 1 mg/m² in 14 patients and 0.7 mg/m² in 4 patients. Reductions with oral dexamethasone occurred in 24 patients and with IV dexamethasone in 7 patients.

In this phase 2 trial of relapsed and refractory patients with a median of 3 prior lines of therapy and prior treatment with lenalidomide and a proteasome inhibitor, we found that the combination of elo-PVd yielded an ORR of 56.3% and a median PFS of 10 months. These response rates are notable given the extensive prior treatment history: pomalidomide (33%), carfilzomib (29%), and anti-CD38 antibody (29%); 13% of patients were penta-drug exposed (ie, exposure to these 3 agents along with lenalidomide and bortezomib). The safety profile of elo-PVd showed adverse events as expected from the drugs used in the regimen, with neutropenia and infections being the more common high-grade adverse events.

In patients with 1 prior line of treatment, the median PFS was 23.4 months, which compares favorably with the PFS of 17.84 months seen in the 1 prior line, lenalidomide-refractory subset of the OPTIMISMM trial with PVd.¹⁴ Within the limitations of cross-trial comparisons and the small numbers in this subset, this finding suggests that elotuzumab adds efficacy to the PVd triplet. Elotuzumab has been evaluated in a similar 4-drug regimen with carfilzomib, lenalidomide, and dexamethasone (KRd) at first relapse.¹⁵ Although the sample size with elo-KRd was small (N = 15), the median PFS was lower at 11.5 months. This may reflect that 67% of the participants in that trial had high-risk cytogenetics or reuse of lenalidomide in a population that was 80% lenalidomide refractory. Moreover, although the findings of elo-PVd may appear similar to ELOQUENT-3,⁷ in which the combination of elotuzumab, pomalidomide, and dexamethasone showed an ORR of 53% and a median PFS of 10.3 months, the results from the elo-PVd regimen underscores the value of including bortezomib. The patients in our trial were more heavily treated, including one-third treated with pomalidomide (which was an exclusion from ELOQUENT-3) and 29% treated with anti-CD38 antibody, which was only 2% of the elotuzumab arm in ELOQUENT-3.

The treatment landscape for multiple myeloma has dramatically evolved since this trial was developed. Anti-CD38 antibodies such as daratumumab and isatuximab have gained a strong presence in relapsed disease, initially building on trials showing single-agent activity as well as increased efficacy when combined with lenalidomide,¹⁶ pomalidomide,^17,18 bortezomib,¹⁹ or carfilzomib.^20,21 Similarly, anti-CD38 antibodies are now a standard of care in newly diagnosed patients, in both transplant-eligible²² and -ineligible patients.²³ However, data on the efficacy of elotuzumab after anti-CD38 antibody treatment are limited. Preclinical data suggest that prior anti-CD38 antibody may weaken the efficacy of elotuzumab. SLAMF7, in addition to being on myeloma cells, is also present on NK cells, and elotuzumab relies on NK cells for cell-mediated antibody-dependent cellular toxicity.^3,4,24 However, CD38 is also expressed on NK cells, and daratumumab treatment lowers NK cell numbers, which in turn may undermine the activity of elotuzumab.²⁵ Indeed, retrospective clinical series suggest that prior exposure to daratumumab reduces the effectiveness of subsequent elotuzumab, but this is not the case for elotuzumab treatment before daratumumab.^26,27 In our study of elo-PVd, 29% of patients (n = 14) had prior anti-CD38 antibody (daratumumab, n = 12; isatuximab, n = 2), with a median of 4 prior lines of treatment, and 100% exposure to prior lenalidomide and proteasome inhibitor, and therefore, they are also classified as triple-class exposed. In this triple-class–exposed patient subset, we show an ORR of 35.7% and a median PFS of 5.82 months. Moreover, these findings were similar in patients for whom anti-CD38 antibody was the last line of treatment (n = 7) before elo-PVd, with an ORR of 28.6% and a median PFS of 7.69 months (95% CI, 2.83 to not reached), although caution is noted in interpretation, given the small number of patients. The responses compare favorably in trials of a similar patient population with anti-CD38 antibody exposure, such as DREAMM-2 with belantamab mafodotin as a single agent, in which the response rate at the 2.5 mg/kg dose was 31% and a median PFS of only 2.9 months,²⁸ as well as with the MAMMOTH retrospective study, in which the PFS was 3.4 months, further emphasizing the value of combination approaches in this vulnerable population.²⁹ It should be noted that the current approval of elotuzumab with pomalidomide and dexamethasone is after 2 prior lines of therapy, including lenalidomide and a proteasome inhibitor. However, the field and practice are moving toward classifying patients and defining refractoriness to treatment by drug class exposure rather than number of lines of therapy, which may highlight the importance of moving elo-Pd and elo-PVd to earlier lines of therapy.^30,31 

Since the conception of this trial, the treatment of triple-class–exposed patients has evolved considerably, with the approvals of 2 chimeric antigen receptor (CAR) T-cell therapies, idecabtagene vicleucel (ide-cel) and ciltacabtagene autoleucel; 2 bispecific antibodies targeting BCMA, teclistamab and elranatamab; and a bispecific antibody targeting GPRC5D, talquetamab. However, access to CAR T-cell therapies is constrained to major medical centers with expertise in the management of cytokine release syndrome, neurotoxicity, and subsequent toxicities. There are also the challenges involved in the wait time for manufacturing of CAR T cells, which can take weeks, along with the coordination of care related to apheresis, bridging therapy, and ensuring caregiver support after administration of CAR T cells. Similarly, bispecific antibodies, which have the appeal of off-the-shelf availability, also have risks of cytokine release syndrome and neurotoxicity, which limit its use to experienced institutions. Recently, with belantamab mafodotin, the addition of pomalidomide and dexamethasone to belantamab mafodotin in the DREAMM-8 study showed significant activity in a triple-class–exposed patient population, but there is an attendant risk of ocular toxicity as well as the burden of evaluation by an eye care professional before dosing over the course of therapy.³² With these considerations, elo-PVd may emerge as a viable alternative when there is limited access to CAR T-cell therapies or bispecific antibodies.

