Introduction

Multiple myeloma (MM) is a common malignant neoplasm of plasma cells. The treatment of MM has evolved rapidly over the last two decades, with the advent of novel therapeutics leading to significant increases in patient survival. Despite substantial improvement in patient outcomes, MM remains incurable in the vast majority of cases and disease relapse is expected. The mainstays of treatment for newly diagnosed and relapsed patients include various combinations of proteosome inhibitors (PIs) (bortezomib, carfilzomib, ixazomib), immunomodulatory drugs (IMIDs) (lenalidomide, pomalidomide, thalidomide), monoclonal anti-CD3 antibodies (mABs) (daratumumab, isatuximab), and autologous stem cell transplantation (ASCT) in select fit patients.

Ultimately, patients become triple-class refractory (refractory to a PI, IMID, MAB) or, increasingly, penta-refractory (refractory to bortezomib, carfilzomib, lenalidomide, pomalidomide, and daratumumab), and new therapeutic options are required. The review focuses on belantamab mafodotin, selinexor, and melflufen as treatment for relapsed/refractory multiple myeloma (RRMM). Published clinical trial data for these agents is summarized in Table 1 (study designs), Table 2 (efficacy data), and Table 3 (toxicity data).

Table 1 Study design of pivotal phase 2/3 trials of melflufen, selinexor, and belantamab mafodotin-based regimens
Full size table
Table 2 Efficacy outcomes of melflufen, selinexor, and belantamab mafodotin based on selected phase 2/3 pivotal clinical trials
Full size table
Table 3 Adverse effects of melflufen, selinexor, and belantamab mafodotin based on selected phase 2/3 pivotal clinical trials
Full size table

Belantamab Mafodotin

Belantamab mafodotin (belamaf) is a first-in-class afucosylated humanized anti-BCMA IgG1 monoclonal antibody conjugated to microtubule disrupting monomethyl auristatin F (MMAF). It binds to FcγRIIIa on plasma cells resulting in activation and recruitment of immune effector cells and enhanced antibody-dependent cell-mediated cytotoxicity (ADCC) [10].

The initial phase 1/2 trial (DREAMM-1) in patients with heavily pre-treated RRMM administered belamaf monotherapy in a dose escalation from 0.03 to 4.60 mg/kg over 1 h by intravenous (IV) every 3 weeks for up to 16 cycles [6]. While no maximum tolerated dose (MTD) was reached in phase 1, the trend towards the increased frequency of grade 3/4 corneal events at higher doses and lower clinical activity at doses ≤ 2.5 mg/kg led to a recommended dose of recommended phase 2 dose (RP2D) of 3.4 mg/kg2. The overall response rate (ORR) in the phase 2 portion was 60% (95% CI: 42.1–76.1); 46.2% (95% CI: 19.2–74.9) in patients who had previously had > 5 prior therapies and 68.2% (95% CI: 45.1–86.1) in those with ≤ 5 prior lines [6]. Notably, single-agent belamaf had a 38.5% (95% CI: 13.9–68.4) in patients refractory to an IMID and PI and daratumumab exposed [7]. At a median follow-up of 12.5 months, the median progressive-free survival (PFS) was 12 months (95% CI: 3.1–not estimable), 6.2 months in those refractory to IMID PIs and daratumumab exposed, 7.9 months in those refractory to IMIDs and PIs, and 15.7 months in those not previously treated with daratumumab [7].

