Recent reports indicate that chimeric antigen receptor T cell (CAR‐T) therapy can induce deep and durable responses for many patients with heavily‐pretreated multiple myeloma (MM), with median progression‐free or event‐free survival ranging between 7 and 12 months (Zhao et al., 2018; Cho et al., 2018; Raje et al., 2019; Xu et al., 2019). Relapse post CAR‐T, however, is common, and treatment options for these patients are limited. Currently, there is minimal published data and no consensus on subsequent therapies after progression on CAR‐T. Most strategies, including CAR‐T retreatment, have proven largely ineffective, establishing a major unmet medical need for this patient population.

Selinexor is an oral, small‐molecule inhibitor of the nuclear export protein exportin 1 (XPO1) that induces accumulation of tumour suppressor proteins in the nucleus, reductions in oncoproteins, cell cycle arrest and apoptosis of cancer cells (Abdul Razak et al., 2016; Chen et al., 2017; Vogl et al., 2018). XPO1 is overexpressed in many cancers, including MM, and elevated levels are correlated with poor prognosis, increased bone lytic lesions and resistance to therapy (Tai et al., 2014). In patients with penta‐exposed, triple‐class refractory MM, selinexor plus low‐dose dexamethasone (Sd) produced an overall response rate of 26·2% and clinical benefit rate (≥ minimal response) of 39·3%, which translated into a survival benefit for responding patients (Chari et al., 2019). The activity of selinexor was preserved regardless of prior therapy, as expected from a drug with a novel mechanism of action; however, the efficacy after CAR‐T therapy has not been specifically described. Here, we report on observations of the activity of Sd alone or administered as a triplet in combination with bortezomib (SVd) or carfilzomib (SKd) in patients with MM whose disease has progressed after CAR‐T therapy.

We identified seven patients across selinexor trials who received lymphodepleting conditioning with fludarabine and/or cyclophosphamide followed by an effective dose of CAR‐T cell therapy [>10⁸ CAR‐positive cells targeting B‐cell maturation antigen (BCMA)] for their MM prior to being enrolled in a trial using a selinexor‐containing regimen. One patient was treated on the STORM study (NCT02336815) with selinexor (80 mg twice‐weekly, days 1 and 3) plus dexamethasone (20 mg twice‐weekly, days 1 and 3); one patient was treated with selinexor (100 mg once‐weekly) plus bortezomib (1·3 mg/m² once‐weekly for 4 of 5 weeks) and dexamethasone (40 mg once‐weekly) in the compassionate use program; and five patients were treated with selinexor (100 mg once‐weekly) plus carfilzomib (20/56 or 20/70 mg/m²) and dexamethasone (40 mg once‐weekly or 20 mg twice‐weekly) in the NCT02199665 trial. Response was assessed by the treating physician per International Myeloma Working Group (IMWG) criteria.

Baseline characteristics for the seven patients are summarized in Table 1. All patients were heavily pretreated (median 10 prior therapeutic regimens; range: 5–15) and had high‐risk cytogenetics; six patients had rapidly progressing disease as evidenced by the percent increase in paraprotein from screening to cycle 1, day 1 (C1D1) (range: 17–91%). Four patients had MM that was refractory to bortezomib, carfilzomib, lenalidomide, pomalidomide and daratumumab in a prior regimen (penta‐refractory), and all patients had progressed after autologous stem cell transplantation, alkylating agents, as well as CAR‐T therapy. The median time to progression on CAR‐T therapy was 4 months (1–7 months). All patients received a selinexor combination immediately after progression on CAR‐T (with the exception of one patient who received two additional lines of therapy prior to receiving Sd). As of the date of data cutoff, the median time on a selinexor‐based regimen was 6·0 months (range: 3·7–8·3 months); four patients’ disease had progressed, one patient had withdrawn consent, and two patients were still responding and on therapy.

Response assessments for each patient are outlined in Fig. 1A,B, and included one stringent complete response, three very good partial responses, two partial responses and one minimal response. Responses (≥ partial response) occurred within the first cycle of treatment for four patients; a noteworthy observation, given the rapidly progressing paraprotein levels at baseline. Interestingly, despite meeting IMWG criteria for having disease progression (>25% increase from nadir) after 6·5 months, Patient 1 continued treatment with Sd for an additional 1·5 months and derived clinical benefit before ending study with paraprotein levels nearly 50% below C1D1.

