Selinexor plus low-dose dexamethasone in Chinese patients with relapsed/refractory multiple myeloma previously treated with an immunomodulatory agent and a proteasome inhibitor (MARCH): a phase II, single-arm study

The open-label, single-arm, phase II MARCH study was conducted at 17 sites in China. The study addresses Chinese regulatory needs by providing a bridge to the global phase II STORM study, using consistent primary endpoint, eligible criteria, treatment regimens, and disease assessment schedule. STORM eligible patients had previously been treated with lenalidomide, pomalidomide, bortezomib, carfilzomib, and daratumumab. They had disease refractory to at least one IMiD, one PI, daratumumab, and their most recent regimen. In contrast, eligible patients in MARCH had been treated and refractory to one IMiD (lenalidomide), one PI (bortezomib), and the most recent regimen defined by domestic drug accessibility. Daratumumab had not been approved in China when MARCH started but was conditionally approved for marketing in China during the course of MARCH. Therefore, a number of patients who failed daratumumab were subsequently enrolled in the study. In addition, eligible RRMM patients had measurable disease based on International Myeloma Working Group (IMWG) guidelines [13], had an Eastern Cooperative Oncology Group (ECOG) [14] performance score of 0–2, and had adequate organ system function including renal, hepatic, and hematopoietic functions. All patients provided written informed consent prior to enrollment. The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines, following approval by independent ethics committees or institutional review boards at each study site. This trial is registered at ClinicalTrials.gov, NCT03944057.

Patients received selinexor 80 mg combined with dexamethasone 20 mg orally on day 1 and day 3 of each week, in 4-week cycles until progressive disease (PD), death, unmanageable toxicity, or withdrawal of consent, whichever occurred first.

The primary endpoint for the study was the overall response rate (ORR) per IMWG as determined by an independent review committee (IRC). Secondary endpoints included overall survival (OS), progression-free survival (PFS), time to response (TTR), duration of response (DOR), safety, and PK profile.

Response assessment was performed every 28 days on day 1 of each cycle. Adverse event (AE) was assessed using the NCI Common Terminology Criteria for Adverse Events (CTCAE) version 4.03. Blood sampling (3 mL per sampling point) was collected on day 1 (single-dose) and day 15 (multiple-dose) of cycle 1 to characterize the selinexor PK profile. Plasma concentrations were determined at the following distinct time points: pre-dose, 0.5 h, 1 h, 1.5 h, 2 h, 3 h, 4 h, 5 h, 6 h, 8 h, 10 h, 24 h, and 48 h. Pharmacokinetics ethnicity sensitivity analysis was performed to evaluate the potential impact of ethnic factors on the acceptability and difference in the data of Chinese and western patients [15‐19].

Efficacy was also evaluated in the specified subgroups, including triple-class refractory and triple-class exposed, as well as baseline age, creatinine clearance, MM subtype, revised-International Staging System (R-ISS) stage, high-risk cytogenetic abnormality by FISH, prior therapy, and the number of regimens.

Plasma sample analysis was performed using validated high-performance liquid chromatography tandem mass spectrometry method to determine the blood concentration of selinexor. PK parameters were obtained by analyzing adequate plasma drug concentration data and calculated using the NCA model with Phoenix WinNonlin v8.3.1. The descriptive statistics of PK parameters were summarized using SAS 9.4.

Between August 2019 and February 2021, 103 patients were assessed for eligibility for the study. A total of 82 patients were enrolled and received study treatment. As of the cut-off date of May 06, 2021, 9 patients were still receiving study treatment, and 37 patients were still on safety or survival follow-up (Fig. 1). Seventy-three patients (89.0%) discontinued study treatment. The main reasons for that included disease progression (48.8%), patients’ decision (24.4%), and adverse events (8.5%). All 82 patients were included in the mITT or safety population for efficacy and safety analyses.

Baseline demographics and disease characteristics are summarized in Table 1. Patients ranged from 42 to 82 years in age, with a median of 60.0 years. The median duration from initial diagnosis was 3.2 years (range 0.2–13.4). Most patients (72%) had baseline R-ISS stage II/III. A total of 55 patients (67.1%) had high-risk cytogenetic abnormalities with 22.0% del (17p13), 25.6% t(4;14), 6.1% t(14;16), and 64.6% 1q amplification. Eighteen patients (22.0%) had baseline plasmacytoma and 18 patients (22.0%) had creatinine clearance < 60 mL/min.