The DREAMM-2 study was a randomized phase 2 trial of single-agent belamaf (2.5 mg/kg vs. 3.4 mg/kg every 3 weeks) in patients with RRMM refractory to a PI, IMID, and anti-CD38 mAB [8]. Approximately one-third of patients in each cohort received an anti-CD38 mAB as the last prior therapy. The ORR was 31% (97.5% CI: 20.8–42.6) at the 2.5 mg/kg dose and 34% (23.9–46.0) at the 3.4 mg/kg dose, with 60% and 59% of responders achieving ≥ VGPR in each cohort, respectively [8]. At a median follow-up of 6.3 months (2.5 mg/kg group)/6.9 months (3.4 mg/kg group), the median PFS was 2.9 months (95% CI: 2.1–3.7) and 4.9 months (2.3–6.2), with OS not yet reached. Longer term median follow-up of 13.0 months was recently reported for the 2.5 mg/kg group, showing a median PFS of 2.8 months (95% CI: 1.6–3.6) and OS of 13.7 months (95% CI: 9.9–NR) [11]. Both study cohorts had similar clinical efficacy, but the 2.5 mg/kg dose had a favorable safety profile and was therefore recommended for further trials. The promising single-agent activity and PFS demonstrated study has supported the development of further clinical trials of belamaf combination regimens.

The forthcoming belamaf studies are summarized in Table 4. DREAMM-3 is a randomized phase 3 trial of belamaf in combination with pomalidomide and dexamethasone (Pd) versus Pd which is currently enrolling, in which belantamab is being administered in 4 cohorts: 1.92 mg/kg, 2.5 mg/kg single dose, 2.5 mg/kg split on days 1 and 8, or 3.4 mg/kg split on days 1 and 8 [12]. This will build upon the data from the Canadian Myeloma Research Group (CMRG) ALGONQUINN trial, which is a phase 1/2 study of belamaf-Pd exploring multiple doses of belamaf in the triplet combination. This study identified an MTD of belamaf 2.5 mg/kg combined with standard dose Pd, and preliminary data on 60 patients with a median of 3 prior lines of therapy showed a promising ORR of 88.9% with ≥ VGPR 62% and median PFS 24.2 months [13]. The combination of belamaf-Pd is also being studied in the DREAMM-8 trial, a large randomized study of belamaf-Pd with belamaf dosed at 2.5 mg/kg cycle 1 and 1.92 mg/kg cycle 2 onwards versus pomalidomide, bortezomib, and dexamethasone (PVd), the results of which are eagerly anticipated [14].

Table 4 Active* clinical trials including belantamab mafodotin (where recruitment is ongoing or anticipated)
Full size table

Preliminary data from part 1 of the DREAMM-4, a phase 1/2 trial of belamaf in combination with pembrolizumab, was recently presented [15]. Patients were dosed at either belamaf 2.5 mg/kg or 3.4 mg/kg with pembrolizumab 200 mg IV every 3 weeks. At a median follow-up duration of 6.8 months in the 2.5 mg/kg group and 4.2 months in the 3.4 mg/kg group, keratopathy was the most common AE with similar AE profiles in both groups and no DLTs [15]. ORR was 67% in the 2.5 mg/kg group and 43% in the 3.4 mg/kg group, with the 2.5 mg/kg dose selected for part 2 [15]. At a median follow-up of 14.7 months, preliminary efficacy results of 34 patients (6 in part 1 and 28 in part 2) who had a median of 5 prior lines of therapy, 29% high-risk cytogenetics and 26% extramedullary disease, showed a median ORR of 47% with 63% of responders achieving ≥ VGPR [16]. Median PFS was months with no new safety signals observed [16].

The DREAMM-5 study is a large phase 1/2 platform study currently accruing in patients post ≥ 3 prior therapies (including a PI, IMID, and anti-CD38 MAB), evaluating the combination of belamaf with GSK3174998 (OX 40 agonist), feladilimab (GSK3359609, ICOS agonist), nirogacestat (gamma-secretase inhibitor), and dostarlimab (anti-PD1 MAB) [17]. Preliminary results of the feladilimab group showed a promising ORR of 48% and a manageable safety profile among 23 heavily pre-treated patients (median 5 prior lines of therapy) [18]. The initial results of 10 patients, with a median of 4.5 prior lines, treated with low-dose belamaf (0.95 mg/kg Q3W) and nirogacestat 100 mg BID in the dose escalation phase of this substudy were recently presented [19]. The ORR was 60% with one-third of responders achieving VGPR, with 70% of patients experiencing grade 1–2 keratopathy, diarrhea, and hypophosphatemia [19].

