Exportin 1 (XPO1) is the major nuclear export protein for tumor suppressor proteins (TSPs; e.g., TP53, CDKN1A, FOXO1), cell cycle regulators (e.g., CDKN1A, CDKN1B, TOB1), and eIF4E-bound oncoprotein mRNAs (e.g., MYC, BCL2L1, MDM2, CCND1).¹ XPO1 is overexpressed in many malignancies, including multiple myeloma (MM), giving cancer cells a growth and survival advantage through enhanced nuclear export of TSPs and oncoprotein mRNAs. Therefore, selective inhibitors of nuclear export (SINE) compounds, which target XPO1, offer a novel approach to the treatment of MM.²

Eltanexor (KPT-8602) is an oral second-generation SINE compound currently in development for the treatment of hematological and solid tumor malignancies. Similar to other SINE compounds, eltanexor covalently binds to a critical cysteine residue (Cys528) in the cargo binding groove of XPO1 and inhibits XPO1-mediated nuclear transport.³ This leads to nuclear accumulation of multiple TSPs and reduction in oncoprotein translation, resulting in cell cycle arrest and apoptosis of cancer cells.³ In preclinical animal models, eltanexor demonstrated markedly reduced penetration across the blood–brain barrier (about 30-fold) and less weight loss, compared to the first-generation compound selinexor, allowing for more frequent dosing.³ Furthermore, eltanexor treatment (15 mg/kg given once daily for 5 days per week [QDx5]) caused 97% inhibition of tumor growth in preclinical mice models of MM (Karyopharm unpublished results). Finally, the strong synergy between eltanexor and dexamethasone antitumor activity observed in preclinical models of acute lymphoblastic leukemia,⁴ supports the addition of low dose of dexamethasone to eltanexor for the treatment of MM; similar strong synergy has been observed in both preclinical and clinical settings with the approved oral XPO1 inhibitor selinexor in MM.⁵ We therefore conducted a phase 1 study to evaluate the safety, tolerability, and efficacy of eltanexor with or without dexamethasone in patients with relapsed refractory MM (RRMM).

Thirty-nine patients were enrolled between January 2016 and August 2017. Baseline characteristics at screening are listed in Table S1. Patients had received a median of seven lines of prior therapy (range 2–14) and 49% (n = 19) had penta-exposed, triple-class refractory myeloma. The median duration of study treatment exposure was 2.7 months (range 0–45.9 months). Twenty-two (56%) patients had discontinued due to progressive disease (PD), nine for an adverse event (AE), three for physician's decision, and five for patient's decision.

Eltanexor was tested at six dose levels (5, 10, 20, 30, and 40 mg all given QDx5, and 60 mg [days 1, 3, 5 of each week] during 28-day cycles) utilizing a 3 + 3 study design (Table S2). Of the first three patients enrolled into the 40 mg dose level, one experienced a dose-limiting toxicity (DLT; > 4 missed doses in the first cycle as a result of nausea, vomiting, and anorexia—all grade 2 [G2]) and one experienced drug delay due to a treatment-related G3 anemia that was followed by PD and end of treatment. Three additional patients were enrolled at this dose level with none experiencing a DLT, but one withdrew consent after 22 days on study due to G2 fatigue and G1 thrombocytopenia. A cohort of patients treated at 60 mg was then enrolled to test a less frequent dose schedule and reduced cumulative dose of eltanexor. Of the four patients in this cohort, one was removed from the study after only one dose of eltanexor due to syncope requiring hospitalization, and each of the remaining three required a dose reduction. Though maximum tolerated dose (MTD) was not reached, based on these safety data the treating physicians and study sponsor agreed that dose escalation would be halted, and patients would be enrolled into two expansion cohorts to test 20 mg (n = 5) and 30 mg (n = 10) eltanexor plus dexamethasone from cycle 1 day 1 (C1D1). One DLT was observed in the 30 mg expansion cohort for G4 thrombocytopenia. In addition, although the small number of patients in these cohorts precluded a formal statistical comparison, patients in the 20 mg expansion cohort appeared to have fewer toxicities than those in the 30 mg expansion cohort (e.g. G4 thrombocytopenia 0% versus 50%, G3 neutropenia 20% versus 30%, and G3 anemia 20% versus 30%). Eltanexor demonstrated pharmacokinetics, pharmacodynamics, and molecular markers of response profiles that were like those of selinexor (Table S5, Figures S3 and S4).