The anti‐myeloma activity seen in patients with disease that was refractory to bortezomib, carfilzomib and/or dexamethasone in a prior line of therapy validates reported data showing that selinexor overcomes resistance to proteasome inhibitors (PIs). Preclinical studies have demonstrated that selinexor synergizes with PIs and glucocorticoids through enhanced suppression of the NF‐κB signaling pathway and potentiation of glucocorticoid receptor transcriptional activity in the presence of dexamethasone, respectively (Kashyap et al., 2016; Argueta et al., 2018). These results are supported with clinical findings from the STOMP (NCT02343042) and NCT02199665 trials, which have demonstrated efficacy of Sd in combination with bortezomib or carfilzomib in patients with MM refractory to PIs (Bahlis et al., 2018; Jakubowiak et al., 2019).

Consistent with other studies testing Sd, SVd or SKd in heavily pretreated patients with MM, the most common adverse events among the patients described here were nausea, fatigue, thrombocytopenia, neutropenia and anaemia (Chen et al., 2017; Bahlis et al., 2018; Vogl et al., 2018; Jakubowiak et al., 2019). Most patients required a dose interruption or reduction in selinexor during the course of treatment. The seven patients presented here were managed for nausea with prophylactic 5‐HT3 antagonists, and additional supportive care agents including rolapitant, omeprazole, lorazepam or olanzapine. In addition, thrombocytopenia was managed and reversible with dose reductions or interruptions combined with weekly doses of romiplostim, a thrombopoietin agonist (typically starting at 1 ug/kg and increasing, if needed, to 10 ug/kg), to stimulate platelet production.

This is the first data set demonstrating anti‐myeloma activity of selinexor‐based regimens in patients who have progressed after CAR‐T therapy. The activity was observed regardless of prior treatment history, with no cross resistance. Though the findings reported here are based on a small group of patients, the responses are intriguing and warrant further investigation. Currently, the available therapeutic options for patients described here include intensive multi‐agent chemotherapy, recycling various combinations or entry onto an investigational trial. As CAR‐T therapy moves into earlier lines of MM treatment and more patients develop resitance to this approach, selinexor‐based regimens may offer important therapeutic benefit, underpinned by a novel mechanism of action which is key in the setting of relapsed and refractory disease, and should be considered (Richardson & Blade, 2014).

最近的报告表明，嵌合抗原受体T细胞（CAR-T）治疗可对许多高度预处理的多发性骨髓瘤（MM）患者产生深远和持久的反应，中位无进展或无事件生存期为7到12个月（Zhao等人，2018 ; Cho等人，2018 ; Raje等人，2019 ; Xu等人，2019）。但是，CAR-T后复发很常见，这些患者的治疗选择有限。目前，CAR-T进展后发表的数据很少，对后续治疗尚无共识。事实证明，大多数策略，包括CAR-T复治，在很大程度上均无效，这为该患者群体带来了严重的医疗需求缺口。

Selinexor是一种核出口蛋白输出蛋白1（XPO1）的口服小分子抑制剂，可诱导肿瘤抑制蛋白在细胞核中积聚，癌蛋白减少，细胞周期阻滞和癌细胞凋亡（Abdul Razak等人，2016年） ; Chen等，2017 ; Vogl等，2018）。XPO1在包括MM在内的许多癌症中均过表达，且水平升高与预后不良，溶骨性病变增加和对治疗的抗性有关（Tai等人，2014）。五线暴露，三级难治性MM患者，selinexor联合小剂量地塞米松（Sd）产生的总缓解率为26·2％，临床受益率（≥最小缓解）为39·3％，这转化为对有反应的患者有生存益处（Chari et al。，2019）。正如具有新作用机制的药物所预期的那样，无论以前的治疗如何，都可以保持selinexor的活性。但是，CAR-T治疗后的疗效尚未具体描述。在此，我们报告了在CAR-T治疗后病情已恶化的MM患者中，单独服用Sd或与硼替佐米（SVd）或卡非佐米（SKd）联用三联体给药的活性观察结果。

