Introduction
Plitidepsin is a cyclic depsipeptide originally extracted from the tunicate Aplidium albicans and currently produced by chemical synthesis that has potent in vitro activity against primary multiple myeloma (MM) tumor cells and a broad spectrum of human MM cell lines [ 1 ]. Plitidepsin interacts with the eukaryotic elongation factor eEF1A2, an oncogene overexpressed in MM cells [ 2 ]. This interaction leads to early oxidative stress, Rac1 activation, initiation of the MAPK cascade and cyclin down-regulation, induction of endoplasmic reticulum stress, and activation/deactivation of several cytoplasmic factors. These events finally result in induction of apoptosis, perturbation of the cell cycle, and a potent antiproliferative effect.
Plitidepsin (P) 5 mg/m 2 i.v. on Day 1 and 15 every four weeks (q4wk) plus low-dose dexamethasone (LD-DXM) (40 mg orally on Day 1, 8, 15 and 22 q4wk) was evaluated versus LD-DXM alone in patients with relapsed/refractory multiple myeloma (r/r MM) exposed to ≥3–6 prior treatment lines in the randomized phase III ADMYRE trial [ 3 ]. Low-dose dexamethasone regimen was used in combination with plitidepsin as experimental arm and as single agent in the control arm of ADMYRE trial because, based on data from last available publications [ 4 ] at the time of protocol writing, dexamethasone administered in combination at reduced doses was associated with a better safety profile while reporting better survival. This was agreed with the Health Authorities. ADMYRE showed a statistically significant prolongation of progression-free survival (PFS) with P + LD-DXM and a tolerable safety profile [ 3 ]. Based on these findings, the P + LD-DXM combination was approved in Australia in 2018 for the treatment of patients with r/r MM.
External Control arms (ECAs) help to evaluate the effectiveness of a new treatment by comparing its outcomes with those of a similar group of patients who received standard care or a different treatment. An ECA can be created from patient-level data from individuals who were external to the investigational trial and selected with statistical methods such as propensity score modeling to provide confidence that the baseline characteristics of the selected external patients are balanced and comparable with the baseline characteristics of the patients treated with the investigational product to whom they are compared [ 5 , 6 ]. After an ECA is created with a similar distribution of baseline characteristics to the investigational arm patients, the differences in outcomes between the two groups of patients can be reliably attributed to the investigational product.
Historical clinical trials (HCT) data and real-world (RWD) data are the common data sources to form an ECA, although HCT data has advantages over RWD data for ECA construction including availability of traditional clinical trial endpoints and covariates, arising from a more similar patient pool in that these patients were participating in a clinical trial, and a rigorous and more complete collection of data often done at prestigious medical centers [ 7 – 9 ]. A scoping review examining the use of ECAs in oncology from 1996 to 2022 showed that 44% of them were conducted in blood-related cancer studies [ 10 ]. ECAs based on HCT or RWD data have been previously used to conduct adjusted comparisons of treatments in r/r MM: teclistamab versus an ECA of four daratumumab clinical trials (POLLUX, CASTOR, EQUULEUS, and APOLLO) [ 11 ]; ciltacabtagene autoleucel in CARTITUDE-1 trial versus an ECA of three daratumumab clinical trials (POLLUX, CASTOR, and EQUULEUS) [ 12 ]; talquetamab in MonumenTAL-1 trial versus an ECA of two observational RWD studies (LocoMMotion and MoMMent) [ 13 ]; or talquetamab in MonumenTAL-1 trial versus an ECA from patients in the Flatiron Health database [ 14 ].
Pomalidomide (POM) is an appropriate comparator to plitidepsin as it was approved while ADMYRE trial was being conducted and both drugs were evaluated in a similar population: patients with r/r MM, who receive at least two prior treatment regimens that include both a proteasome inhibitor and an immunomodulatory drug. POM plus LD- DXM was evaluated in the pivotal randomized phase III study MM-003 in r/r MM patients [ 15 ]. In an indirect comparison (ADMYRE versus MM-003), the median overall survival of POM + LD-DXM (OS = 12.7 months) was quite similar to the median OS of 11.6 months observed for P + LD-DXM in ADMYRE [ 3 ], while the safety profile of P + LD-DXM compared favorably with respect to POM + LD-DXM. Grade ≥ 3 neutropenia (16% for P + LD-DXM vs. 48% with POM + LD-DXM) and febrile neutropenia (0.6% vs. 10%) were lower with P + LD-DXM. Severe complications derived from low neutrophil counts were less frequent with P + LD-DXM: grade 2 infection (1.8% vs. 30% with grade ≥ 3 infection with POM + LD-DXM); grade ≥ 3 pneumonia (2.6% vs. 14%). Gastrointestinal and muscular events were more common with P + LD-DXM.
To provide a direct matched comparison with an ECA between P + LD-DXM (ADMYRE data) and POM + LD-DXM, we show here an analysis using individual patient-level data from several contemporary POM trials.
Materials and methods
An ECA was prospectively designed using patient-level external data from HCT including the population with the same disease condition evaluated in ADMYRE (adult patients with r/r MM). Only patients from HCT who were assigned to receive POM + LD-DXM were eligible for inclusion in the ECA, which was constructed using a propensity score approach to balance the ECA patients with the important baseline characteristics and prognostic factors of the ADMYRE patients. HCT which included adult patients with r/r MM were extracted from Medidata Enterprise Data Store (MEDS). MEDS is a collection of thousands of previous clinical trials with patient-level data recorded through the Medidata electronic data capture system, Rave ® .