Treatment-related adverse events (TRAEs) occurring in ≥ 10% of all patients are listed in Table S3. The most common TRAEs of any grade were thrombocytopenia (82%), nausea (54%), and neutropenia (51%). The most common G3 or G4 TRAEs were thrombocytopenia (54%), neutropenia (33%), and anemia (18%). Dose reductions occurred in 38% of patients, and dose interruptions in 72%. In addition to dose reductions and interruptions, thrombocytopenia was managed with platelet transfusions (28%) and the administration of thrombopoietin (TPO) receptor agonists (21%), romiplostim, or eltrombopag. The effectiveness of dose interruptions / TPO administration for alleviating thrombocytopenia is consistent with selinexor studies demonstrating that XPO1 inhibition prevents maturation of the megakaryocyte and not direct cytotoxicity.⁶ Nausea was managed, on an as needed basis, with prophylactic anti-emetic agent in 74% of patients, typically ondansetron or prochlorperazine, and 33% required an additional anti-nausea medication. There were four serious AEs deemed by the treating physician to be at least possibly related to eltanexor treatment: syncope (60 mg; resolved), acute kidney injury (30 mg + 20 mg dexamethasone, resolved with sequelae), accidental overdose of study drug (5 mg, resolved), and lung infection (20 mg + 20 mg dexamethasone, resolved). Seven deaths occurred while on study (treatment period plus an additional 30 days of safety follow-up): five due to PD, one due to a thromboembolic event occurring in a patient with a medical history of hypertension and deep vein thrombosis, and one due to a subdural hemorrhage (in the context of G4 thrombocytopenia that was not related to eltanexor) during hospitalization.

The AEs associated with eltanexor are consistent with those for the first-generation SINE compound, selinexor; however, the incidence and severity of nausea, decreased appetite, hyponatremia, and fatigue were lower than what was reported for selinexor in MM.⁵ Furthermore, the incidence of neurological AEs (including confusion, syncope, altered mental status, dizziness, depressed level of consciousness, and delirium) in this study was low, with 3 of 39 (8%) patients having a treatment-related neurological AE while on study, and only 1 of the 3 patients having a grade 3 neurological event (syncope). Tolerability of eltanexor was also evident, as patient withdrawal due to TRAEs was relatively low at 8%, while 36% of the patients remained on treatment for over 6 months.

The overall response rate (ORR) for the 35 efficacy evaluable patients was 20% (N = 7) and the clinical benefit rate (CBR) was 41%. This included one (2.9%) very good partial response (VGPR), six PRs, and nine minimal responses (MRs). Twelve patients had a best overall response of stable disease (SD) and seven of PD (Table S4). Nine of the 20 (45%) patients treated in the dose escalation phase showed a reduction in M-protein during the first cycle of treatment (Figure S1A). Dexamethasone (20 mg days 1, 3) was added to the treatment regimen of 15 of these patients. Best overall response for the 20 patients in the dose escalation phase included 2 PRs, 5 MRs, 8 SD, and 5 PD.

In the dose expansion cohort (n = 15), 20 mg and 30 mg eltanexor (QDx5) plus 20 mg dexamethasone (days 1, 3) starting at C1D1 produced an ORR of 33% (one VGPR and four PRs) and CBR of 60% (four MR). Four patients had SD and two had PD. Deeper and faster responses were observed when dexamethasone was started from C1D1 compared to after C2D1 (Figure S1A,B). While the ORR was higher for patients in the dose expansion cohort treated at 30 mg compared to 20 mg eltanexor (40% vs 20% respectively), the CBR was equivalent (60% for both groups).