DREAMM-6 is an ongoing two-part phase 1/2 study of belamaf combined with either lenalidomide or bortezomib for RRMM after ≥ 1 prior treatment. Preliminary results of 18 patients on the belamaf (2.5 mg/kg single dose cohort), bortezomib, and dexamethasone arm showed an ORR of 78% with ≥ VGPR rate of 50%, with universal ocular toxicity that was manageable [20]. An interim analysis of 45 patients with median 3 prior lines treated on the belamaf (1.9 mg kg Q8W or Q4W; 2.5 mg/kg Q4W or Q4W split dose) (50% days 1 and 8) with lenalidomide and dexamethasone cohort was recently presented [21]. The ORR was 75% in the belamaf 1.9 mg/kg Q4W group (n = 12), 42% in the 1.9 mg/kg Q8W group (n-4), 63% in the 2.5 mg/kg group (n = 16), and 69% in the 2.5 mg/kg Q4W split group (n = 13), with median PFS not yet reached at the time of data cutoff [21]. No new safety signals were identified. The phase 3 DREAMM-7 study is ongoing, randomizing patients with relapsed MM and ≥ 1 prior therapy to belamaf in combination with bortezomib and dexamethasone (BVd) vs. daratumumab, bortezomib, and dexamethasone (DVd) [22].

Belamaf is also being evaluated in patients with newly diagnosed MM. Preliminary results of a phase 1/2 study of the combination of belamaf, lenalidomide, and dexamethasone in newly diagnosed patients ineligible for transplant showed no new safety signals with the first 18 patients enrolled [23]. DREAMM-9 is an ongoing phase 3 study in newly diagnosed transplant-ineligible patients which randomizes patients to different dose intensities and schedules of belamaf in combination with bortezomib, lenalidomide, and dexamethasone (VRd). A preliminary report on 12 patients showed no new or unexpected safety signals [24].

Toxicity

The most challenging adverse event observed with belamaf is corneal toxicity, which has been described previously with monomethyl auristatin F (MMAF) antibody combinations [25], and is attributable to MMAF, possibly due to non-specific drug uptake by dividing cells in the basal epithelial layer of the cornea [26]. Corneal examinations including a slit-lamp examination for keratopathy and best-corrected visual acuity (BCVA) assessment are currently considered standard for monitoring patients receiving belamaf, with examination findings combined and graded for toxicity [27]. The keratopathy and visual acuity (KVA scale) has been developed for identifying and grading corneal events [11]. Corneal examination findings range from mild to severe superficial keratopathy with or without microcysts, sub-epithelial haze, or stromal opacity, as well as corneal epithelial defects such as ulcers [27]. Symptomatically, ocular manifestations include dry eyes, photophobia, blurry vision, and reduced visual acuity [6]. The majority of patients experience corneal toxicity, and it is the most common reason for dose delays and reductions [27]. Fortunately, the vast majority of patients recover with a median time to recovery of 86.5 days (range 8–358) [7,8,9, 11].

The mainstay of management of corneal events is dose reduction/modification [11]. The original trial protocols recommended prophylactic corticosteroid eye drops four times daily for 7 days and preservative-free lubricating drops 4–8 times daily [6, 8]. An ocular substudy of DREAMM-2 evaluated corticosteroid eye drops given in one eye only and showed no difference in ocular events between the two eyes. This suggests that corticosteroid eyedrops are not effective prophylaxis for preventing corneal complications [8]. Currently, it is recommended that patients use preservative-free lubricating eye drops at least 4 times daily for the duration of their treatment [11].

