Selinexor: Targeting a novel pathway in multiple myeloma

Abstract Selinexor is an orally bioavailable selective inhibitor of nuclear export compound that inhibits exportin‐1 (XPO1), a novel therapeutic target that is overexpressed in multiple myeloma (MM) and is responsible for the transport of ∼220 nuclear proteins to the cytoplasm, including tumour suppressor proteins. Inhibition of this process has demonstrated substantial antimyeloma activity in preclinical studies, both alone and in combination with established MM therapeutics. Based on a clinical trial programme encompassing multiple combination regimens, selinexor‐based therapy has been approved for the treatment of relapsed/refractory MM (RRMM), with selinexor‐dexamethasone approved in the later‐relapse setting for penta‐refractory patients and selinexor‐bortezomib‐dexamethasone approved for patients who have received ≥1 prior therapy. Here, we provide a comprehensive review of the clinical data on selinexor‐based regimens, including recent updates from the 2022 American Society of Hematology annual meeting, and summarise ongoing studies of this novel targeted agent in newly diagnosed MM and RRMM.


INTRODUCTION
The past 20 years has seen rapid, extensive developments and innovations in the treatment of multiple myeloma (MM) [1,2]. The therapeutic armamentarium has expanded to include multiple new classes of agents in increasingly complex regimens, which have largely replaced old chemotherapy regimens. As a consequence, the 5-year relative survival rate has increased from 34.5% in the year 2000 to almost 60% at the present time [3], and a growing proportion of patients with newly diagnosed MM (NDMM) can expect to survive for more than 10 years, thanks to increasingly durable frontline responses [4], and/or therapies targeting B-cell maturation antigen (BCMA) have emerged recently as a fourth key drug class for the treatment of MM, following the approvals of two BCMA-directed chimeric antigen receptor (CAR) T-cell therapies, idecabtagene vicleucel (ide-cel) and ciltacabtagene autoleucel (cilta-cel) [11], the bispecific antibody teclistamab, which targets BCMA on MM cells and CD3 on T cells [8,12] and the anti-BCMA antibody-drug conjugate belantamab mafodotin [13] (which, following accelerated approval, has had US marketing authorisation withdrawn following the phase 3 DREAMM-3 trial not meeting its primary endpoint, but which remains under investigation in combination regimens in multiple studies including DREAMM-5, DREAMM-7 and DREAMM-8 based on promising efficacy in preliminary studies).
The range of treatment options is important in the context of MM being a highly heterogeneous disease with a variable and often unpredictable disease course [14]. MM is extremely complex at diagnosis and at relapse due to increasing numbers of genomic events and clonal evolution, which results in the disease displaying numerous mechanisms of resistance [1]. As a consequence, 'one size does not fit all' in the treatment of patients with NDMM or RRMM [15]. Furthermore, due to the multiply mutagenic nature of the disease, patients will typically receive at least several lines of therapy over the course of their disease, adding further complexity to treatment selection-thus, a range of distinct treatment options is needed, with optimized sequencing, and with therapies with activity in later-line settings in patients with RRMM [16].
In particular, given the widespread use of PIs, lenalidomide, pomalidomide and anti-CD38 mAbs in frontline approaches and/or earlier lines of treatment in the relapse setting, there is a specific need for efficacious treatment options in the triple-class refractory setting-in which patients have MM that is refractory to a PI, an immunomodulatory drug and an anti-CD38 mAb [17]-and the penta-refractory setting (disease that is refractory to two PIs, two immunomodulatory drugs and an anti-CD38 mAb) [18], which may arise as soon as after only two lines of therapy, in order to provide alternatives following failure of these standards of care.