Fourteen (40%) patients were treated for over 6 months, with four patients receiving treatment for over 1 year, one of whom has been receiving treatment for over 2 years and another for 3.8 years (Figure 1). The median progression free survival (PFS) and overall survival for all patients was 4.47 (95% CI 2.40, 9.26) months and 17.81 (95% CI 11.70, NR) months, respectively (Figure S2), an improved efficacy compared to selinexor QDx2: median PFS 3.7 months, and median OS 8.6 months.⁵ Such apparent improved efficacy of eltanexor QDx5 compared to selinexor QDx2 was also observed in preclinical models of MM (Karyopharm unpublished results).

The median PFS was higher for patients in the dose expansion cohort treated at 20 mg compared to 30 mg eltanexor (8.64 months [95% CI 8.48, 13.83] vs 3.02 months [95% CI 2.99, 9.26], respectively). Based on the totality of safety and efficacy data, a recommended phase 2 dose (RP2D) of 20 mg eltanexor (QDx5) plus 20 mg dexamethasone (days 1, 3) weekly was identified.

In conclusion, when given in combination with dexamethasone from C1D1, eltanexor demonstrated significant efficacy with 33% ORR, 60% CBR, and 80% of patients having a reduction in their myeloma markers from baseline. Furthermore, eltanexor was generally well-tolerated. Across all doses and schedules, these heavily pretreated patients (median 7 prior lines) had a median PFS of 4.47 months, suggesting both clinical benefit and tolerability. While an MTD was not reached in this study, based on the totality of the safety and efficacy data, an RP2D of 20 mg eltanexor (QDx5) plus 20 mg dexamethasone (days 1, 3) weekly was identified. These results support the novel approach of inhibiting XPO1 for the treatment of MM.

输出蛋白 1 (XPO1) 是肿瘤抑制蛋白（TSP；例如，TP53、CDKN1A、FOXO1）、细胞周期调节因子（例如，CDKN1A、CDKN1B、TOB1）和 eIF4E 结合的癌蛋白 mRNA（例如，MYC ）的主要核输出蛋白, BCL2L1 , MDM2 , CCND1 ). ¹ XPO1 在包括多发性骨髓瘤 (MM) 在内的许多恶性肿瘤中过度表达，通过增强 TSP 和癌蛋白 mRNA 的核输出为癌细胞提供生长和存活优势。因此，靶向 XPO1 的选择性核输出抑制剂 (SINE) 化合物为治疗 MM 提供了一种新方法。^2个

Eltanexor (KPT-8602) 是一种口服的第二代 SINE 化合物，目前正在开发中，用于治疗血液学和实体瘤恶性肿瘤。与其他 SINE 化合物类似，eltanexor 与 XPO1 货物结合沟中的关键半胱氨酸残基 (Cys528) 共价结合，并抑制 XPO1 介导的核转运。³这会导致多种 TSP 的核积累和癌蛋白翻译的减少，从而导致癌细胞的细胞周期停滞和凋亡。³在临床前动物模型中，与第一代化合物 selinexor 相比，eltanexor 的血脑屏障渗透率显着降低（约 30 倍），体重减轻也更少，因此可以更频繁地给药。^3个此外，eltanexor 治疗（15 mg/kg，每天一次，每周 5 天 [QDx5]）在 MM 的临床前小鼠模型中导致 97% 的肿瘤生长抑制（Karyopharm 未发表的结果）。最后，在急性淋巴细胞白血病的临床前模型中观察到 eltanexor 和地塞米松抗肿瘤活性之间的强协同作用，⁴支持在 eltanexor 中加入低剂量地塞米松用于治疗 MM；在临床前和临床环境中，已在 MM 中观察到批准的口服 XPO1 抑制剂 selinexor 具有类似的强协同作用。⁵因此，我们开展了一项 1 期研究，以评估 eltanexor 联合或不联合地塞米松治疗复发难治性 MM (RRMM) 患者的安全性、耐受性和疗效。