It is currently unclear how ocular toxicity and necessary dose interruption affect patient response and outcomes. A study of 38 consecutive heavily treated patients (median 8 prior lines) who received ≥ 1 dose of belamaf 2.5 mg/kg evaluated this question, with a median duration of belamaf treatment of 2 months and median 11 months of follow-up [28]. The ORR for the overall cohort was 20% and the median PFS was 2 months [28]. As expected, 75% of patients in this cohort experienced corneal toxicity, with 58% having decreased BCVA and 69% keratopathy, occurring at a median time of 1.4 months into therapy and improving after withholding belamaf for median 2.4 months [28]. In this study, belamaf was permanently discontinued or delayed due to keratopathy in 10 of the 38 patients (26%), with 7 of those patients experiencing MM progression within 3 months while off belamaf [28]. This study raises the issue of belamaf feasibility in the context of advanced MM in the real world, though larger studies are needed to evaluate this question.

Thrombocytopenia is commonly reported [6, 8] and is also consistent with previous reports associated with MMAF [25]. It can be significant, but it recovers between doses and bleeding events are rare [7, 8].

Selinexor

Selinexor is an oral selective inhibitor of nuclear export (SINE) small molecule that inhibits exportin 1 (XPO1) function, preventing the export of tumor suppressor proteins, oncoprotein mRNAs, and the glucocorticoid receptor [29, 30]. The nuclear retention of tumor suppressors, in addition to suppression of BCL-2 and MYC [31], leads to myeloma cell death [32].

The initial phase 1 study of selinexor in patients with hematologic malignancies, including heavily treated MM, explored a range of selinexor doses (45–160 mg) twice weekly [33]. The ORR was 23%, and based on tolerance and efficacy, the dose of 60 mg twice weekly was selected as the RP2D for hematologic malignancies [33]. Subsequently, part 1 of the STORM phase II trial of 79 MM patients (median of 7 prior lines, 48 and 31 of whom were quad- and penta-refractory, respectively), using selinexor 80 mg and dexamethasone 20 mg twice weekly was performed [34]. The ORR was 21% in quad-refractory patients and 20% in penta-refractory patients, with median PFS of 2.3 months and median OS of 9.3 months [34].

Part 2 of the STORM trial evaluated selinexor 80 mg and dexamethasone 20 mg twice weekly in a heavily treated patient population, with inclusion criteria requiring exposure to lenalidomide, pomalidomide, bortezomib, carfilzomib, glucocorticoids, an alkylator, and daratumumab. Patients also had to be triple refractory (refractory to a PI, IMID, and anti-CD38 mAb) and refractory to the last line of therapy [3]. One hundred and twenty-two patients were enrolled with a median of 7 prior lines of therapy, 68% penta-refractory and 96% refractory to carfilzomib, pomalidomide, and daratumumab [3]. The overall response rate was 26% with ≥ VGPR rate of 5%, median PFS 3.7 months, and OS 8.6 months [3].

With respect to combination trials, a phase 1 study of 21 patients, median 4 prior lines of therapy and 81% refractory to an IMID and PI, treated with selinexor in combination with carfilzomib and dexamethasone (SKd) established a recommended RP2D of selinexor 60 mg, carfilzomib 20/27 mg/m2, and dexamethasone 20 mg, all twice weekly [35]. ORR was 48% with ≥ VGPR rate of 14% for the entire cohort; for 13 patients treated at the RP2D, the ORR was 38% with ≥ VGPR rate 15% [35]. A phase I study of the all-oral combination of selinexor, ixazomib, and dexamethasone (SId) was performed in 18 patients with 5 prior lines of therapy, establishing an MTD of selinexor 80 mg weekly in combination with dexamethasone 20 mg weekly and ixazomib 4 mg on days 1, 8, and 15 of a 28-day cycle, with an ORR of 22% in the cohort [36].

The phase 1b/2 STOMP study is an ongoing trial evaluating the safety and efficacy of selinexor in 10 combinations and 11 treatment arms. The combination arm of selinexor, bortezomib, and dexamethasone (SVd) has been reported, which included 42 patients with median 3 prior lines of treatment, of whom 50% were PI refractory and 45% were PI + IMID refractory [4]. The RP2D of SVd was determined at selinexor 100 mg, bortezomib 1.3 mg/m2, and dexamethasone 40 mg all given once weekly in 35-day cycles, dosing at which no grade 3 or 4 nausea of vomiting was observed [4]. The ORR of the overall SVd cohort was 63% with 31% achieving a ≥ VGPR; ORR was 58% and 21% ≥ VGPR in those treated at the RP2D [4]. Median PFS was 9.0 months for the entire cohort, 6.1 months for the 21 patients who were PI refractory, and 17.8 months for the 19 patients who were not [4].