Given these evolving needs, multiple novel agents with new targets are emerging to expand further the therapeutic armamentarium. With MM being a cancer sensitive to immune surveillance, multiple immunebased therapies-in addition to ide-cel, cilta-cel and teclistamab-are being developed that show substantial activity in RRMM [19]. These include novel antibody-drug conjugates [13] and bispecific antibodies/T cell engagers such as talquetamab [8,12] that target other antigens specific to MM cells, such as CD38, G protein-coupled receptor, class C, group 5, member D (GPRC5D) and Fc receptor-homolog 5 (FcRH5), and co-opt or stimulate the patient's immune system to fight against the disease. Furthermore, following on from the immunomodulatory drugs, a next-generation class of agents-cereblon E3 ligase modulators [20]-is also showing promising early results in the RRMM setting.
In addition to these agents, other small molecule targeted therapies have been developed to exploit specific characteristics of MM cells and pathways that are upregulated in cancer cells more broadly.
Having such options is valuable and important in the RRMM setting following prior mAb-based and immunomodulatory drug-based treatment and, increasingly, post CAR T-cell or bispecific antibody therapy.
Among agents that have been approved for the treatment of RRMM is the selective inhibitor of nuclear export (SINE) compound selinexor, which affects a novel target-exportin-1 (XPO1)-and signalling pathway in MM cells. Selinexor has demonstrated substantial activity in multiple different combination regimens in RRMM, including as postimmunotherapy treatment. Here, we review this novel targeted agent, its unique target and its preclinical and clinical activity and clinical safety in RRMM.

XPO1 INHIBITION-A NOVEL TARGET IN MULTIPLE MYELOMA
The role of XPO1, also known as Chromosome Region Maintenance 1, in normal cells and its potential as an anticancer therapeutic target has been extensively reviewed previously [24][25][26][27][28]. Briefly, XPO1 is responsible for the transport of approximately 220 nuclear proteins to the cytoplasm via nuclear pore complexes [25], including a number of crucial tumour suppressor proteins (TSPs) such as Rb, p53, p21 and p27 [28] (Figure 1). XPO1 thus plays a critical role in cell cycle regulation and cellular proliferation associated with its various cargo proteins. XPO1 overexpression has been shown to be a hallmark of a number of cancers, including multiple solid tumours [24,25]. Elevated expression of XPO1 enables cancer cells to escape TSPs by exporting them to the cytoplasm where they are unable to function, resulting in dysregulation of growth signalling and increased anti-apoptotic signalling.
It also elevates cytosolic levels of pro-survival proteins such as cellular inhibitor of apoptosis proteins, survivin and myeloid cell leukemia-1 (MCL1). Thus, overexpression of XPO1 is associated with poor prognosis and drug resistance in various tumours, including osteosarcoma, glioma, pancreatic cancer and ovarian carcinoma [24]. There is therefore a strong rationale for inhibiting XPO1 in cancer [24,29]. Inhibition of XPO1 thus impacts tumour cells via 3 core mechanisms: increasing nuclear levels and activation of TSPs; trapping oncoprotein mRNA in the nucleus, leading to reduced oncoprotein levels; and retaining activated glucocorticoid receptor (GR) in the nucleus ( Figure 1) [24,29]. XPO1 inhibition has also been shown to inhibit neutrophil extracellular trap formation, which has been associated with cancer progression [30]. Initial evaluations of XPO1 inhibition demonstrated antitumor activity in a range of tumour types, including lung cancer [25] and leukaemias [31], as well as in gastric cancer related to accumulation of p53 [32].
The rationale for targeting XPO1 in MM reflects that in other tumour types. XPO1 has been shown to be overexpressed in MM [29,33,34], and an RNA interference screening analysis found it to be among the most vulnerable potential therapeutic targets in the disease [34]. XPO1 plays a number of key roles of relevance in MM [29] F I G U R E 1 Schematic of the role of XPO1 in transporting various cargoes from the nucleus to the cytoplasm and the effects of XPO1 inhibition with selinexor [27,29]. CDK, cyclin-dependent kinase; cIAP, cellular inhibitor of apoptosis protein; NF-κB, nuclear factor-κB; GR, glucocorticoid receptor; IL, interleukin; MCL1, myeloid cell leukaemia sequence 1; PUMA, P53 up-regulated modulator of apoptosis; Rb, retinoblastoma; TSP, tumour suppressor protein; VEGF, vascular endothelial growth factor; XPO1, exportin-1.