2016 年 1 月至 2017 年 8 月期间招募了 39 名患者。筛选时的基线特征列于表 S1 中。患者接受的先前治疗的中位数为 7 线（范围 2-14），49%（n  = 19）患有五线暴露、三级难治性骨髓瘤。研究治疗暴露的中位持续时间为 2.7 个月（范围 0-45.9 个月）。22 名 (56%) 患者因疾病进展 (PD) 停药，9 名患者因不良事件 (AE) 停药，3 名患者因医生决定停药，5 名患者因患者决定停药。

Eltanexor 使用 3 + 3 研究设计在六个剂量水平（5、10、20、30 和 40 mg 均给予 QDx5，以及 60 mg [每周第 1、3、5 天]，28 天周期）进行测试（表 S2）。在进入 40 mg 剂量水平的前三名患者中，一名患者经历了剂量限制性毒性（DLT；由于恶心、呕吐和厌食，第一个周期中错过了 4 次以上的剂量——均为 2 级 [G2]）和一名患者因治疗相关的 G3 期贫血而出现药物延迟，随后出现 PD 和治疗结束。在该剂量水平下招募了另外三名患者，没有人经历 DLT，但由于 G2 疲劳和 G1 血小板减少症，一名患者在研究 22 天后撤回同意。然后招募一组接受 60 mg 治疗的患者，以测试频率较低的剂量方案和减少的 eltanexor 累积剂量。在该队列中的四名患者中，一名患者因晕厥需要住院治疗而仅服用一剂 eltanexor 后退出研究，其余三名患者均需要减少剂量。尽管未达到最大耐受剂量 (MTD)，但根据这些安全数据，治疗医师和研究发起人同意停止剂量递增，并且患者将被纳入两个扩展队列以测试 20 mg（n  = 5) 和 30 mg ( n  = 10) 来自第 1 周期第 1 天的 eltanexor 加地塞米松 (C1D1)。在 G4 血小板减少症的 30 mg 扩展队列中观察到一个 DLT。此外，虽然这些队列中的少数患者排除了正式的统计比较，但 20 mg 扩展队列中的患者似乎比 30 mg 扩展队列中的患者具有更少的毒性（例如 G4 血小板减少症 0% 对 50%，G3 中性粒细胞减少症20% 对 30%，G3 贫血 20% 对 30%）。Eltanexor 的药代动力学、药效学和反应谱的分子标志物与 selinexor 相似（表 S5，图 S3 和 S4）。

发生在 ≥ 10% 的所有患者中的治疗相关不良事件 (TRAE) 列于表 S3。最常见的任何级别的 TRAE 是血小板减少症 (82%)、恶心 (54%) 和中性粒细胞减少症 (51%)。最常见的 G3 或 G4 TRAE 是血小板减少症 (54%)、中性粒细胞减少症 (33%) 和贫血 (18%)。38% 的患者出现剂量减少，72% 的患者出现剂量中断。除了减少剂量和中断治疗外，还通过输注血小板 (28%) 和使用血小板生成素 (TPO) 受体激动剂 (21%)、romiplostim 或艾曲波帕来控制血小板减少症。剂量中断/TPO 给药缓解血小板减少症的有效性与 selinexor 研究一致，表明 XPO1 抑制可防止巨核细胞成熟而不是直接细胞毒性。^6个根据需要，74% 的患者使用预防性止吐剂（通常是昂丹司琼或丙氯拉嗪）控制恶心，33% 的患者需要额外的抗恶心药物。治疗医师认为有四种严重的 AE 至少可能与 eltanexor 治疗相关：晕厥（60 mg；已解决）、急性肾损伤（30 mg + 20 mg 地塞米松，已解决并有后遗症）、研究药物意外过量（ 5 mg，解决）和肺部感染（20 mg + 20 mg 地塞米松，解决）。研究期间发生 7 例死亡（治疗期加上额外的 30 天安全随访）：5 例死于 PD，1 例死于血栓栓塞事件，该事件发生在有高血压和深静脉血栓形成病史的患者身上，

