A Study of Melphalan Flufenamide (Melflufen)-Dex or Pomalidomide-dex for RRMM Patients Refractory to Lenalidomide: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of Melphalan Flufenamide (Melflufen)-Dex or Pomalidomide-dex for RRMM Patients Refractory to Lenalidomide (NCT03151811), a randomized phase 3 study. The primary outcome is Progression Free Survival (PFS).

2. Clinical Question

3. Trial Design

This was a randomized, controlled, open-label, parallel-group Phase 3 study sponsored by Oncopeptides AB. The registry describes it as a multicenter trial and records an enrollment of 495. Patients were assigned to melflufen plus dexamethasone or pomalidomide plus dexamethasone. The record also lists single masking, while describing the study as open-label; it does not specify which individuals, if any, were masked.

Eligibility focused on relapsed or refractory multiple myeloma after 2-4 prior lines of therapy, with lenalidomide refractoriness in the last line. The registry's definition required progression on lenalidomide or within 60 days of the last dose. The two groups followed different drug schedules, while dexamethasone was scheduled on the same cycle days in both arms. The registry states that treatment was to continue until confirmed progression, unacceptable toxicity, or a decision by the patient or investigator that continued treatment was not in the patient's best interest.

The registry lists the study start as 2017-06-12, primary completion as 2021-02-03, and status as terminated. The reason for termination is not reported in the registry record. The primary endpoint used the Full Analysis Set as described in the registry, with all randomized patients included in the primary outcome analysis. A fuller account of the randomization procedure, allocation concealment, and the operational details of masking is not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: 246 patients were randomized to treatment Arm A, 249 patients to treatment Arm B (FAS). Of the 495 randomized patients, 228 patients in Arm A received treatment and 246 patients in Arm B received treatment (SAF). The protocol had pre-defined treatment criteria for specific laboratory values, and if the patient's values deteriorated between screening and C1D1, treatment could not be initiated. In total, 664 patients were screened, 495 randomized, 474 treated, 21 randomized but not treated.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized patients (Full Analysis Set as per protocol) were included into the primary outcome measure analysis.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: All patients that were enrolled defined as patients assigned a unique patient number by Interactive Response Technology system at the time of enrollment.

6. Primary Endpoint

The registry lists the primary outcome as Progression Free Survival (PFS), measured over From date of randomization until first evidence of confirmed disease progression or death due to any cause (whichever occurred first), up to the data cutoff date of 03 Feb 2021 (ie, assessed up to approximately 43 months)..

Progression-free survival measures the time from randomization until the first of confirmed disease progression or death from any cause. In this trial, progression was assessed by an Independent Review Committee according to International Myeloma Working Group Uniform Response Criteria. The endpoint therefore combines disease progression and death, rather than treating either outcome in isolation.

Time-to-event endpoints are used when the timing of an event is relevant and participants may have different lengths of observation. A median gives a point on the time scale by which half of the observed event-time distribution has occurred, subject to the analysis method and available follow-up. It does not describe every participant's course, and it is not an estimate of the treatment difference by itself.

The registry reports median progression-free survival in months with a 95% confidence interval for each arm. It also reports a log-rank test for the comparison. The confidence intervals express uncertainty around each arm's median; they are not a confidence interval for a between-arm difference. A direct estimate such as a hazard ratio and its confidence interval is not reported in the posted results. The timeframe is from randomization through the data cutoff of 03 Feb 2021, described as assessed up to approximately 43 months.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical ideas help distinguish the reported arm summaries from the comparison between groups. The primary endpoint is a time-to-event outcome: the event is confirmed progression or death, whichever occurs first, and the time origin is randomization. The registry specifies independent review for progression assessment and the criteria used. The analysis is described as including all randomized patients in the Full Analysis Set.

What does a median time-to-event estimate mean?

In general, a median time-to-event estimate is the point in the estimated event-time distribution at which the estimated probability of remaining event-free reaches the median threshold. It is a summary of a distribution, not a prediction for an individual. Some participants may have an event earlier and others later; some may have no recorded event by the analysis cutoff. Time-to-event methods are designed to use observed follow-up while accounting for participants whose event time is not observed before the analysis ends, provided the necessary censoring information and assumptions are handled appropriately.

The registry reports medians and arm-specific 95% confidence intervals. A confidence interval describes statistical uncertainty in an estimated quantity under the model and method used. The interval for one arm's median is not itself an interval for the difference between medians, nor does overlap or non-overlap of two arm-specific intervals replace a formal comparison. A direct comparative estimate, such as a hazard ratio with an interval, would answer a different and more explicitly comparative question. Such an estimate is not reported in the posted results.

What does the log-rank test assess?