( Figure 1); in addition to exporting TSPs and cell cycle regulators, XPO1 exports immune response regulators such as IκB, the inhibitor of the transcription factor nuclear factor-κB (NF-kB), leading to dysregulated cellular growth signalling and an anti-apoptotic state. Further, cargoes of XPO1 also include the oncoprotein mRNAs cellular myelocytomatosis (c-MYC), cyclin D1 and murine double minute 2 (MDM2), and their transport to the cytoplasm thus promotes the synthesis of these oncoproteins, which are known to be upregulated in MM [29].

PRECLINICAL ACTIVITY OF XPO1 INHIBITION IN MULTIPLE MYELOMA
Selinexor, an orally bioavailable SINE compound (Figure 1), is one of a number of potent small molecule inhibitors of XPO1 that bind covalently to the cargo-binding groove of XPO1 to prevent its nuclear transport role [29]. This mechanism of action was demonstrated using CRISPR/Cas9 genome editing to create a homozygous gene mutation in XPO1 that resulted in a mutation in the binding site of selinexor, thereby conferring resistance to selinexor [35]. As a small molecule inhibitor, selinexor is known to cross the blood-brain barrier, as evidenced by its demonstrated activity in glioblastoma [36]; this property also potentially mediates some of the common toxicities associated with selinexor, as discussed below. Of importance in the context of some of the critical signalling pathways in MM, preclinical data have shown that selinexor in MM models reactivates multiple TSPs relevant to MM, inhibits NF-kB signalling, reduces c-Myc levels, and reactivates GR signalling in combination with dexamethasone [24-26, 29, 33]. Selinexor has also been shown to preferentially disrupt the 3D nuclear organization of telomeres in cancer cells versus normal cells, resulting in antitumor activity [37], and to result in antitumor activity in hypoxia-induced bortezomib-resistant MM cells, resensitising the cells to bortezomib [38]. Conversely, a number of mechanisms of selinexor resistance have been suggested from preclinical and ex vivo studies, including the upregulation of alternative export pathways in selinexor-refractory MM [39,40].
Selinexor has demonstrated synergistic activity in combination with various MM agents in vitro and in vivo. Synergistic effects were seen in combination with dexamethasone via the induction of GR expression and inhibition of the mammalian target of rapamycin pathway in MM cells [41], as well as in ex vivo primary MM cells, with MYC-regulated genes associated with sensitivity to the combination [40]. Selinexor also results in synergistic antimyeloma activity in combination with PIs, potentially via the mechanism of nuclear localization of IκB and the resultant inhibition of NF-κB activity [42,43]; specific activity has been demonstrated in combination with both bortezomib [38,44] and carfilzomib [45], with the latter associated with mechanisms including a reduction in expression of the pro-survival protein B-cell lymphoma 2 (Bcl-2) and induction of caspase-10-dependent apoptosis. Other combinations with synergistic activity in preclinical studies include F I G U R E 2 US FDA and EMA approvals of selinexor in relapsed/refractory multiple myeloma. dex, dexamethasone; EMA, European Medicines Agency; FDA, Food and Drug Administration; US, United States; Vd, bortezomib-dexamethasone. selinexor plus pomalidomide [40], the conventional MM chemotherapy drugs melphalan [46] and doxorubicin [47], the latter mediated through inhibition of nuclear export of topoisomerase II alpha and the resultant doxorubicin-induced DNA damage, and the investigational pan-RAF (rapidly accelerated fibrosarcoma) inhibitor TAK-580 [48], with synergistic effects mediated by the FOXO3a (forkhead box O3)-Bim (Bcl-2 interacting mediator of cell death) signalling pathway. These preclinical findings have provided the mechanistic rationales for the subsequent clinical investigation of combination regimens, notably of selinexor in combination with dexamethasone, with or without a PI.