Objectives of the analysis
The objectives of this analysis were to assess the difference in OS by comparing the ADMYRE P + LD-DXM patients to the ECA POM + LD-DXM patients, and by comparing the ECA POM + LD-DXM patients to the ADMYRE LD-DXM alone patients. Safety, based on the occurrence of treatment-related adverse events (TRAEs), was also assessed. PFS, the primary endpoint in the ADMYRE study [ 3 ], was not included in this analysis because different International Myeloma Working Group criteria versions were used to calculate PFS events: without confirmation of disease progression (PD) in ADMYRE, and with confirmation of PD in the POM studies. Subsequent therapies could start without PD confirmation in the ADMYRE trial, while subsequent therapies could have been guided to start only after PD confirmation in the POM trials. Therefore, unbiased comparison for PFS based on the confirmation of PD between ADMYRE and POM trials cannot be fully achieved.
Construction of the external control arms
Multiple steps for searching and selecting the qualified historical clinical studies to be used for ECA construction were implemented, including both trial-level and patient-level selection. The initial examination began with a search for all clinical trials of r/r MM where the study treatment contained POM + LD-DXM across major trial repositories (NIH/ClinicalTrials.gov, EudraCT, UMIN-CTR) and publication databases (PubMed) and required trial confirmation (via design description or publication of results) to enroll at least some patients treated with POM + LD-DXM; to recruit at least some patients between 1 June 2010 and 31 December 2017 (the ADMYRE trial was conducted between 29 June 2010 and 19 May 2017); and an interventional trial design. Eight contemporary trials ( NCT01053949 , NCT01311687 , NCT01712789 , NCT02654132 , NCT02726581 , NCT02916420 , NCT02990338 and NCT03151811 ) met all the criteria described in Table 1 and were selected. Trial design elements (including objectives, endpoints measured, comparability of control arm treatments and masking status) were evaluated to ensure comparability with those of the ADMYRE trial.
Table 1
Key features of the ADMYRE trial and historical clinical trials
a Patients meeting the following requirements were included: patients studied in a trial covered by Medidata data sharing agreements; patients studied in a trial currently meeting Medidata completion eligibility requirements; patients studied in a trial that collects targeted endpoints and adequate planned follow-up, and patients who met the trial- and or patient-level project criteria. Abbreviations: LD-DXM low-dose dexamethasone, ORR overall response rate, OS overall survival, P plitidepsin, PFS progression-free survival, POM pomalidomide
| Study attributes | ADMYRE | Historical data a |
| Trial phase | III | II/III |
| Intervention(s) | P + LD-DXM and LD-DXM alone | POM + LD-DXM |
| Intervention assignment | Parallel assignment | Single arm Parallel assignment |
| Masking | None (open label) | None (open label) |
| Study start year | 2010 | 2009–2017 |
| Study completion year | 2017 | 2013–2022 |
| Key endpoints | OS, PFS, ORR, safety | OS, PFS, ORR, safety |
| Planned follow-up duration (primary endpoints) | Approximately 5 years | Approximately 5 years |
| Region | Global | Global |
For all studies included in this analysis, patients provided written informed consent, and an independent ethics committee or institutional review board at each study center approved the study protocol. These studies were conducted in accordance with the Declaration of Helsinki and the International Conference on Harmonisation Guidelines for Good Clinical Practice.
After selecting the eight candidate HCTs, those studies that were covered by Medidata data sharing agreements and met Medidata completion eligibility requirements were further evaluated using the patient-level data. Only HCT patients who were assigned to receive the comparator of interest, POM + LD-DXM, and met similar key eligibility criteria from the ADMYRE study (Supplementary Table 1) were included in the potential ECA eligible pool. For ECA-eligible patients, POM was to be received orally at a dose of 4 mg on Day 1 through 21 and low-dose LD-DXM 40 mg orally (20 mg for patients > 75 years) per day on Day 1, 8, 15, and 22 q4wk. In ADMYRE trial, plitidepsin 5 mg/m² was administered as an intravenous infusion on Day 1 and 15 q4wk in combination with low-dose LD-DXM (40 mg orally on Day 1, 8, 15 and 22 q4wk vs. low-dose LD-DXM (40 mg orally on Day 1, 8, 15 and 22 q4wk).
Two separate ECAs were created using the propensity score method to conduct two sets of comparative analyses in matched populations:
ECA1: P + LD-DXM (ADMYRE) was compared to the POM + LD-DXM arm.
ECA2: POM + LD-DXM was compared to the LD-DXM arm (ADMYRE).
The resultant number of patients for these comparisons is summarized in Fig. 1 . Median (range) patient age in the ADMYRE matched P + LD-DXM population was 64.0 (36.0–85.0) years and median (range) patient age in the POM + LD-DXM matched population (ECA1) was 64.9 (37.0–87.0) years. Median (range) patient age in the POM + LD-DXM matched population (ECA2) was 66.0 (37.0–87.0) years and median (range) patient age in the ADMYRE matched LD-DXM population was 66.0 (42.0–85.0) years (Supplementary Table 2).