In general, the log-rank test compares the event experience of groups over follow-up, using the timing of events and the participants still at risk at those times. It is a test of evidence for a difference in event-time distributions under its assumptions. The test's p-value does not state how large the difference is. It also does not indicate the probability that a treatment is effective or that a result will repeat in another study.

The registry reports a log-rank p-value of 0.031 and identifies the analysis as superiority. The posted summary does not include a hazard ratio, its confidence interval, or further details about the test implementation. The p-value can therefore be reported as the registry's inferential result, but it should not be used to reconstruct a comparative effect estimate from the medians. The medians alone do not contain the event and censoring information needed for that reconstruction.

Why are time-to-event comparisons different from simple recalculations?

For outcomes represented by a count or proportion, an unadjusted comparison can sometimes be recalculated from group counts and denominators. A median time-to-event endpoint is different: its analysis depends on the event-time and censoring patterns, not just the number randomized and the two medians. For that reason, no simple unadjusted comparative estimate is presented from the posted summary statistics. The registry's own log-rank analysis is the available comparative test result.

In general, a time-to-event model or test may also rely on assumptions about how event risk behaves over time. A hazard ratio, when reported, summarizes a relative event rate under a proportional-hazards interpretation; it is not the same as a difference between medians. The registry record does not report a hazard ratio for this primary comparison, so no such interpretation can be made from these results.

How should the confidence intervals and p-value be read together?

The confidence intervals describe uncertainty around the two median estimates separately. The log-rank p-value addresses a group comparison using the event-time data. These statistics answer related but not interchangeable questions: the medians describe typical event timing in each arm, while the test addresses evidence of a distributional difference. Neither the p-value nor the arm-specific intervals alone provide a complete account of effect magnitude and precision.

General statistical interpretation also requires attention to analysis population, follow-up, censoring, and missing outcome information. The registry identifies the primary analysis population but does not report the detailed procedures for missing data, censoring, or sensitivity analyses. Those omissions limit independent assessment of how robust the result is to alternative assumptions. The planning sample-size illustrations for the secondary binary endpoint are not observed power calculations and are not a substitute for the primary time-to-event analysis.

10. Analyses Reported by the Sponsor

The registry reports the sponsor's primary comparison as a log-rank test, with p = 0.031, and reports arm-specific median progression-free survival estimates with 95% confidence intervals. The posted summary does not give a hazard ratio or a confidence interval for the between-arm effect. Because the endpoint is time-to-event, the two medians and their confidence intervals are not enough to reconstruct the log-rank result or a comparative hazard ratio. No simple recalculated primary comparison is therefore presented.

The record does not flag a discrepancy between the posted primary result and a recalculation. The available primary inferential result is the registry's own analysis, rather than an unadjusted estimate rebuilt from summary statistics. The log-rank p-value should be read as a test result, not as a measure of effect size or precision. The registry does not report enough detail to independently reproduce the full analysis from the posted arm summaries.

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary endpoint and multiple secondary outcomes, but a multiplicity-control procedure or formal testing hierarchy is not reported in the registry record. When several endpoints or comparisons are tested, the chance of at least one apparently notable result can rise unless the analysis plan accounts for the number and order of tests. This is a general statistical concern, not evidence that any particular posted result is false. Without a documented approach, secondary p-values, if available, would need cautious interpretation; the record does not list reported analyses for the posted overall response outcome.

Procedures for missing outcome data, censoring, and sensitivity analyses are not reported in the registry record. These details matter for time-to-event outcomes because participants may have different amounts of follow-up, and some event times may not be observed before the data cutoff. The registry gives the primary endpoint timeframe and analysis population, but those details do not establish how missing assessments or deviations from planned follow-up were handled.

The primary analysis population is described as all randomized patients in the Full Analysis Set, as per protocol. The record does not report further detail on how this set was defined or whether additional analysis populations were used for other outcomes. Interim analyses, sample-size justification, and the statistical plan are not reported in the registry record. Consequently, the available information does not establish whether the primary analysis was adjusted for interim looks, how missingness was addressed, or how secondary testing was intended to support conclusions. These gaps constrain independent evaluation of the analysis strategy, even though the registry does report a primary endpoint and a log-rank result.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumonia (9.3%), covid-19 pneumonia (5.7%), thrombocytopenia (4.8%), atrial fibrillation (3.7%), anaemia (3.5%).

The registry reports serious adverse event information for participants at risk in each arm, alongside counts for other adverse events and deaths. For serious adverse events, the record lists 99 affected among 228 at risk in the melflufen plus dexamethasone arm and 124 among 246 at risk in the pomalidomide plus dexamethasone arm. These are counts with their stated denominators; they should not be converted into comparisons of risk without considering the registry's definitions, observation periods, and analysis methods.