CLINICAL STUDIES OF SELINEXOR IN MULTIPLE MYELOMA
Based on the findings from clinical studies of selinexor-based therapy in RRMM, selinexor is currently approved in the US and EU in combination with bortezomib-dexamethasone (Vd) for patients who have received ≥1 prior therapy and in combination with dexamethasone in patients who have received ≥4 prior therapies and are refractory to ≥2 PIs, ≥2 immunomodulatory drugs, and an anti-CD38 mAb [49,50]. Figure 2 illustrates the timeline for these approvals [51]. It is also approved in the US for patients with diffuse large B-cell lymphoma following ≥1 prior therapies [49].
Consequently, selinexor-based therapy is currently incorporated into both European and US treatment guidelines for MM.
Within the European Society for Medical Oncology guidelines [10], selinexor-Vd is included as a treatment option for second-line therapy after daratumumab plus (1) lenalidomide-dexamethasone (Dara-Rd), (2) bortezomib, melphalan, and prednisone (Dara-VMP), or (3) bortezomib, thalidomide, and dexamethasone (Dara-VTd) and after lenalidomide-bortezomib-dexamethasone (RVd) in bortezomibsensitive and lenalidomide-sensitive or lenalidomide-refractory patients. Selinexor-Vd is also suggested as third-line therapy or beyond in PI-sensitive, lenalidomide-refractory patients and selinexor-dexamethasone is recommended as an option for tripleclass refractory patients. The US National Comprehensive Cancer Network guidelines [9] incorporate a wider range of suggested regimens based on data from multiple clinical trials; for patients with RRMM following 1-3 prior therapies, the guidelines include once-weekly selinexor-Vd, selinexor-daratumumab-dexamethasone and selinexor-carfilzomib-dexamethasone (Kd) as options, as well as selinexor-pomalidomide-dexamethasone (pom-dex) for patients who have received two prior therapies including a PI and an immunomodulatory drug and who are refractory to their last prior therapy.
Selinexor-dexamethasone is also included as an option, in line with the US label.
The clinical trials reviewed below form the basis for these approvals and recommendations for selinexor-based therapy.

Selinexor alone or in combination with dexamethasone for RRMM
In the earliest clinical studies in RRMM, selinexor was investigated alone and in combination with dexamethasone in the late-relapse setting, including in patients who were triple-class refractory and/or penta-class exposed or penta-refractory (Table 1) [52,53]. Selinexor ± dexamethasone was evaluated at 12 different dose levels and using four distinct dosing schedules (28-or 21-day cycles) in a phase 1 study in 84 patients with RRMM or Waldenström's macroglobulinemia who had received a median of six prior therapies [52]. Although no maximum tolerated dose was determined, twice-weekly selinexor 45 and 60 mg/m 2 in combination with dexamethasone in 28-day cycles were selected for study in expansion cohorts based on other clinical study data, with the lower dose level proving more tolerable and the addition of dexamethasone improving gastrointestinal tolerability compared with single-agent selinexor [52]. Activity was limited overall, associated with the use of multiple low dose levels and limited numbers of patients receiving added dexamethasone (Table 1); however, 6 of 12 patients treated with twice-weekly selinexor 45 mg/m 2 plus dexamethasone responded, with 1 achieving a complete response (CR).
Based on these findings, twice-weekly selinexor at a fixed dose of 80 mg (approximately equivalent to 45 mg/m 2 ) plus dexamethasone 20 mg was studied in the phase 2 STORM trial [53]. In part 1 of the study, patients (n = 51) were initially dosed on days 1, 3, 8, 10, 15 and 17 of 28-day cycles, but following a protocol amendment continuous twice-weekly dosing (i.e., on additional days 22 and 24; n = 28) was used with no apparent impact on safety profile or activity.