与 eltanexor 相关的 AE 与第一代 SINE 化合物 selinexor 的相关；然而，恶心、食欲下降、低钠血症和疲劳的发生率和严重程度低于 MM 中报告的 selinexor。⁵此外，本研究中神经系统 AE（包括意识模糊、晕厥、精神状态改变、头晕、意识水平低下和谵妄）的发生率很低，39 名患者中有 3 名 (8%) 患有与治疗相关的神经系统 AE在研究期间，3 名患者中只有 1 名出现 3 级神经系统事件（晕厥）。eltanexor 的耐受性也很明显，因为 TRAE 导致的患者退出率相对较低，为 8%，而 36% 的患者继续接受治疗超过 6 个月。

35 例疗效可评估患者的总缓解率 (ORR) 为 20% ( N  = 7)，临床获益率 (CBR) 为 41%。这包括一个 (2.9%) 非常好的部分反应 (VGPR)、六个 PR 和九个最小反应 (MR)。12 名患者的最佳总体反应为稳定疾病 (SD)，7 名患者为 PD（表 S4）。在剂量递增阶段接受治疗的 20 名患者中有 9 名 (45%) 在第一个治疗周期期间表现出 M 蛋白减少（图 S1A）。其中 15 名患者的治疗方案中加入了地塞米松（第 1、3 天 20 mg）。在剂量递增阶段，20 名患者的最佳总体反应包括 2 名 PR、5 名 MR、8 名 SD 和 5 名 PD。

在剂量扩展队列（n  = 15）中，从 C1D1 开始使用 20 mg 和 30 mg eltanexor (QDx5) 加 20 mg 地塞米松（第 1、3 天）产生了 33% 的 ORR（一个 VGPR 和四个 PR）和 60 的 CBR %（四个 MR）。四名患者有 SD，两名有 PD。与 C2D1 之后相比，从 C1D1 开始使用地塞米松时观察到更深更快的反应（图 S1A、B）。虽然接受 30 mg eltanexor 治疗的剂量扩展队列患者的 ORR 高于接受 20 mg eltanexor 的患者（分别为 40% 和 20%），但 CBR 相当（两组均为 60%）。

14 名 (40%) 患者接受治疗超过 6 个月，其中 4 名患者接受治疗超过 1 年，其中一名患者已接受治疗超过 2 年，另一名患者已接受治疗 3.8 年（图 1）。所有患者的中位无进展生存期 (PFS) 和总生存期分别为 4.47 (95% CI 2.40, 9.26) 个月和 17.81 (95% CI 11.70, NR) 个月（图 S2），与 selinexor QDx2 相比疗效有所提高：中位 PFS 3.7 个月，中位 OS 8.6 个月。⁵在 MM 的临床前模型中也观察到了 eltanexor QDx5 与 selinexor QDx2 相比明显改善的疗效（Karyopharm 未发表的结果）。

与 30 mg eltanexor 相比，剂量扩展队列中接受 20 mg eltanexor 治疗的患者的中位 PFS 更高（分别为 8.64 个月 [95% CI 8.48, 13.83] 和 3.02 个月 [95% CI 2.99, 9.26]）。根据安全性和有效性的整体数据，确定了每周 20 mg eltanexor (QDx5) 加 20 mg 地塞米松（第 1、3 天）的推荐 2 期剂量 (RP2D)。

总之，当与来自 C1D1 的地塞米松联合给药时，eltanexor 显示出显着的疗效，ORR 为 33%，CBR 为 60%，80% 的患者骨髓瘤标志物较基线有所降低。此外，eltanexor 通常具有良好的耐受性。在所有剂量和方案中，这些经过大量预处理的患者（中位 7 个先前线）的中位 PFS 为 4.47 个月，表明临床获益和耐受性。虽然本研究未达到 MTD，但根据安全性和有效性数据的整体性，确定了每周 20 mg eltanexor (QDx5) 加 20 mg 地塞米松（第 1、3 天）的 RP2D。这些结果支持抑制 XPO1 治疗 MM 的新方法。