The listed serious event terms include pneumonia at 9.3%, COVID-19 pneumonia at 5.7%, thrombocytopenia at 4.8%, atrial fibrillation at 3.7%, anaemia at 3.5%, and urinary tract infection at 2.4%. The denominator and further details for these term-specific percentages are not reported in the registry record. The record also lists other adverse event counts and deaths, but raw counts alone do not establish comparative safety, severity, duration, or causality. These figures describe reported safety observations and are not a complete assessment of treatment risks.

13. What the Result Does, and Does Not, Mean

The posted primary result describes a difference in the reported progression-free survival experience between the randomized groups, with the registry's log-rank test giving p = 0.031. The arm-specific medians and confidence intervals offer summaries of the event-time distributions and their uncertainty. They do not, on their own, establish how large or clinically important a comparative effect is. The registry does not report a hazard ratio or a confidence interval for the between-arm effect in these results.

The finding applies to the trial population described in the record: patients with relapsed or refractory multiple myeloma after 2-4 prior lines of therapy who were refractory to lenalidomide in their last line. It does not by itself establish what would be observed in populations with different prior treatment histories or eligibility characteristics. The study was randomized and controlled, but it was also described as open-label. The registry's independent review process for progression assessment is relevant context, while further operational details are not reported.

A p-value is not the probability that the null hypothesis is true, and it does not measure the chance that the observed result arose only by chance in a broad, everyday sense. It is a result of a statistical test under its assumptions. The registry's terminated status also does not explain why the study stopped; the reason is not reported in the registry record. Neither the primary finding nor the safety counts should be expanded into treatment advice or a conclusion about an individual's likely outcome. The results are best read as the reported trial comparison, with the uncertainty and reporting gaps described alongside it.

14. Why This Trial Matters Statistically

This trial illustrates why time-to-event endpoints need to be interpreted through both arm summaries and a comparative analysis. The median progression-free survival values provide an accessible description of typical event timing, while the log-rank test evaluates the group event-time distributions. The two types of result are connected, but they do not answer the same question. Treating the medians as if they were a complete estimate of the treatment contrast would overlook the event and censoring information used by the time-to-event analysis.

The registry also shows the importance of distinguishing statistical evidence from effect magnitude. A reported p-value can indicate evidence against a test's null hypothesis, but without a comparative estimate and its confidence interval, readers have less direct information about the size and precision of the relative effect. This is particularly important when a record reports arm-specific uncertainty but not uncertainty around the between-arm contrast.

Finally, the trial's primary and secondary endpoints address different outcomes, while multiplicity procedures and missing-data methods are not reported in the registry record. This makes transparent reporting of the analysis plan and endpoint hierarchy important to a careful reading. The study provides a useful example of how a posted registry result can contain informative outcome summaries and a formal test, yet still leave key details needed for complete statistical appraisal unreported.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several trial-specific reporting gaps limit interpretation. The registry does not report a hazard ratio or confidence interval for the primary between-arm effect, so the relative magnitude and precision of the time-to-event comparison cannot be assessed from the posted primary results. The arm-specific median confidence intervals do not fill that gap. Detailed information about event counts, censoring, and follow-up distribution is also not reported in the registry record.

The record describes the trial as open-label and lists single masking, but does not identify which roles were masked. Further details about allocation concealment and the randomization procedure are not reported. The record does not report the reason for the terminated status, nor does it provide the statistical plan, sample-size justification, interim analysis procedures, or missing-data and sensitivity-analysis methods. Multiplicity handling for the secondary outcomes is also not reported.

Safety summaries include counts and selected serious event terms, but the term-specific percentage denominators and fuller exposure-adjusted comparisons are not reported in the registry record. The treatment schedules and the selected eligibility criteria define the context in which the results were obtained; the findings should not be assumed to describe populations outside that context. These limitations do not negate the registry's reported results, but they restrict independent evaluation of their robustness and broader applicability.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of Melphalan Flufenamide (Melflufen)-Dex or Pomalidomide-dex for RRMM Patients Refractory to Lenalidomide — NCT03151811.
• Registry record last updated 2024-01-30; status last verified 2023-12.
• Results first posted 2022-07-27.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT03151811 is a randomized, controlled, open-label, Phase 3 parallel-group study of melflufen plus dexamethasone versus pomalidomide plus dexamethasone for patients with relapsed or refractory multiple myeloma refractory to lenalidomide. The registry records enrollment of 495 and status as terminated. The primary endpoint was progression-free survival, assessed from randomization until confirmed progression or death from any cause.

The posted primary results report median progression-free survival of 6.83 months for melflufen plus dexamethasone and 4.93 months for pomalidomide plus dexamethasone, with arm-specific 95% confidence intervals and a log-rank p-value of 0.031. A comparative hazard ratio and its confidence interval are not reported. The record also lists secondary efficacy and safety outcomes, while key details about multiplicity, missing data, and the reason for termination are not reported in the registry record.