Propensity score and covariates assessment
Propensity scores, which is considered as a summary score combining all baseline characteristics, were defined as the probability of a patient belonging to the ADMYRE trial and being assigned to P + LD-DXM in ECA1 construction or LD-DXM in ECA2 construction, conditional on the baseline characteristics (i.e., potential confounders) using logistic regression, i.e., where T = 1 if a patient was enrolled in ADMYRE and was assigned to P+LD-DXM (ECA1) or LD-DXM (ECA2), or T = 0 if a patient was assigned to receive POM+LD-DXM in HCTs; and X is a vector representing the covariates included in the propensity score model. Greedy nearest-neighbor matching algorithm [ 16 ] without replacement and a fixed 1-to-2 matching ratio was used. After a performance status was estimated for each patient from both ADMYRE and eligible ECA populations, the common support region was constructed and only subjects within this region were eligible for matching. The common support region was defined as the largest interval that contains propensity scores for subjects in both the ADMYRE and HCT groups, extended by 0.25 times the pooled estimate of the standard deviation of the logit of the propensity score in the ADMYRE and HCT groups, which is the default definition within the SAS PSMATCH procedure (SAS Institute Inc. 2017. SAS/STAT ® 14.3 User’s Guide. Cary, NC). Supplementary Fig. 1 presents two sets of box plots illustrating the distribution of propensity scores for patients that were included in the before- and after-matching sets. The comparability between the P + LD-DXM arm and POM+LD-DXM arm (ECA1) or between POM + LD-DXM arm (ECA2) and ADMYRE LD-DXM arm was improved after matching. The difference between the arms reduced substantially as evidenced by the placement of the median lines and mean markers.
The propensity score was used as a balancing score (see Supplementary Table 3 and Supplementary Table 4 for details on covariate selection). The distribution of measured baseline covariates was assessed in the after-matching analysis sets if was similar between the ADMYRE P + LD-DXM patients or ADMYRE LD-DXM randomized patients and the corresponding ECA patients in ECA1 and ECA2, respectively. Absolute standardized difference in covariate means was computed and compared following Austin’s notations [ 17 ]. The absolute standardized differences should generally be less than 0.25 [ 18 ]. An absolute standardized difference of less than 0.10 was taken to indicate a negligible difference in the mean or prevalence of a covariate between treatment groups [ 19 ]. The propensity score model that best achieved these thresholds was selected and implemented. After balance was achieved, the efficacy analysis of OS and safety analysis were performed in each ECA. Supplementary Fig. 2 shows the standardized differences after matching all fall within the highlighted range between − 0.10 and 0.10, indicating a negligible difference in the mean or prevalence of each covariate.
Statistical methods
SAS software (version 9.4 or higher) was used for all statistical analyses. Summary descriptive statistics by treatment group are provided. For continuous variables, the descriptive statistics included the number of observations, mean, 95% confidence interval (CI), standard deviation, median, first quartile, third quartile, minimum, and maximum. For categorical variables, the descriptive statistics contained the frequency and percentage.
OS was defined as the time (in months) from the index date to death due to any cause or last contact date (censoring). The index date was defined as the date of randomization for patients from randomized studies, including ADMYRE and randomized HCT, or date of enrolment for patients from single-arm HCT. OS was estimated using the Kaplan-Meier method and compared between the ADMYRE and ECA patients using the after-matching analysis sets. Two-sided log-rank tests were used for testing the differences between the ADMYRE target groups and corresponding ECAs. The hazard ratio (HR) between the two treatment groups in each ECA analysis and its 95% CI using Wald’s confidence limits was estimated using a Cox regression model, including treatment group (i.e., ADMYRE vs. ECA) as a covariate. Ties were handled with Efron’s approximation. In ECA1 analysis, if the upper limit of the 95%CI for HR was below a prespecified non-inferiority margin of 1.3, it can be concluded that ADMYRE P + LD-DXM is non-inferior to POM + LD-DXM. This non-inferiority margin of 1.3 is based on a systematic review from 74 non-inferiority trials of cancer drug therapies, and on a systematic literature review of 285 published oncology clinical trials where the median non-inferiority margin was a HR of 1.29 [ 20 , 21 ]. In the ECA2 analysis, if the upper limit of the 95%CI for HR was less than 1, POM + LD-DXM can be concluded as superior to LD-DXM (ADMYRE).
All statistical tests were two-sided at alpha level of 0.05.
Results
ECA1: P + LD-DXM arm (ADMYRE) versus POM + LD-DXM arm
The HR of OS for the P + LD-DXM arm (ADMYRE) relative to POM + LD-DXM (ECA1) was 1.009 (95%CI, 0.812–1.254) (two-sided log-rank test p = 0.9336) (Fig. 2 ). Kaplan-Meier plot shows that the P + LD-DXM arm and the POM + LD-DXM arm have similar survival estimates, with the curves close to one another and crossing at multiple points. As the 95%CI for the HR (0.812–1.254) falls below the pre-specified non-inferiority margin of 1.3, a strong signal of non-inferiority was found in OS for the ADMYRE P + LD-DXM treatment compared to POM + LD-DXM. Additionally, the median survival time and the log-rank p-value do not indicate signs of a difference between the arms.
A lower percentage of patients with any grade ≥ 3 TRAEs (50.3% vs. 63.9%) was found for the P + LD-DXM patients compared to the POM + LD-DXM patients (Table 2 ). Overall, the safety profile of P + LD-DXM (ADMYRE) when compared to POM + LD-DXM is characterized by a low rate of hematological TRAEs (e.g., grade ≥ 3 neutropenia 2.5% vs. 37.1%; grade ≥ 3 thrombocytopenia 2.5% vs. 13.2%) and infections (8.1% vs. 18.7%), while gastrointestinal, biochemical and musculoskeletal events were more common with P + LD-DXM.