All patients were quad-refractory (refractory to bortezomib, carfilzomib, lenalidomide, pomalidomide) and 39% were penta-refractory (also refractory to CD38 mAb). The overall response rate (ORR) was 21% in all patients (Table 1), 20% in penta-refractory patients, 25% in patients with del17p, and 35% in those with any high-risk TA B L E 1 Response and outcomes with selinexor ± dexamethasone in RRMM.
cytogenetics (del17p, t(4;14), t(14;16)). Based on this promising activity, STORM was expanded to include a phase 2b component that enrolled 122 triple-class refractory, penta-exposed patients, 68% of whom were penta-refractory [54]. As in the latter part of the phase 2 component, patients received continuous twice-weekly dosing of selinexor 80 mg plus dexamethasone 20 mg. An ORR of 26% was achieved, which included two stringent CRs (sCR) and two partial responses in patients who had relapsed following CAR T-cell therapy; ORR was similar regardless of which agents patients were refractory to, and was 25% in the penta-refractory cohort. Outcomes were promising (Table 1), with a median overall survival (OS) of 15.6 months among patients achieving at least a minimal response in this highly refractory and heavily pre-treated patient population [54].
As STORM was a single-arm study without a comparator group, the OS achieved with selinexor-dexamethasone was instead compared with a similar cohort of real-world triple-class-refractory patients [55].

Selinexor-based triplet regimens for RRMM
with Vd-compared with Vd alone-in the phase 3 BOSTON trial [61] ( Table 2). The BOSTON trial was informed by initial experience with selinexor-Vd in the phase 1b/2 STOMP study [62], which has investigated several different selinexor-based triplet regimens (Table 3). with twice-weekly regimens [62]. STOMP also demonstrated the triplet to have substantial activity in a patient population that had received a median of 3 prior regimens; the ORR was 63%, including 30% ≥very good partial response (VGPR), among all patients and 58% at the dose recommended for BOSTON [62]. Notably, when considering use of the triplet earlier in the treatment algorithm, response rates were particularly high in patients who were not refractory to a PI, with an ORR of 84%, including 37% ≥VGPR, and PFS was also prolonged compared with the overall population (median 17.8 versus 9.0 months,  (Table 3). Similarly promising response rates, with ORRs of 48%-70% and ≥VGPR rates of 14%-27%, were reported from two phase 1 studies of selinexor-Kd (Table 4) [81,82] and three analyses of patients from STOMP (Table 3) [72,73] and other selinexor studies (Table 4) [86]; importantly, as well as being heavily pre-treated, the patients in these studies and analyses were refractory to multiple prior therapies and had been exposed to novel immunotherapy agents, including CAR T-cell therapy, anti-BCMA therapies more broadly. The feasibility and activity of selinexor-based regimens in the post-immune-based therapy setting, including after CD38 mAbs [72], BCMA-targeting agents (ORR 64%, 6-month PFS 75% with selinexor-based triplets/quadruplets in 11 heavily pretreated patients who had received prior anti-BCMA therapy [73]), or CAR-T cell therapy [81,86] (including responses in six of 7 patients, with durations of up to 7.4+ months, with selinexor plus dex, Vd or Kd in patients with a median of 10 prior regimens and who were refractory to prior CAR T-cell therapy [86]) is valuable. This evolving position in the MM treatment algorithm represents a potential unmet need due to the increasing use of bispecific antibodies, belantamab mafodotin and CAR T-cell therapies earlier in the treatment course, associated with their greater efficacy compared with other regimens in this setting [89,90]. Selinexor-based therapy in the later-relapse setting in regions where CAR T-cell therapy is not available or accessible, or in patients for whom is it not appropriate, may also be useful [91].
With the aim of personalizing therapy for patients with RRMM, several analyses have been conducted into potential markers of selinexor activity or resistance. These markers may help identify patients for whom selinexor-based treatment is a more/less promising option.