Preliminary results of the selinexor, pomalidomide, and dexamethasone (SPd) arm of STOMP were recently reported [37]. Twenty patients were treated with the selinexor 60 mg weekly and 19 patients with 40 mg weekly, both in combination with standard pomalidomide 4 mg on days 1–21 and weekly dexamethasone 40 mg. This group is less heavily pre-treated, with median 2 prior lines of therapy, 85% PI and 79% IMID refractory. In the 60 mg group, ORR was 65% with median PFS 8.9 months, whereas the ORR in the 40 mg group was 42% with median PFS not yet reached; no new safety concerns were identified [37].

The phase III Boston trial of 402 patients with relapsed MM compared SVd to Vd in patients treated with 1–3 prior lines of therapy [5]. At a median follow-up of 13.2 months for the SVd group, the median PFS was 13.9 months as compared to the Vd cohort who had a median of 16.5 months of follow-up and a median PFS of 9.46 months [5]. A summary of active and accruing trials for selinexor is summarized in Table 5.

Table 5 Active* clinical trials including selinexor (where recruitment is ongoing or anticipated)
Full size table

Toxicity

The main challenges in safety and tolerability in patients treated in clinical trials with selinexor are gastrointestinal in nature, with the most frequent non-hematologic AEs being nausea (62–73%), vomiting (31–44%), anorexia (49–60%), and diarrhea (42–43%), the majority of which were grades 1–2 [4, 34], with 5–8% of patients having grade 3–4 nausea and 5–7% having grade 3–4 diarrhea [4, 34]. Grade 3 hyponatremia occurred in 22% of patients [3, 34]. The most common grade 3–4 hematologic toxicity is thrombocytopenia (50–59%), followed by anemia (19–44%) and neutropenia (21–26%) [3, 4, 34], with grade 3 bleeding seen in only a minority of patients (3–5%) [3, 34].

Routine use of prophylactic and therapeutic anti-emetics is recommended. Administration of 8 mg of ondansetron (or other 5-HT3 agonists) prior to selinexor dosing, as well as up to three times daily as needed afterwards, is encouraged. We and others have found low-dose olanzapine (2.5 mg) two times daily as needed can also be effective for mitigating nausea for patients treated with selinexor [36]. Additional anti-emetic agents including low-dose benzodiazepines, megestrol, neurokinin-1 receptor antagonists, and cannabinoids have been described in patients treated with selinexor [38].

Thrombocytopenia is predictable and multifactorial, due to the drug in addition to additional factors such as prior myelosuppressive therapy and MM. Close monitoring of platelets is recommended and off-label use of eltrombopag or romiplostim can be considered for severe thrombocytopenia, in additional platelet transfusion if clinically indicated [38].

Melflufen

Melflufen is an alkylating agent and a melphalan prodrug that was developed to reduce drug resistance, increase tumor specificity, and reduce toxicity [39, 40]. It is rapidly incorporated into tumor cells due to its high lipophilicity and is hydrolyzed into melphalan by aminonopeptidase N, an enzyme highly expressed in MM cells, leading to an accumulation of melphalan inside MM cells and resultant cytotoxicity [39]. Pre-clinical studies demonstrated increased potency and the ability to overcome resistance to melphalan and novel agents, endorsing further clinical studies to improve outcomes in patients with RRMM [39, 40].