Table 2
Treatment-related adverse events (≥ 5% of patients) (ECA1)
Data shown are n (%) of patients. Ordered by frequency in the first column A dash (“-“) indicates that the value did not reach the threshold of 5% of patients TRAEs were coded using MedDRA version 16.0. For a patient who had multiple occurrences of the same SOC and PT, the patient was counted only once for the corresponding SOC and PT. SOC is presented alphabetically and PTs within each SOC are presented in descending order of prevalence in the P + LD-DXM arm Abbreviations: ALT alanine aminotransferase, AST aspartate aminotransferase, CPK creatine phosphokinase, ECA external control arm, MedDRA Medical Dictionary for Regulatory Activities, P + LD-DXM plitidepsin plus low-dose dexamethasone, POM + LD-DXM pomalidomide plus low-dose dexamethasone, PT preferred term, SOC system organ class, TRAE treatment-related adverse event
| System Organ Class Preferred Term | All grades | Grade ≥ 3 | ||
| P + LD-DXM (ADMYRE) ( n = 161) | POM + LD-DXM (ECA1) ( n = 310) | P + LD-DXM (ADMYRE) ( n = 161) | POM + LD-DXM (ECA1) ( n = 310) | |
| Patients with any TRAE | 138 (85.7) | 265 (85.5) | - | - |
| Patients with any grade ≥ 3 TRAE | - | - | 81 (50.3) | 198 (63.9) |
| Blood and lymphatic system disorders | 25 (15.5) | 161 (51.9) | 17 (10.6) | 141 (45.5) |
| Anaemia | 16 (9.9) | 59 (19.0) | 10 (6.2) | 42 (13.5) |
| Neutropenia | 4 (2.5) | 127 (41.0) | 4 (2.5) | 115 (37.1) |
| Thrombocytopenia | 4 (2.5) | 57 (18.4) | 4 (2.5) | 41 (13.2) |
| Leukopenia | 0 | 22 (7.1) | - | - |
| Gastrointestinal disorders | 85 (52.8) | 85 (27.4) | - | - |
| Nausea | 61 (37.9) | 22 (7.1) | - | - |
| Vomiting | 28 (17.4) | 5 (1.6) | - | - |
| Diarrhoea | 23 (14.3) | 25 (8.1) | - | - |
| Abdominal pain upper | 10 (6.2) | 4 (1.3) | - | - |
| Constipation | 5 (3.1) | 23 (7.4) | - | - |
| General disorders and administration site conditions | 81 (50.3) | 126 (40.6) | 22 (13.7) | 28 (9.0) |
| Fatigue | 60 (37.3) | 87 (28.1) | 18 (11.2) | 18 (5.8) |
| Oedema peripheral | 18 (11.2) | 28 (9.0) | - | - |
| Pyrexia | 13 (8.1) | 24 (7.7) | - | - |
| Infections and infestations | 31 (19.3) | 104 (33.5) | 13 (8.1) | 58 (18.7) |
| Pneumonia | 7 (4.3) | 33 (10.6) | 4 (2.5) | 29 (9.4) |
| Upper respiratory tract infection | 3 (1.9) | 16 (5.2) | - | - |
| Investigations | 64 (39.8) | 46 (14.8) | 41 (25.5) | 28 (9.0) |
| Blood CPK increased | 26 (16.1) | 0 | 23 (14.3) | 0 |
| ALT increased | 25 (15.5) | 2 (0.6) | 14 (8.7) | 1 (0.3) |
| AST increased | 15 (9.3) | 0 | 9 (5.6) | 0 |
| Electrocardiogram QT prolonged | 9 (5.6) | 0 | - | - |
| Neutrophil count decreased | 0 | 19 (6.1) | 0 | 18 (5.8) |
| Metabolism and nutrition disorders | 32 (19.9) | 29 (9.4) | - | - |
| Decreased appetite | 21 (13.0) | 7 (2.3) | - | - |
| Hyperglycaemia | 8 (5.0) | 13 (4.2) | - | - |
| Musculoskeletal and connective tissue disorders | 48 (29.8) | 48 (15.5) | 18 (11.2) | 3 (1.0) |
| Myalgia | 24 (14.9) | 4 (1.3) | 9 (5.6) | 0 |
| Muscular weakness | 16 (9.9) | 9 (2.9) | - | - |
| Muscle spasms | 4 (2.5) | 31 (10.0) | - | - |
| Nervous system disorders | 16 (9.9) | 62 (20.0) | - | - |
| Peripheral sensory neuropathy | 1 (0.6) | 19 (6.1) | - | - |
| Tremor | 0 | 18 (5.8) | - | - |
| Psychiatric disorders | 20 (12.4) | 55 (17.7) | - | - |
| Insomnia | 11 (6.8) | 21 (6.8) | - | - |
| Respiratory, thoracic and mediastinal disorders | 22 (13.7) | 67 (21.6) | - | - |
| Dyspnoea | 11 (6.8) | 24 (7.7) | - | - |
| Cough | 4 (2.5) | 18 (5.8) | - | - |
| Skin and subcutaneous tissue disorders | 8 (5.0) | 49 (15.8) | - | - |
| Rash | 3 (1.9) | 16 (5.2) | - | - |
ECA2: POM + LD-DXM versus LD-DXM Arm (ADMYRE)
The HR of OS for the POM + LD-DXM arm (ECA2) relative to the ADMYRE LD-DXM arm was 0.762 (95%CI, 0.566–1.026) (two-sided log-rank test p = 0.0733) (Fig. 3 ).
POM + LD-DXM was associated with a higher rate of grade ≥ 3 TRAEs compared to the control arm used in the ADMYRE (LD-DXM): 60.0% vs. 10.4% of patients (Table 3 ), particularly notable for hematological and infection events.