Patients enrolled had highly refractory disease and were heavily pretreated. The median number of prior regimens was 5 (range 1–16). Among 82 patients, 18 patients (22.0%) received autologous stem cell transplantation (ASCT) and 10 patients (12.2%) had prior CAR-T therapy. All 82 patients (100%) were documented refractory to lenalidomide and bortezomib, of whom 20 patients (24.4%) were also refractory to daratumumab. These 20 patients constituted the triple-class refractory population, with the median prior regimen number of 6 (range 1–14).

The primary efficacy endpoint for the study was ORR per IMWG by an IRC. The ORR in the mITT population was 29.3% (95% CI 19.7, 40.4), which included 4 (4.9%) patients with VGPR and 20 (24.4%) patients with PR. The lower limit of the 95% CI for ORR (19.7%) was greater than the pre-defined threshold of 15%, confirming that the primary endpoint had been reached and that the ORR of 29.3% could be considered to have demonstrated statistically significant efficacy of the Sd regimen (p = .0001). The median time to PR or better was 1 month (range 0.5–8.4), and the median duration of response (DOR) was 4.7 months (95% CI 2.02, not estimable [NE]) (Table 2). Figure 2 illustrates the duration of response for all responders. It was observed that a deeper response could be obtained with a longer treatment duration.

A subgroup analysis for ORR was also performed (Fig. 3). It is noted that both triple-class refractory and triple-class exposed patients could benefit from the Sd regimen, with ORR of 25.0% and 30.4%, respectively. ORR was 25.5% in patients with high-risk cytogenetic abnormalities, including 22.2% in those with del 17p13. In addition, ORR was 28.6% in elderly patients (65–74 years), 44.4% in those with baseline R-ISS III, 42.9% in IgD MM, 33.3% in those with baseline creatinine clearance 30–<60 mL/min, 32.8% in those with prior regimen number ≥ 4, and 33.3% in those with prior stem cell transplant. In summary, the response rate was generally consistent across prespecified subgroups.

Of note, ORR was 50.0% in the 10 patients who had prior CAR-T therapy, with the median duration of response of 1.4 months. As expected, these patients had a longer disease history and received more prior regimens than the total population. The median duration from MM’s initial diagnosis was 5.2 years, and the median number of prior regimens was 9.5. Five patients experienced very rapid disease progression as indicated by a median of 46.2% increase of tumor burden from screening to C1D1. The median PFS was 1.9 months and median OS was not reached with the estimated 12-month OS rate of 68.6%. AEs reported in this subgroup were consistent with those reported in the overall population.

As of the cut-off date, with a median follow-up of 10.6 months, the median PFS in the mITT population was 3.7 months (95% CI 2.02, 4.66). The median PFS in the triple-class refractory population was 2.9 months (95% CI 1.65, 5.62). For patients who had a PR or better, the median PFS was 11.1 months (95% CI 4.66, NE) (Fig. 4).

The median OS in the mITT population was 13.2 months (95% CI 11.90, NE). The median OS in the triple-class refractory population was 11.9 months (95% CI 1.95, NE). For patients who had an MR or better, the median OS was not reached (95% CI 13.22, NE); the estimated 1-year survival rate was 60.1% (95% CI 46.9, 71.0) (Fig. 5).

The median treatment duration of Sd was 13.7 weeks (range 1.1–70.0). 54.9% of patients received treatment ≥12 weeks and 28% of patients ≥24 weeks. The median weekly dose received was 113.7 mg (range 55.4–160.0). The majority of patients had dose modifications of selinexor, including dose interruptions in 75.6% of patients and dose reductions in 84.1% of patients. Among the patients with dose reductions, the median number of dose reductions was 2; 33 patients (40.2%) had a reduction to 100–120 mg weekly and 17 patients (20.7%) had a reduction to 80 mg weekly. The median time to the first dose reduction was 22.0 days (range 4–120). The median treatment duration was 16.6 weeks (range 2–70) for the patients who had dose reduction vs 4.3 weeks (range 1–27) for those who did not have. Eight patients re-escalated the dose as prespecified after AE was stabilized for more than 4 weeks.