The initial phase 1/2 study in RRMM which included patients with ≥ 2 prior lines of treatment, PI and IMID exposed, established the maximum tolerated dose (MTD) of melflufen of 40 mg every 21 days with dexamethasone 40 mg weekly [41]. The phase 2 portion showed an ORR of 31% [41], with a median PFS of 5.7 months and overall survival (OS) of 20.7 months at a median follow-up of 46.0 months [42]. The phase 2 HORIZON study followed, which included patients with ≥ 2 prior therapies, also PI and IMID exposed as well as pomalidomide and/or daratumumab refractory [43]. Seventy-six percent of patients included in HORIZON were triple-class refractory, 59% alkylator refractory, and 35% had extramedullary disease [42]. With a median of 5.5 prior treatments, ORR was 29% with a median PFS of 4.2 months and OS of 11.6 months [1], including ORR of 26%, median PFS of 4.0 months, and OS of 11.3 months in patients with triple-class refractory disease [1].

The instructive phase 3 OCEAN study was a large trial of melflufen and dexamethasone versus pomalidomide and dexamethasone (Pd) in patients with RRMM after 2–4 prior lines of therapy. With the most recent update, 474 patients were randomized: 246 to receive melflufen and dexamethasone and 249 to Pd. Median prior therapies was 3 in both groups, with 50% having a prior transplant [2]. At a median follow-up of 15.5 months, the median PFS was 6.8 months in the melflufen group and 4.9 months in the Pd group, with median OS of 19.8 months and 25.0 months, respectively [2]. While exploratory analyses of PFS favored melflufen in most subgroups, Pd showed improved OS in patients who had previously had ASCT. Further analyses specifically evaluating patients with and without prior ASCT were performed. In the no prior ASCT cohort, the median PFS was 9.3 months in the melflufen group compared to 4.6 months in the Pd group, with a median OS of 21.6 months for melflufen and 16.5 months for Pd [2]. In contrast, patients with prior ASCT had a median PFS of 4.4 months for melflufen and 5.2 months for Pd, with median OS of 16.7 months and 31.0 months, respectively [2]. Post hoc multivariate analysis revealed prior ASCT and ECOG PS 1–2 vs. 0 as risk factors affecting OS [2]. The FDA requested suspension of enrollment based on this preliminary data, and subsequently, the sponsor withdrew melflufen from the US market [44].

With this development, two additional studies have since been terminated (the ANCHOR phase 1/2 study evaluating the combination of melflufen with bortezomib or daratumumab in RRMM and the phase 2 BRIDGE trial of patients with RRMM and renal impairment).

Toxicity

Melflufen is generally well tolerated, with the most common grade 3–4 AEs being hematologic and low frequencies of non-hematologic AEs [1, 41, 45, 46]. While rates of neutropenia (54–79%), thrombocytopenia (63–76%), and anemia (43%) were high, grade 3 (1–16%) and 4 (0–3%) bleeding events are low [2, 42]. Non-hematologic AEs are less common and include pneumonia (3–10%) and other infections (11%–10); gastrointestinal events including diarrhea (12–27%), constipation (7–15%), and nausea/vomiting (13%) are common and mostly grades 1–2 with rare grade 3 events [2, 42].

Conclusions

This review focused on the respective roles of belantamab mafodotin, selinexor, and melflufen as treatment for RRMM. The majority of patients with RRMM will be exposed and/or refractory to PIs, IMIDs, and anti-CD38 MABs, given the increasing use of triplet and quadruplet therapies earlier in the disease course, and this population continues to be one of an unmet clinical need. While there are many novel therapeutics being studied in this space, including bi-specific antibodies and chimeric antigen receptor T cell (CAR-T) therapies, optimal therapy and its sequencing is not yet defined. Belantamab mafodotin and selinexor both have single-agent activity in RRMM with a reasonable safety profile, although the duration of response as single agents remains limited and combination use is recommended for the majority of patients. Evaluation of their respective safeties and efficacy in combinations in both newly diagnosed and relapsed diseases is ongoing. Melflufen has been withdrawn from the US market and currently is not recommended for the treatment of MM. Ongoing studies are needed to define the optimal partners, timing, and sequence for the use of belamaf and selinexor as treatments for MM.