Table 3
Treatment-related adverse events (≥ 5% of patients) (ECA2)
Data shown are n (%) of patients. Ordered by frequency in the first column A dash (“-“) indicates that the value did not reach the threshold of 5% of patients TRAEs were coded using MedDRA version 16.0. For a patient who had multiple occurrences of the same SOC and PT, the patient was counted only once for the corresponding SOC and PT. SOC is presented alphabetically and PTs within each SOC are presented in descending order of prevalence in the P + LD-DXM arm Abbreviations: ALT alanine aminotransferase, AST aspartate aminotransferase, CPK creatine phosphokinase, ECA external control arm, MedDRA Medical Dictionary for Regulatory Activities, P + LD-DXM plitidepsin plus low-dose dexamethasone, POM + LD-DXM pomalidomide plus low-dose dexamethasone, PT preferred term, SOC system organ class, TRAE treatment-related adverse event
| System Organ Class Preferred Term | All grades | Grade ≥ 3 | ||
| POM + LD-DXM (ECA2) ( n = 145) | LD-DXM (ADMYRE) ( n = 77) | POM + LD-DXM (ECA2) ( n = 145) | LD-DXM (ADMYRE) ( n = 77) | |
| Patients with any TRAE | 124 (85.5) | 35 (45.5) | - | - |
| Patients with any grade ≥ 3 TRAE | - | - | 87 (60.0) | 8 (10.4) |
| Blood and lymphatic system disorders | 72 (49.7) | 1 (1.3) | 64 (44.1) | 1 (1.3) |
| Neutropenia | 55 (37.9) | 0 | 49 (33.8) | 0 |
| Anaemia | 34 (23.4) | 1 (1.3) | 25 (17.2) | 1 (1.3) |
| Thrombocytopenia | 25 (17.2) | 0 | 16 (11.0) | 0 |
| Leukopenia | 8 (5.5) | 0 | - | - |
| Gastrointestinal disorders | 42 (29.0) | 13 (16.9) | - | - |
| Constipation | 15 (10.3) | 1 (1.3) | - | - |
| Diarrhoea | 13 (9.0) | 2 (2.6) | - | - |
| Nausea | 8 (5.5) | 7 (9.1) | - | - |
| Dyspepsia | 4 (2.8) | 4 (5.2) | - | - |
| General disorders and administration site conditions | 59 (40.7) | 10 (13.0) | 14 (9.7) | 1 (1.3) |
| Fatigue | 40 (27.6) | 7 (9.1) | - | - |
| Pyrexia | 14 (9.7) | 2 (2.6) | - | - |
| Oedema peripheral | 12 (8.3) | 1 (1.3) | - | - |
| Infections and infestations | 46 (31.7) | 8 (10.4) | 29 (20.0) | 3 (3.9) |
| Pneumonia | 19 (13.1) | 0 | 17 (11.7) | 0 |
| Bronchitis | 8 (5.5) | 0 | - | - |
| Investigations | 27 (18.6) | 0 | 18 (12.4) | 0 |
| Neutrophil count decreased | 14 (9.7) | 0 | 14 (9.7) | 0 |
| Musculoskeletal and connective tissue disorders | 25 (17.2) | 9 (11.7) | - | - |
| Muscle spasms | 18 (12.4) | 1 (1.3) | - | - |
| Nervous system disorders | 29 (20.0) | 5 (6.5) | - | - |
| Peripheral sensory neuropathy | 10 (6.9) | 0 | - | - |
| Psychiatric disorders | 26 (17.9) | 11 (14.3) | - | - |
| Insomnia | 10 (6.9) | 8 (10.4) | - | - |
| Respiratory, thoracic and mediastinal disorders | 29 (20.0) | 5 (6.5) | - | - |
| Dyspnoea | 12 (8.3) | 2 (2.6) | - | - |
| Cough | 8 (5.5) | 1 (1.3) | - | - |
| Skin and subcutaneous tissue disorders | 27 (18.6) | 5 (6.5) | - | - |
| Rash | 8 (5.5) | 0 | - | - |
Discussion
Overall survival results showed non-inferiority for patients assigned to be treated with P + LD-DXM from ADMYRE [ 3 ] compared to matched patients treated with POM + LD-DXM in HCT. No difference in OS was observed between P + LD-DXM and POM + LD-DXM (HR = 1.009; 95%CI, 0.812–1.254). Additionally, the log-rank test p-value of 0.9336 does not indicate a statistically significant difference in OS.
The treatment effect for POM + LD-DXM obtained by comparing POM + LD-DXM in HCT to the LD-DXM control data from the ADMYRE trial (HR = 0.762; 95%CI, 0.566–1.026) was similar to the treatment effect previously demonstrated for P + LD-DXM compared to LD-DXM in the ADMYRE trial (HR = 0.797, 95%CI, 0.596–1.067) [ 3 ].
Subsequent line therapies may have an impact on overall survival outcomes. The fact that trials were contemporary might help to reduce the diversity of further therapies and then its likely effect on survival. Nevertheless, it has to be remarked that the similar effect on OS observed for the arms evaluated was obtained despite a lower percentage of patients treated in the ADMYRE trial arms received subsequent therapy: 47.9% (P + LD-DXM) vs. 65.6% (POM + LD-DXM) in ECA1; and 55.1% (POM + LD-DXM) vs. 71.1% (LD-DXM) in ECA2 (Supplementary Table 2).
