Aponermin or placebo in combination with thalidomide and dexamethasone in the treatment of relapsed or refractory multiple myeloma (CPT-MM301): a randomised, double-blinded, placebo-controlled, phase 3 trial

CPT-MM301 was a multicentre, randomized, double-blinded, placebo-controlled phase 3 trial conducted in China. The trial protocol was designed by sponsors and investigators, and was approved by the independent ethics committees of Beijing Chao-Yang Hospital Capital Medical University. The trial was conducted in accordance with the principles of Good Clinical Practice and the Declaration of Helsinki. All patients provided written informed consent before enrolment. The data were collected, analysed, and interpreted by investigators and sponsor. Investigators had full accessibility to all data. The trial was registered at http://​www.​chictr.​org.​cn as ChiCTR-IPR-15006024, 17/11/2014.

Eligible participants were enrolled in this study. Key inclusion criteria were as follows: participants' age: 18–75 years; previous treatment with two or more regimens for MM and not considered for bone marrow transplantation; M-protein levels needed to meet at least one of the following criteria: serum M-protein ≥ 10 g/L (IgG, IgM, IgD subtype), or ≥ 7.5 g/L (IgA subtype), or urinary M-protein ≥ 200 mg/24 h; absolute neutrophil count ≥ 1.0 × 10⁹/L; platelet ≥ 50 × 10⁹/L; aspartate transaminase (AST) ≤ 2.5 × upper limit of normal (ULN); alanine transaminase (ALT) ≤ 2.5 × ULN; alkaline phosphatase ≤ 2.5 × ULN; total bilirubin (TBIL) ≤ 1.5 × ULN; creatinine clearance rate ≥ 30 ml/min. Key exclusion criteria included: refractoriness to TD or lenalidomide plus dexamethasone (RD) regimens of the last treatment; received any anti-MM drug treatment within 4 weeks before the trial; participated in aponermin clinical trials previously; had serious organic or mental diseases. (see Supplementary file).

An allocation ratio preserving biases coin randomization was used in this trial (block size of 6) [23]. The random allocation sequence was generated using SAS 9.2 and uploaded to an interactive web response system (IWRS) by an unblinded system administrator. Eligible patients were enrolled and randomly assigned (2:1) by investigators via the IWRS to receive either aponermin plus TD (aponermin group), or placebo plus TD (placebo group). Randomization was stratified according to the number of prior therapeutic regimens (≤ 3 or > 3), the status of TD/RD therapy (yes vs. no), and International Staging System (ISS) stage (stage I vs. stage II or III). Aponermin and placebo were packaged in a blinded manner under the supervision of a statistician according to the drug list. The packaging and labels of aponermin and placebo were identical to ensure that they remained masked to the treatment assignment. The investigators, participants, research staff, members of the independent assessment committee (IAC), and sponsor study team were masked to the treatment location.

In each cycle, patients were administered 10 mg/kg of aponermin or placebo via intravenous infusion on days 1–5, oral thalidomide 150 mg on days 1–28, and oral dexamethasone 40 mg on days 1–4 for 18 cycles (28 days for each cycle). In the first cycle, thalidomide was administered on days 2–28, and dexamethasone on days 2–5 to observe the changes in AST, ALT, and lactate dehydrogenase (LDH) after the first dose of aponermin alone. Treatment was continued for up to 18 cycles or until progressive disease (PD), unacceptable toxicities, or withdrawal from the study. After completing 18 cycles of treatment, patients might receive further treatment based on the investigators' opinions.

The primary endpoint was PFS, defined as the time from the date of the randomization to the date of the first documented PD or death from any cause during the study, whichever occurred first. Secondary endpoints included overall survival (OS), ORR, duration of response (DOR), time to response (TTR), time to progression (TTP), safety, and health-related quality of life (HRQoL). The exploratory endpoint was to evaluate the efficacy among high-risk patients defined as [t(14;16)], [t(4;14)], or [del(17p)] by fluorescence in situ hybridization, or chromosome 13 deletion with hypodiploidy by G-band staining [24].

Serum and urine monoclonal proteins and serum free light chain levels were measured at a central laboratory. Disease status were assessed by the investigators at the baseline and the end of each cycle. The International Myeloma Working Group consensus criteria (IMWG criteria) were used to assess responses and PD. For patients in remission, if treatment was discontinued due to intolerable adverse event (AE) or the completion of 18 cycles of treatment, the disease status was assessed every six weeks until PD, death or next anti-myeloma therapy started. We required all responses and PD be assessed by IAC. PFS, ORR, DOR, TTR, and TTP were calculated based on the responses and PD assessed by IAC.