我们在selinexor试验中确定了七名患者，他们接受了氟达拉滨和/或环磷酰胺的淋巴清除调节，然后接受有效剂量的CAR-T细胞疗法[> 10 ⁸靶向B细胞成熟抗原（BCMA）的CAR阳性细胞]参加使用含selinexor方案的试验。在STORM研究（NCT02336815）中，一名患者接受了selinexor（80 mg，每周两次，第1和3天）和地塞米松（20 mg，每周两次，第1和3天）的治疗；一名患者接受了赛琳索（每周一次100 mg）加硼替佐米（1·3 mg / m ^2）治疗在同情使用计划中，每周一次，共5周中的4周）和地塞米松（每周一次40 mg）；在NCT02199665试验中，有5名患者接受了selinexor（每周一次100 mg）加carfilzomib（20/56或20/70 mg / m ²）和地塞米松（每周40 mg或每周两次20 mg）治疗。根据国际骨髓瘤工作组（IMWG）标准由主治医师评估反应。

表1总结了这7例患者的基线特征。所有患者均接受了严格的预处理（中位数10种先前的治疗方案；范围：5-15），并且具有高风险的细胞遗传学特征。从筛查到第1轮第1天（C1D1），副蛋白的百分比增加证明了6名患者的疾病进展迅速（范围：17–91％）。四例患者在先前的治疗方案中对硼替佐米，卡非佐米，来那度胺，泊马利度胺和达拉妥单抗难治性MM（戊型难治性），并且所有患者均在自体干细胞移植，烷化剂和CAR-T治疗后进展。CAR-T治疗进展的中位时间为4个月（1–7个月）。所有患者在CAR‐T进展后立即接受了selinexor联合治疗（一名患者在接受Sd之前接受了另外两种疗法的治疗除外）。截至数据截稿之日，基于selinexor方案的中位时间为6·0个月（范围：3·7-8·3个月）。4名患者的病情已恶化，1名患者撤回了同意书，2名患者仍在应答并接受治疗。

图1A，B概述了每个患者的反应评估，包括一个严格的完全反应，三个非常好的部分反应，两个部分反应和一个最小反应。在治疗的第一个周期内有四名患者出现反应（≥部分反应）；鉴于基线的副蛋白水平迅速增长，值得一提。有趣的是，尽管在6·5个月后达到IMWG疾病进展标准（从最低点增加25％以上），但患者1继续用Sd治疗了1·5个月，并在结束研究前获得了临床益处，副蛋白水平接近50％低于C1D1。

在先前治疗中对硼替佐米，卡非佐米和/或地塞米松难治的疾病患者中观察到的抗骨髓瘤活性验证了报道的数据，表明赛力克索克服了对蛋白酶体抑制剂（PIs）的耐药性。临床前研究表明，在存在地塞米松的情况下，selinexor分别通过增强对NF-κB信号通路的抑制和增强糖皮质激素受体的转录活性而与PI和糖皮质激素协同作用（Kashyap等人，2016 ; Argueta等人，2018））。这些结果得到STOMP（NCT02343042）和NCT02199665试验的临床结果的支持，这些临床结果证明了Sd与硼替佐米或卡非佐米联合治疗对PI难治性MM的疗效（Bahlis等，2018 ; Jakubowiak等，2019））。

与在经过大量预处理的MM患者中测试Sd，SVd或SKd的其他研究一致，此处描述的患者中最常见的不良事件是恶心，疲劳，血小板减少，中性粒细胞减少和贫血（Chen等，2017 ; Bahlis等，2018年; Vogl等人，2018年; Jakubowiak等人，2019年）。在治疗过程中，大多数患者需要中断剂量或降低赛瑞昔的剂量。此处介绍的7名患者使用预防性5-HT3拮抗剂和其他支持治疗剂治疗恶心，包括罗拉吡坦，奥美拉唑，劳拉西m或奥氮平。另外，通过减少剂量或中断剂量和每周一次剂量的血小板生成素激动剂romiplostim（通常从1 ug / kg开始，并在需要时增加至10 ug / kg）相结合，可以控制血小板减少症并使其可逆，以刺激血小板生成。

这是第一个数据证明在CAR-T治疗后进展的患者中，使用基于selinexor方案的抗骨髓瘤活性。无论先前的治疗史如何，均观察到活性，没有交叉耐药性。尽管此处报告的发现仅基于一小部分患者，但反应令人感兴趣，需要进一步调查。目前，这里描述的患者可用的治疗选择包括强化多药化疗，循环使用各种组合或进入研究试验。随着CAR‐T治疗进入MM治疗的早期阶段，并且越来越多的患者开始对此方法产生抗药性，以新的作用机制为基础的基于selinexor的治疗方案可能会提供重要的治疗益处，这是复发和难治性疾病的关键，2014）。