In the ECA1 patient-level comparison, the safety profile of P + LD-DXM (ADMYRE) when compared to POM + LD-DXM (ECA1) is characterized by a low rate of hematological adverse events and infections, with a lower percentage of grade ≥ 3 TRAEs, while gastrointestinal, biochemical and musculoskeletal events were more common with P + LD-DXM. The different safety profile of plitidepsin to that of other agents approved for r/r MM like pomalidomide should be noted. Plitidepsin is not associated with renal impairment, neutropenia, thrombocytopenia, thromboembolism, rash or neurotoxicity, or infections, as is the case with many agents used in r/r MM in clinical trials contemporaneous to ADMYRE [ 15 , 22 – 27 ]. This differentiated safety profile makes plitidepsin a valuable treatment option in patients known to be at risk of particular adverse event known to occur with the use of these agents.
eEF1A2, a protein which is overexpressed in MM, has been identified as the primary target for plitidepsin [ 2 , 28 ]. Of note, all recently introduced new anti-MM drugs have mechanisms of activity not targeting eEF1A2.
In the other patient-level comparison (ECA2), POM + LD-DXM was associated with a higher rate of events compared to the control arm used in ADMYRE (LD-DXM). Therefore, the finding of more toxicity reported for P + LD-DXM in the ADMYRE trial [ 3 ], which was expected from a combination when compared to a monotherapy, was also observed with POM + LD-DXM.
Conclusions
This patient-level external data comparison shows that P + LD-DXM is non-inferior to POM + LD-DXM in terms of overall survival in a similar population than that evaluated in the AMDYRE trial. The safety profile of P + LD-DXM is characterized by a low rate of hematological adverse events and infections compared to POM + LD-DXM. A new therapy with a novel mechanism of action that shows a PFS benefit, as proven in ADMYRE study [ 3 ], along with a likelihood of survival benefit in a largely treatment-refractory population, provide plitidepsin with a specific place into the available options of the haemato-oncologist treating r/r MM and could be considered a new therapeutic option in this setting.
Supplementary Information
Below is the link to the electronic supplementary material.
Supplementary Material 1 (DOCX 168 KB)
Author contributions
Conceptualization, H.L., S.M., C.K., J.J., A.N., J.G., R.D., X.Y and M.V.M. Formal analysis, S.M., A.N., S.E., C.K., J.G, V.A., R.D. and X.Y.; Investigation, S.M., C.K., J.J., A.N., J.G., V.A., R.D., X.Y., and M.V.M.; Writing—original draft, S.M., C.K., V.A. and M.V.M.; Writing—review and editing, H.L., E.T., M.B., S.M., C.K., J.J., A.N., S.E., J.G., V.A., R.D., X.Y., and M.V.M. All authors have read and agreed to the published version of the manuscript.
Funding
Open access funding provided by FEDER European Funds and the Junta de Castilla y León under the Research and Innovation Strategy for Smart Specialization (RIS3) of Castilla y León 2021-2027. Both the ADMYRE study and the ECA analysis were funded by PharmaMar S.A.
Data availability
The original contributions presented in the analysis have been included in the article; further inquiries can be directly addressed to the corresponding author.
Declarations
Competing interests
H.L received research support by AMGEN and Sanofi; and was speaker or participated in advisory boards for Janssen, BMS, Pfizer, Oncopetides, Menarini/Stemline, and GSK. E.T. was in advisory boards for Amgen, AstraZeneca, EUSA Pharma, BMS, GSK, J&J, Menarini/Stemline, Pfizer, Sanofi, and Takeda; received honoraria from Amgen, Antengene, AstraZeneca, EUSA Pharma, BMS, Forus, GSK, J&J, Menarini/Stemline, Novartis, Pfizer, Sanofi, Swixx, and Takeda; received research funding from Amgen, GSK, J&J, Sanofi, and Takeda; and received travel expenses from Takeda. M.B. received honoraria from Sanofi, Celgene, Amgen, Janssen, Novartis, Bristol Myers Squibb, and AbbVie; served on advisory boards for Janssen and GlaxoSmithKline; and received research funding from Sanofi, Celgene, Amgen, Janssen, Novartis, Bristol Myers Squibb, and Mundipharma. S.M., C.K., J.J., A.N., S.E., J.G. and V.A. are employees of PharmaMar S.A. C.K., J.J., A.N., J.G. and V.A. are stock ownerships in PharmaMar S.A. R.D. and X.Y. are employees and stock ownerships of Medidata, a Dassault Systèmes Company. M.V.M. received honoraria derived from lectures and participation in advisory boards from Johnson & Johnson, BMS, Amgen, GSK, AbbVie, Pfizer, Regeneron, Roche, Sanofi, Oncopeptides, and Kite.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