Given the rapidly changing treatment landscape, the place of elotuzumab is undergoing further examination. In newly diagnosed patients, the addition of elotuzumab did not improve PFS in several trials: transplant-ineligible (ELOQUENT-1),³³ transplant-eligible (GMMG-HD6),³⁴ or high-risk patients (SWOG-1211)³⁵; but it did improve measurable residual disease negativity rate when combined with KRd in the DSMM XVII study.³⁶ On the contrary, in relapsed disease, elotuzumab has well-established efficacy in significantly improving PFS in combinations with lenalidomide and pomalidomide. Notably, elotuzumab successfully surmounted the higher bar of statistically significantly improving OS in these combinations.^8,9 In contrast, daratumumab³⁷ or isatuximab³⁸ combinations with pomalidomide numerically improved OS but not at a statistically significant level, even though OS benefit was seen in other settings with other regimens. Although cellular therapies and bispecific antibodies have shown unprecedented responses, relapses will inevitably occur because real-world experience has shown a median PFS of 8.5 months with ide-cel³⁹ and a 6-month PFS of 77% with ciltacabtagene autoleucel,⁴⁰ supporting an ongoing need for novel combinations and other treatment strategies using available drugs as well as new agents to improve outcomes. Indeed, there are ongoing trials, for example, exploring elotuzumab with mezigdomide and dexamethasone,⁴¹ including a trial focused on patients with prior anti-BCMA therapy (NCT05981209), and elotuzumab with iberdomide as consolidation after ide-cel (NCT06518551). Overall, the findings from this phase 2 trial of elo-PVd support the use of this combination in this changing landscape and merit further exploration.

The authors thank the patients and their caregivers and family members for their participation in the trial. The study drugs were provided by Bristol Myers Squibb (BMS) and Celgene.

This is an investigator-initiated study with funding provided by BMS and Celgene. Additional support was provided by the Multiple Myeloma Research Foundation and the Myeloma ACE Fund. N.S.R. is supported by the Paula and Rodger Riney Foundation.

Contribution: A.J.Y. and N.S.R. designed the study; A.J.Y., J.P.L., E.L.C., B.C.L., O.N., R.S.F., C.E.C., E.K.O., G.B., A.R.B., R.L.S., K.C.A., P.G.R., and N.S.R. recruited patients for the study; C.C.H., J.N.B., M.T.G., K.J.L., C.A.R., and D.X.A. cared for the patients; S.J.S., C.A.R., and D.X.A. provided trial support; A.J.Y., R.R., J.G.L., and N.S.R. analyzed the data; and all authors were involved in drafting, contributing to, and approving the manuscript.

Conflict-of-interest disclosure: A.J.Y. reports consulting for AbbVie, Adaptive Biotechnologies, Amgen, Bristol Myers Squibb (BMS), Celgene, GlaxoSmithKline (GSK), Johnson & Johnson (Janssen), Karyopharm, Oncopeptides, Pfizer, Prothena, Regeneron, Sanofi, Sebia, and Takeda; and research funding from Amgen, BMS, GSK, Johnson & Johnson (Janssen), and Sanofi (to institution). B.C.L. reports consulting for AbbVie, BMS, Janssen, Pfizer, and Sanofi; and research funding from Amgen and Janssen. O.N. reports consulting for BMS, Janssen, Takeda, Sanofi, and GPCR Therapeutics; honoraria from Pfizer; and research funding from Janssen. C.E.C. reports consulting for AbbVie, Genentech, GSK. Janssen, Pfizer, and Sanofi; speakers bureau fees from Binding Site; and research funding from GSK. E.K.O. reports consulting for Sanofi. G.B. reports consulting for Prothena. A.R.B. reports consulting for Adaptive, BeiGene, CSL Behring, Genzyme, Karyopharm, Pharmacyclics, and Sanofi. P.G.R. reports consulting for BMS/Celgene, Karyopharm, Oncopeptides, Sanofi, and GSK; and research funding from Oncopeptides. K.C.A. reports consulting for BMS, Janssen, Sanofi, Pfizer, Amgen, AstraZeneca, Genentech, and GSK; leadership roles with C4 Therapeutics, Starton Therapeutics, Window Therapeutics, Dynamic Cell Therapies, and Predicta Biosciences; stock or ownership interests with C4 Therapeutics, OncoPep, Raqia Therapeutics, NextRNA, Starton Therapeutics, Window Therapeutics, and Predicta Biosciences; and patents or royalties with C4 Therapeutics and OncoPep. N.S.R. reports consulting for AbbVie, Amgen, BMS, Janssen, Pfizer, Immuneel Therapeutics, GSK, K36 Therapeutics, Sanofi, and AstraZeneca; and research funding from Pfizer. The remaining authors declare no competing financial interests.

The current affiliation for C.E.C. is Karmanos Cancer Institute, Lansing, MI.

The current affiliation for E.K.O. is Dana-Farber Cancer Institute, Boston, MA.

The current affiliation for J.G.L. is Huntsman Cancer Institute, University of Utah, Salt Lake City, UT.

Correspondence: Noopur S. Raje, Massachusetts General Hospital Cancer Center, 55 Fruit St, Boston, MA 02114; email: nraje@mgh.harvard.edu.