In terms of treatment-emergent adverse event (TEAE), overall, all 82 patients had at least one TEAE. The most common (≥50%) TEAEs included thrombocytopenia (87.8%), anemia (86.6%), leukopenia (82.9%), nausea (78.0%), lymphopenia (76.8%), neutropenia (72.0%), hyponatremia (67.1%), weight decreased (65.9%), decreased appetite (62.2%), asthenia (59.8%), hyperglycemia (56.1%), and vomiting (50.0%). The most frequent grade 3/4 TEAE (>10%) were anemia (57.3%), thrombocytopenia (51.2%), lymphopenia (42.7%), neutropenia (40.2%), hyponatremia (29.3%), lung infection (26.8%), hypokalemia (12.2%), and asthenia (9.8%)/fatigue (2.4%) (Table 3). The most common adverse events leading to dose interruption or reduction were thrombocytopenia, neutropenia, and lung infection.

Most grade 3/4 TEAEs were hematologic events, which could be closely monitored and well managed through routine blood tests, dose modification, and supportive treatment, such as blood transfusion, growth factor, or platelet stimulating factor. For example, 72.2% of patients with thrombocytopenia and 57.6% of patients with neutropenia/febrile neutropenia received supportive treatment. The incidence of grade 3/4 thrombocytopenia was comparable with that of other selinexor studies and not associated with severe hemorrhagic events. Less than 10% of patients experienced hemorrhagic events within 5 days of thrombocytopenia, with the majority (>70%) being grade 1/2. Only 3 patients discontinued treatment due to the events. Most non-hematologic TEAEs were gastrointestinal or constitutional events, predominantly limited to grade 1/2 severity and reversible through dose modification and supportive treatment. Multiple non-hematologic TEAEs could occur at the same time and correlated with each other. Treatment for nausea and vomiting was critical for successful safety management, which simultaneously improved decreased appetite, weight decrease, and asthenia/fatigue. Thus, patients were requested to receive prophylactic antiemetic and appetite enhancing therapy, including 5-HT3 receptor antagonist, NK1 receptor antagonist, appetite stimulant, and/or olanzapine. The mechanism of hyponatremia remains unknown, although it was often seen in association with loss of body fluid and/or imbalance of electrolytes. The incidence of grade 3 hyponatremia was 29.3%, but the majority of the patients had serum sodium levels between 125 and 130 mmol/L with no obvious symptoms. Laboratory monitoring is necessary, and the events were rapidly controlled with sodium supplementation. The incidence of lung infection in the study was expected. Lung infection is a common complication in patients with hematological tumors, especially in elderly patients with myeloma who have received multiple lines of therapy. At baseline, about 50% of the patients complicated with underlying respiratory diseases (including chronic bronchitis and chronic obstructive pulmonary disease). Most of the events improved with anti-infection treatment. The incidence of the events leading to treatment discontinuation was only 2.4%.

Serious adverse events (SAEs) were reported in 45 (54.9%) patients, with lung infection and thrombocytopenia (14.6% each) as the most common SAE. Grade 5 AEs occurred in 7 patients (8.5%), including infective exacerbation of bronchiectasis, sudden death, pulmonary embolism, lung infection, sepsis, and intracranial hemorrhage.

Most of TEAEs occurred in the first 2 cycles, and the incidence would decrease afterwards. A summary of incidence by study treatment period for the TEAEs of special interest is also provided in Table 4.

PK sample collection was performed in 18 patients, of whom 12 received multiple-dose PK sampling. Following a single dose of selinexor 80 mg on C1D1, selinexor PK profile (Fig. 6) was characterized by rapid absorption with median T _max of 2.88 h, mean t _1/2 of 6 h, mean C _max of 862 ng/mL, and mean AUC_0–48h of 5804 h ng/mL. PK results on C1D15 were similar to C1D1, and R _ac(Cmax) and R _ac(AUC) were 1.04 and 1.10, respectively, indicating no significant accumulation following twice-weekly administration. No significant differences were found in t _1/2, AUC_0–t, T _max, CL/F, and V_d/F between single dose and multiple dose. No significant inter-individual variability was seen as evidenced by the geometric mean percent coefficient of variation range for AUC_0–∞ (30.7–32.6%) and C _max (22.0–35.3%).

Comparing PK parameters obtained from MARCH and five global selinexor studies [15‐19], T _max and t _1/2 were similar, with median T _max of 2 to 2.95 h and mean t _1/2 of 5.84 to 6.24 h; C _max, AUC_0–t, and AUC_0–inf were slightly higher in MARCH, but considering inter-individual variation, it generally showed an overlapping distribution within these studies (Supplementary 1). There was no significant ethnic difference observed between Chinese and western populations.