1. Mitsiades CS, Ocio EM, Pandiella A, Maiso P, Gajate C, Garayoa M, Vilanova D, Montero JC, Mitsiades N, McMullan CJ, Munshi NC, Hideshima T, Chauhan D, Aviles P, Otero G, Faircloth G, Mateos MV, Richardson PG, Mollinedo F, San-Miguel JF, Anderson KC (2008) Aplidin, a marine organism-derived compound with potent antimyeloma activity in vitro and in vivo. Cancer Res 68(13):5216–5225
2. Losada A, Munoz-Alonso MJ, Garcia C, Sanchez-Murcia PA, Martinez-Leal JF, Dominguez JM, Lillo MP, Gago F, Galmarini CM (2016) Translation elongation factor eEF1A2 is a novel anticancer target for the marine natural product plitidepsin. Sci Rep 6:35100. 10.1038/srep35100
3. Spicka I, Ocio EM, Oakervee HE, Greil R, Banh RH, Huang SY, D’Rozario JM, Dimopoulos MA, Martinez S, Extremera S, Kahatt C, Alfaro V, Carella AM, Meuleman N, Hajek R, Symeonidis A, Min CK, Cannell P, Ludwig H, Sonneveld P, Mateos MV (2019) Randomized phase III study (ADMYRE) of plitidepsin in combination with dexamethasone vs. dexamethasone alone in patients with relapsed/refractory multiple myeloma. Ann Hematol 98(9):2139–2150. 10.1007/s00277-019-03739-2
4. Rajkumar SV, Jacobus S, Callander NS, Fonseca R, Vesole DH, Williams ME, Abonour R, Siegel DS, Katz M, Greipp PR, Eastern Cooperative Oncology G (2010) Lenalidomide plus high-dose dexamethasone versus lenalidomide plus low-dose dexamethasone as initial therapy for newly diagnosed multiple myeloma: an open-label randomised controlled trial. Lancet Oncol 11(1):29–37. 10.1016/S1470-2045(09)70284-0
5. Rosenbaum PR, Rubin DB (1983) The central role of the propensity score in observational studies for causal effect. Biometrika 70(1):41–55
6. Lim J, Walley R, Yuan J, Liu J, Dabral A, Best N, Grieve A, Hampson L, Wolfram J, Woodward P, Yong F, Zhang X, Bowen E (2018) Minimizing patient burden through the use of historical subject-level data in innovative confirmatory clinical trials: review of methods and opportunities. Ther Innov Regul Sci 52(5):546–559. 10.1177/2168479018778282
7. Ghadessi M, Tang R, Zhou J, Liu R, Wang C, Toyoizumi K, Mei C, Zhang L, Deng CQ, Beckman RA (2020) A roadmap to using historical controls in clinical trials - by drug information association adaptive design scientific working group (DIA-ADSWG). Orphanet J Rare Dis 15(1):69. 10.1186/s13023-020-1332-x
8. Bartlett VL, Dhruva SS, Shah ND, Ryan P, Ross JS (2019) Feasibility of using real-world data to replicate clinical trial evidence. JAMA Netw Open 2(10):e1912869. 10.1001/jamanetworkopen.2019.12869
9. Miksad RA, Abernethy AP (2018) Harnessing the power of real-world evidence (RWE): a checklist to ensure regulatory-grade data quality. Clin Pharmacol Ther 103(2):202–205. 10.1002/cpt.946
10. Farah E, Kenney M, Warkentin MT, Cheung WY, Brenner DR (2024) Examining external control arms in oncology: a scoping review of applications to date. Cancer Med 13(13):e7447. 10.1002/cam4.7447
11. Mateos MV, Chari A, Usmani SZ, Goldschmidt H, Weisel K, Qi K, Londhe A, Nair S, Lin X, Pei L, Ammann E, Kobos R, Smit J, Parekh T, Marshall A, Slavcev M, Moreau P (2023) Comparative efficacy of Teclistamab versus physician’s choice of therapy in the long-term follow-up of APOLLO, POLLUX, CASTOR, and EQUULEUS clinical trials in patients with triple-class exposed relapsed or refractory multiple myeloma. Clin Lymphoma Myeloma Leuk 23(5):385–393. 10.1016/j.clml.2023.02.006
12. Weisel K, Martin T, Krishnan A, Jagannath S, Londhe A, Nair S, Diels J, Vogel M, Schecter JM, Banerjee A, Berdeja JG, Nesheiwat T, Garrett A, Qi K, Valluri S, Usmani SZ, Yong K (2022) Comparative efficacy of ciltacabtagene autoleucel in CARTITUDE-1 vs physician’s choice of therapy in the long-term follow-up of POLLUX, CASTOR, and EQUULEUS clinical trials for the treatment of patients with relapsed or refractory multiple myeloma. Clin Drug Investig 42(1):29–41. 10.1007/s40261-021-01100-y
13. Einsele H, Moreau P, Bahlis N, Bhutani M, Vincent L, Karlin L, Perrot A, Goldschmidt H, van de Donk N, Ocio EM, Martinez-Lopez J, Rodriguez-Otero P, Dytfeld D, Diels J, Strulev V, Haddad I, Renaud T, Ammann E, Cabrieto J, Perualila N, Gan R, Zhang Y, Parekh T, Albrecht C, Weisel K, Mateos MV (2024) Comparative efficacy of Talquetamab vs. current treatments in the LocoMMotion and MoMMent studies in patients with triple-class-exposed relapsed/refractory multiple myeloma. Adv Ther 41(4):1576–1593. 10.1007/s12325-024-02797-x
14. Ye JC, Biran N, Nair S, Lin X, Qi K, Londhe A, Ammann E, Renaud T, Kane C, Parekh T, Gray K, Peterson S, Costa LJ (2025) Talquetamab versus real-world physician’s choice treatment: comparative effectiveness in patients with triple-class exposed relapsed/refractory multiple myeloma. Clin Lymphoma Myeloma Leuk 25(2):124–134 e125. 10.1016/j.clml.2024.08.003