Using RNA sequencing of CD138+ cells from 100 patients in the BOSTON study, a three-gene signature comprising WNT10A, DUSP1 and ETV7 was identified that predicted response to selinexor-Vd [92], suggesting that elevated interferon-mediated signalling may sensitize MM cells to selinexor. The signature was validated not only in a cohort of patients from STOMP and in the real-world RRMM setting but also in patients with glioblastoma, suggesting it as a potential pan-tumour signature of selinexor sensitivity. In a separate analysis using hematologic malignancy cell lines and samples from patients with myelodysplastic syndrome, a number of potential protein biomarkers of selinexor activity were identified, including XPO1, NF-κB(p65), MCL-1 and p53 protein levels [93], a finding that is supportive of the synergistic mechanism between selinexor and PIs. Other genetic markers identified as possible mediators of selinexor sensitivity include ABCC4 (MRP4) [94] and ASB8 (ankyrin repeat and SOCS box containing 8), the latter potentially representing a shared modulator of activity across cancer types [95] associated with ASB8 promoting selinexor-induced proteasomal degradation of XPO1 [95]. Meanwhile, a potential biomarker predictive for resistance to selinexor is downregulation of TGFβ-SMAD4 pathway signalling, which has been associated with poorer outcome [95].

Safety and tolerability of selinexor-based therapy for RRMM
Based on results from clinical trials (Table 5) and with extensive experience of using selinexor-based therapy, clinicians have learned to manage and mitigate the significant toxicity associated with selinexor with aggressive supportive care and timely dose de-escalation (Table 6) [96,97]; notably, the use of the once-weekly regimen and lower doses of selinexor in triplet combinations results in a more manageable safety profile. Common toxicities include gastrointestinal and haematological toxicities and fatigue [68,98] (Table 5). The safety profile and rates of toxicities are distinct between selinexor regimens that utilise twice-weekly dosing, such as with selinexor-dexamethasone in the STORM study [53,54], and those employing a once-weekly dosing schedule, such as used with selinexor-Vd in BOSTON [61].
Lower rates of common gastrointestinal adverse events (AEs), haematological AEs, fatigue and other common AEs such as infections and hyponatremia are seen with once-weekly dosing, and initial toxicity management guidelines for twice-weekly selinexor-dexamethasone [96,98,99] have been more recently revised to reflect the safety profile of the less intensive weekly regimen (Table 6) [97]; in this context, it should also be recognised that early-phase studies such as STORM were conducted in a more heavily pre-treated population than BOSTON, which may also have contributed to the differing safety profiles. Nevertheless, as detailed in the management guidelines [96][97][98][99] and in the US and EU labels for selinexor [49,50], common toxicitiesnotably gastrointestinal toxicities-remain important considerations for treatment associated with both regimens, and these are generally manageable through optimal supportive care ( Table 6) and, where required, dose reductions. Indeed, in an analysis of the BOSTON trial, not only did dose reductions result in substantially lower rates of key AEs but also they enabled patients to remain on selinexor-Vd treatment for a prolonged period of time (compared to those who did not have dose reductions), and this translated into improved therapeutic outcomes with the regimen [100].
As shown in Table 5, thrombocytopenia was generally the most common grade ≥3 AE reported with selinexor-based therapy [98], with a lower rate seen with selinexor-Vd in BOSTON compared to selinexor-dexamethasone in STORM. Nevertheless, selinexor-Vd did result in an increased rate of grade ≥3 thrombocytopenia compared to Vd in BOSTON, associated with the different causative mechanisms of thrombocytopenia with selinexor and bortezomib [101,102]; while bortezomib results in transient, cyclical decreases in platelet counts due to inhibition of platelet budding from megakaryocytes, selinexorinduced thrombocytopenia arises from inhibition of thrombopoietin signalling early during megakaryopoiesis. Consequently, the kinetics of the AEs differ, with selinexor-induced thrombocytopenia occurring approximately 2-3 weeks after starting treatment and commonly recovering with a 1-2-week treatment interruption [97]. However, with both bortezomib-induced and selinexor-induced thrombocytopenia, associated bleeding events are rare [101,102].
Fatigue and gastrointestinal AEs are the most common nonhaematological AEs reported with selinexor-based treatment, although rates of grade ≥3 events are generally limited (Table 5).