The HRQoL questionnaires were completed at baseline and the end of every cycle using the European Organization for Research and Treatment of Cancer (EORTC) questionnaires: the generic EORTC QLQ-C30 and the myeloma-specific QLQ-MY20.

The AEs and serious adverse events (SAEs) were collected up to 28 days following the last treatment dose and graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events Version 4.03.

The sample size was determined based on a conservative estimation of the median PFS of 5.0 months in the aponermin group and 3.5 months in the placebo group. According to ethical opinion, the proportion of patients in the aponermin and placebo groups was 2:1. It was estimated that a total of 286 events of PD or death over a 30-months enrollment period and a 6-months follow-up would be required to have a statistical power of 80%. This is to show superiority at a hazard ratio (HR) of 0.70 using a log-rank test (one-sided alpha is 0.025). A total of 313 patients were needed based on the calculation before factoring in the dropout rate. However, considering a 25% dropout rate, at least 417 patients were required (278 in the aponermin group and 139 in the placebo group).

No interim analyse was done in this trial. A final analysis was performed when the last participant had been enrolled for 6 months.

In this study, all efficacy analyses were based on the modified intention-to-treat population (mITT), including all randomized patients who received at least one dose of the trial treatment. The primary endpoint, PFS, was compared using a prespecified stratified log-rank test. The Kaplan–Meier method was used to estimate the median PFS and depict the curve. The 95% confidence interval (CI) was estimated using the Brookmeyer-Crowley formula. HR and 95% CI were estimated using a stratified Cox proportional hazards model. The proportional hazards assumption was assessed and met by the Cox regression model, using time-dependent explanatory variables. The aforementioned strata variables were the same as those used in the randomization. Other time-to-event data, including OS, DOR, TTR, TTP, and subgroup analyses of PFS and OS, were analysed using a method similar to PFS. The ORR, clinical benefit rate and the rate of each response were compared between the groups using chi-squared tests. The Clopper–Pearson method was used to calculate the 95% CI. Subgroup analyses of ORR were performed using similar method. Scores for the EORTC QLQ-30 and MY20 were calculated according to the developer's scoring manual. The raw scores from the scales in both questionnaires were standardized by linear transformation to range 0–100. Descriptive statistics for the baseline scores for each domain were summarized, and the differences between groups were assessed using a group t-test. A mixed-model measure analysis was used to estimate the treatment effects over time for each domain (longitudinal analysis) and assess the differences between groups. The safety analysis included all patients who received at least one dose of the trial treatment. AEs were coded using MedDRA version 22.1. The frequency of AEs was reported.

From February 25, 2015, to July 3, 2019, 508 patients were screened in 36 hospitals in China. A total of 417 patients were eligible and randomly assigned (2:1) to the aponermin group (278 patients) or the placebo group (139 patients). Of these, 276 patients in the aponermin group and 139 in the placebo group received the study treatment and were included in the analysis of efficacy and safety. By the final analysis date (January 3, 2020), 257 (93.1%) patients in the aponermin group and 135 (97.0%) patients in the placebo group had discontinued treatment. The reasons for the discontinuation of the intervention are shown in Fig. 1. The median treatment duration for the aponermin group was significantly longer than that of the placebo group (5 vs. 3 cycles).

The baseline characteristics were well balanced between the two groups (Table 1). The median age was 59 years (range, 26–75), 42.2% were women. The median time since the initial diagnosis of MM was 2.6 years (range, 0.2–14.7). The median number of prior treatment regimens was three (range, 2–25), of which 46.0% had received at least four regimens. All patients had received previous glucocorticoids; 73.6%, proteasome inhibitor (PI); and 86.6%, immunomodulator (IMiD).

At a median follow-up of 17.2 months (95% CI, 15.1–28.2), 203 (73.6%) events of PD or death occurred in the aponermin group and 111 (79.9%) in the placebo group as assessed by the IAC. The median PFS was 5.5 months (95% CI, 4.7–6.5) in the aponermin group vs. 3.1 months (95% CI, 2.0–3.9) in the placebo group (HR, 0.62; 95% CI, 0.40–0.78; P < 0.001) (Fig. 2A). A significantly prolonged PFS was also observed based on the investigator's evaluations (Supplementary file: Table S1).

The prespecified subgroup analysis showed that the effect of aponermin on prolonging PFS compared to placebo was consistent for most subgroups (Fig. 2B). The PFS benefits of the aponermin group vs. placebo group were particularly evident in the subgroups of the patients with refractory MM (median 6.4 vs. 3.7 months; HR, 0.59; 95% CI, 0.38–0.89), the patients with prior therapy of IMiD and PI (median 4.7 vs. 2.1 months; HR, 0.55; 95% CI, 0.41–0.74), and the patients who were refractory to both IMiD and PI (median 3.7 vs. 1.8 months; HR, 0.36; 95% CI, 0.17–0.74).