15. San Miguel J, Weisel K, Moreau P, Lacy M, Song K, Delforge M, Karlin L, Goldschmidt H, Banos A, Oriol A, Alegre A, Chen C, Cavo M, Garderet L, Ivanova V, Martinez-Lopez J, Belch A, Palumbo A, Schey S, Sonneveld P, Yu X, Sternas L, Jacques C, Zaki M, Dimopoulos M (2013) Pomalidomide plus low-dose dexamethasone versus high-dose dexamethasone alone for patients with relapsed and refractory multiple myeloma (MM-003): a randomised, open-label, phase 3 trial. Lancet Oncol 14(11):1055–1066. 10.1016/S1470-2045(13)70380-2
16. Austin PC (2013) The performance of different propensity score methods for estimating marginal hazard ratios. Stat Med 32(16):2837–2849. 10.1002/sim.5705
17. Austin PC (2011) An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivar Behav Res 46(3):399–424. 10.1080/00273171.2011.568786
18. Stuart EA, Rubin DB (2007) Best practices in quasi-experimental designs: matching methods for causal inferences. In: Osborne JW (ed) Best practices in quantitative methods. Sage Publications, Thousand Oaks
19. Normand ST, Landrum MB, Guadagnoli E, Ayanian JZ, Ryan TJ, Cleary PD, McNeil BJ (2001) Validating recommendations for coronary angiography following acute myocardial infarction in the elderly: a matched analysis using propensity scores. J Clin Epidemiol 54(4):387–398. 10.1016/s0895-4356(00)00321-8
20. Gyawali B, Tessema FA, Jung EH, Kesselheim AS (2019) Assessing the justification, funding, success, and survival outcomes of randomized noninferiority trials of cancer drugs: a systematic review and pooled analysis. JAMA Netw Open 2(8):e199570. 10.1001/jamanetworkopen.2019.9570
21. Hashim M, Vincken T, Kroi F, Gebregergish S, Spencer M, Wang J, Kampfenkel T, Lam A, He J (2021) A systematic review of noninferiority margins in oncology clinical trials. J Comp Eff Res 10(6):443–455. 10.2217/cer-2020-0200
22. Dimopoulos M, Spencer A, Attal M, Prince HM, Harousseau JL, Dmoszynska A, San Miguel J, Hellmann A, Facon T, Foa R, Corso A, Masliak Z, Olesnyckyj M, Yu Z, Patin J, Zeldis JB, Knight RD, Multiple Myeloma Study I (2007) Lenalidomide plus dexamethasone for relapsed or refractory multiple myeloma. N Engl J Med 357(21):2123–2132. 10.1056/NEJMoa070594
23. Richardson PG, Sonneveld P, Schuster MW, Irwin D, Stadtmauer EA, Facon T, Harousseau JL, Ben-Yehuda D, Lonial S, Goldschmidt H, Reece D, San-Miguel JF, Blade J, Boccadoro M, Cavenagh J, Dalton WS, Boral AL, Esseltine DL, Porter JB, Schenkein D, Anderson KC (2005) Bortezomib or high-dose dexamethasone for relapsed multiple myeloma. N Engl J Med 352(24):2487–2498
24. Lonial S, Weiss BM, Usmani SZ, Singhal S, Chari A, Bahlis NJ, Belch A, Krishnan A, Vescio RA, Mateos MV, Mazumder A, Orlowski RZ, Sutherland HJ, Blade J, Scott EC, Oriol A, Berdeja J, Gharibo M, Stevens DA, LeBlanc R, Sebag M, Callander N, Jakubowiak A, White D, de la Rubia J, Richardson PG, Lisby S, Feng H, Uhlar CM, Khan I, Ahmadi T, Voorhees PM (2016) Daratumumab monotherapy in patients with treatment-refractory multiple myeloma (SIRIUS): an open-label, randomised, phase 2 trial. Lancet 387(10027):1551–1560. 10.1016/S0140-6736(15)01120-4
25. Siegel DS, Martin T, Wang M, Vij R, Jakubowiak AJ, Lonial S, Trudel S, Kukreti V, Bahlis N, Alsina M, Chanan-Khan A, Buadi F, Reu FJ, Somlo G, Zonder J, Song K, Stewart AK, Stadtmauer E, Kunkel L, Wear S, Wong AF, Orlowski RZ, Jagannath S (2012) A phase 2 study of single-agent carfilzomib (PX-171-003-A1) in patients with relapsed and refractory multiple myeloma. Blood 120(14):2817–2825. 10.1182/blood-2012-05-425934
26. Richardson PG, Schlossman RL, Alsina M, Weber DM, Coutre SE, Gasparetto C, Mukhopadhyay S, Ondovik MS, Khan M, Paley CS, Lonial S (2013) PANORAMA 2: Panobinostat in combination with bortezomib and dexamethasone in patients with relapsed and bortezomib-refractory myeloma. Blood 122(14):2331–2337. 10.1182/blood-2013-01-481325
27. Richardson PG, Hungria VTM, Yoon S-S, Beksac M, Dimopoulos MA, Elghandour A, Jedrzejczak WW, Guenther A, Nakorn TN, Siritanaratkul N, Schlossman RL, Hou J, Moreau P, Lonial S, Lee JH, Einsele H, Sopala M, Bengoudifa B-R, Corrado C, Binlich F, San-Miguel JF (2016) Panobinostat plus bortezomib and dexamethasone in previously treated multiple myeloma: outcomes by prior treatment. Blood 127(6):713–721. 10.1182/blood-2015-09-665018
28. Leisch M, Egle A, Greil R (2019) Plitidepsin: a potential new treatment for relapsed/refractory multiple myeloma. Future Oncol 15(2):109–120. 10.2217/fon-2018-0492
Associated Data
Supplementary Materials
Supplementary Material 1 (DOCX 168 KB)
Data Availability Statement
The original contributions presented in the analysis have been included in the article; further inquiries can be directly addressed to the corresponding author.