US PI [49] Expert recommendations (for once-weekly selinexor) [97]
Patient selection • Selinexor-Vd: patients who have received ≥1 prior therapy • Selinexor-dex: patients who have received ≥4 prior therapies; disease refractory to ≥2 PIs, ≥2 immunomodulatory drugs, an anti-CD38 mAb • Weekly selinexor-based regimen with a PI or an immunomodulatory drug in patients progressing on an anti-CD38 mAb [97] • US: selinexor-Vd, selinexor-dara-dex, selinexor-Kd following 1-3 prior therapies; selinexor-pom-dex following 2 prior therapies including a PI and immunomodulatory drug (and refractory to last prior therapy) [9] Baseline evaluations / actions with selinexor-based therapy is possibly an effect mediated by the central nervous system (CNS) and resulting from selinexor crossing the blood-brain barrier [98]. Another common grade 3/4 toxicity with selinexor-based regimens is hyponatremia, which was reported in 22% of patients in part 2 of STORM; as with the other common AEs, this is manageable with supportive care (Table 6) [96,97,99]. Importantly, given its approval in combination with Vd, selinexor does not appear to be associated with an increased risk of peripheral sensory neuropathy, a key dose-limiting toxicity of bortezomib; indeed, in the BOSTON trial, the rate of any-grade peripheral neuropathy was lower with selinexor-Vd versus Vd (32% vs. 47%) [61], presumably due to the use showing a greater increase in the 'sensory' scale of the instrument [103]; in contrast, the 'autonomic' scale of the QLQ-CIPN20 showed a larger increase with selinexor-Vd. More broadly, analyses of qualityof-life data from STORM have suggested some limited decreases in domain scores associated with selinexor-dexamethasone treatment [104,105], potentially associated with the toxicity burden, albeit that the majority of patients did not demonstrate minimally important difference decreases.

Ongoing studies of selinexor in relapsed/refractory and newly diagnosed MM
Multiple studies of selinexor-based regimens are ongoing in both RRMM and NDMM ( Table 7). The randomized phase 3 EMN29 trial is comparing selinexor plus pomalidomide-dexamethasone with elotuzumab-pomalidomide-dexamethasone in patients with 1-4 prior lines, while the MUKtwelve trial is evaluating the addition of selinexor to cyclophosphamide-prednisolone in patients with ≥2 prior lines and exposure to a PI and an immunomodulatory drug [106]. Quadruplet regimens are also under investigation, including selinexor-pomalidomide-dexamethasone with or without carfilzomib in the SCOPE study and selinexor-Kd plus daratumumab, and combinations with other recently approved novel agents are also being studied, such as selinexor plus belantamab mafodotin and dexamethasone in an ongoing cohort of the STOMP study. Studies are also focused on various high-risk settings such as penta-refractory RRMM and RRMM with extramedullary disease. In the NDMM setting, selinexor is being evaluated in combination with lenalidomide, versus lenalidomide alone, as post-ASCT maintenance, while other studies are evaluating selinexor-Vd plus lenalidomide in high-risk NDMM and selinexor-pomalidomide-dexamethasone in patients with CNS involvement (SPODUMENE study) ( Table 7). The latter setting is of interest in the context of selinexor crossing the blood-brain barrier [107] and thereby potentially offering specific benefit in this setting, with optimised dosing.

CONCLUSIONS
In conclusion, XPO1 inhibition has been shown to be a rational tar- compounds exploit a novel mechanism of action and thus provide an alternative approach for targeting MM, which is valuable in the context of patients commonly requiring multiple lines of therapy over their treatment course. Selinexor-and potentially the next-generation SINE compounds, such as eltanexor, which has a substantially reduced ability to cross the blood-brain barrier that may translate into less toxicity such as fatigue, and thus potentially higher effective drug levels [29, 108]-will thus continue to play an important role in the treatment of patients with RRMM.

AUTHOR CONTRIBUTIONS
PGR developed scope of manuscript. All authors reviewed and revised the manuscript outline, first draft and final draft and approved the manuscript for submission.
Therapeutics. Research funding from Takeda, Janssen. Elizabeth K.