At a median follow-up of 30.1 months (95% CI, 25.9–34.0), 151 (54.7%) deaths occurred in the aponermin group, and 88 (63.3%) in the placebo group. The median OS was 21.8 months (95% CI, 17.3–27.3) in the aponermin group and 17.0 months (95% CI, 12.2–23.2) in the placebo group (HR, 0.72; 95% CI, 0.56–0.94; P = 0.02) (Supplementary file: Figure S1). In an updated analysis of OS with a median follow-up of 48.0 months (95% CI, 40.0–55.7), 6.0 months extension was observed in the aponermin group compared to that in the placebo group (median, 22.4 vs. 16.4 months; HR, 0.70; 95% CI, 0.55–0.89; P = 0.003) (Fig. 3A). The prespecified subgroup analysis showed a significant effect of the aponermin group compared with the placebo group on OS for most of the subgroups (Fig. 3B).

The median DOR was 15.2 months (95% CI, 10.0–18.2) in the aponermin group vs. 9.8 months (95% CI, 4.4–14.7) in the placebo group (HR, 0.55; 95% CI, 0.29–1.05; P = 0.07). The median TTR were both 1.9 months in the two groups (HR, 0.89; 95% CI, 0.53–1.51; P = 0.67). TTP was longer in the aponermin group than in the placebo group (median, 5.8 vs. 3.5 months; HR, 0.61; 95% CI, 0.48–0.78; P < 0.001) (Table 2).

According to the assessment of the IAC, the ORR was 30.4% (95% CI, 25.1%–36.2%) in the aponermin group vs. 13.7% (95% CI, 8.4%–20.5%) in the placebo group (P < 0.001). The rates of very good partial response (VGPR) or better (14.1% vs. 2.2%), VGPR (12.0% vs. 1.4%), and clinical benefit (MR or better) (45.3% vs. 29.5%) were all superior in the aponermin group than those in the placebo group (Table 2). In a prespecified subgroup for patients who had achieved PR or better responses, greater benefits in PFS (median, 17.6 vs. 10.7 months; HR, 0.584; 95% CI, 0.31–1.11; P = 0.10) and OS (median, 42.9 vs. 31.6 months; HR, 0.41; 95% CI, 0.18–0.92; P = 0.03) were observed. The ORR and other response rates assessed by the investigators were similar to those assessed by the IAC (Supplementary file: Table S1).

AEs that occurred in 15% or more of the patients in either group are shown in Table 3. The main grade 3 or 4 AEs included neutropenia (26.8% vs. 26.6%), pneumonia (25.0% vs. 23.7%), and hyperglycemia (21.0% vs. 12.2%) in the aponermin and placebo groups. Hepatotoxicity occurred at a significantly higher frequency in the aponermin group than in the placebo group (ALT, 52.2% vs. 24.5%; AST, 51.1% vs. 19.4%; LDH, 44.9% vs. 21.6%). However, most of the ALT and AST elevations were grade 1 to 2, and approximately 10% of patients exhibited grade 3 or 4 elevations in the aponermin group. All hepatotoxicity events were transient and returned to normal or baseline levels before the next treatment. The incidence of dose adjustment or discontinuation due to hepatotoxicity was < 3%. No case of liver failure or death due to aponermin-related hepatotoxicity was reported.

The incidence of AEs leading to treatment termination were similar between the two groups (8.7% vs. 7.2%), and the most common AE was infectious pneumonia (1.8% vs. 3.6%). All of the AEs were transient and reversible.

SAEs were reported in 112 (40.6%) of 276 patients in the aponermin group and 52 (37.4%) of 139 patients in the placebo group. Pneumonia was the most common SAE (20.3% vs. 20.9%). SAEs occurred in 1% or more of the patients in either group are shown in Table S2.

The mean scores for each domain of the EORTC QLQ-C30 and MY20 at baseline have no difference between the two groups (Supplementary file: Table S3). The Least-Squares (LS) mean changes in scores from baseline over the treatment cycles showed that the difference between groups favored the aponermin group over the placebo group for global health status, emotional functioning, social functioning, fatigue, constipation, and financial difficulties in the QLQ-C30 (all P-values < 0.05). For other domains, no significant differences between groups were observed. Future perspective, body image, and disease symptoms of QLQ-MY20 in the aponermin group were significantly better than those in the placebo group (all P-values < 0.05). The LS mean changes for disease symptoms and side effects of treatment were stable across treatments in the aponermin group and was not worse than that in the placebo group. (Supplementary file: Table S4).