This page separates reported trial results from statistical interpretation. Numerical results and trial characteristics are restricted to the ClinicalTrials.gov record and the linked publication record identified in that data.
1. Trial at a Glance
DREAMM-3 was a randomized, parallel, phase 3 trial comparing single-agent belantamab mafodotin with pomalidomide plus low-dose dexamethasone in participants with relapsed/refractory multiple myeloma. The registered primary endpoint was progression-free survival based on investigator-assessed response according to International Myeloma Working Group criteria.
| Feature | DREAMM-3 |
|---|---|
| Trial name | DREAMM-3 |
| Phase | Phase 3 |
| Condition | Multiple Myeloma |
| Population | Participants with relapsed/refractory multiple myeloma |
| Design | Randomized, parallel, unmasked |
| Allocation | Randomized |
| Primary purpose | Treatment |
| Enrollment | 325 |
| Primary endpoint | Progression-Free Survival (PFS) Based on Investigator-Assessed Response as Per International Myeloma Working Group (IMWG) |
| Primary endpoint time frame | Up to 27 months |
| Trial status | Active, not recruiting |
| Start | 2020-04-02 |
| Primary completion | 2022-09-12 |
| Lead sponsor | GlaxoSmithKline |
| Sponsor type | Industry |
| ClinicalTrials.gov | NCT04162210 |
2. Clinical Question
The primary statistical question was whether time from randomization to progression or death differed between participants assigned to single-agent belantamab mafodotin and those assigned to pomalidomide plus low-dose dexamethasone.
Population
Participants with relapsed/refractory multiple myeloma.
Intervention
Belantamab mafodotin.
Comparator
Pomalidomide plus low-dose dexamethasone (Pom/dex).
Primary question
How does progression-free survival compare between the two randomized treatment groups?
The registry identifies the primary endpoint as a time-to-event outcome and reports the primary comparison using a log-rank test with a hazard ratio as the effect measure. The efficacy analysis population was the intention-to-treat population, defined as all randomized participants whether or not randomized treatment was administered.
3. Trial Design
Belantamab mafodotin
- Single-agent belantamab mafodotin.
- Primary comparison against Pom/dex.
- Primary endpoint analyzed using the ITT population.
Pomalidomide plus dexamethasone
- Pomalidomide plus low-dose dexamethasone.
- Comparator treatment group for the randomized analysis.
- Primary endpoint analyzed using the ITT population.
4. Randomization and Analysis Population
The primary efficacy analysis used the intention-to-treat (ITT) population. The registry defines this population as including all randomized participants whether or not randomized treatment was administered.
This distinction is important for causal interpretation. Once randomization occurs, retaining participants in their assigned groups preserves the treatment comparison created by the randomization process. Treatment discontinuation or failure to receive randomized treatment does not automatically remove a participant from the primary efficacy population.
| Population / concept | Registry information |
|---|---|
| Primary efficacy population | Intent-to-treat (ITT) |
| ITT definition | All randomized participants whether or not randomized treatment was administered |
| Primary comparison | Belantamab Mafodotin vs Pomalidomide Plus Dexamethasone |
| Secondary OS population | ITT Population |
5. Endpoints
Primary Endpoint: Progression-Free Survival
| Element | Registered information |
|---|---|
| Endpoint | Progression-Free Survival (PFS) Based on Investigator-Assessed Response as Per International Myeloma Working Group (IMWG) |
| Time frame | Up to 27 months |
| Endpoint type | Time-to-event |
| Definition | PFS is time from randomization until earliest date of progressive disease (PD), or death due to any cause per investigator-assessed response per IMWG. |
| Progressive disease definition | PD is ≥25% increase from nadir in any of the following: serum M-protein (absolute increase ≥0.5 gram per deciliter [g/dL]), urine M-protein (absolute increase ≥200 mg/24hr), or difference between involved/uninvolved FLC levels (absolute increase >10 mg/dL). |
The endpoint therefore combines two possible event pathways: documented progressive disease or death from any cause. Participants who have not experienced either event by their last evaluable observation contribute censored follow-up to the time-to-event analysis.
Secondary Endpoint: Overall Survival
| Element | Registered analysis information |
|---|---|
| Endpoint | Overall Survival (OS) |
| Time frame | Up to approximately 263 weeks |
| Endpoint type | Time-to-event |
| Analysis population | ITT Population |
| Comparison | Belantamab Mafodotin vs Pomalidomide Plus Dexamethasone |
PFS and OS are related but distinct endpoints. PFS counts progression or death as the event, whereas OS uses death as the event. Consequently, a treatment comparison can produce different hazard-ratio estimates for the two endpoints even though both are analyzed as time-to-event outcomes.
6. Statistical Methodology
Log-rank test
The registry reports the log-rank test as the statistical method for both the primary PFS analysis and the secondary OS analysis. The log-rank test compares the observed pattern of event occurrence over follow-up between randomized groups rather than reducing the data to a single time point.
The log-rank framework is designed for censored time-to-event data and therefore uses information contributed by participants throughout follow-up.
Hazard ratio
The effect measure reported for both analyses was the hazard ratio (HR). A hazard ratio compares the estimated instantaneous event rates between treatment groups within the statistical framework used for the analysis.
An HR below 1 indicates a lower estimated event hazard in the numerator group; an HR above 1 indicates a higher estimated event hazard. The HR is not itself an absolute probability or a difference in median survival.
Stratified analysis and covariate adjustment
The primary and secondary analyses are described as incorporating covariate adjustment and stratification. The registry notes that the one-sided p-value from the stratified log-rank test was adjusted for previous treatment with anti-CD38, ISS staging, and number of prior lines of therapy.
| Adjustment / stratification factor | Role identified in the registry analysis |
|---|---|
| Previous treatment with anti-CD38 | Adjustment / stratification factor |
| ISS staging | Adjustment / stratification factor |
| Number of prior lines of therapy | Adjustment / stratification factor |
One-sided testing with a two-sided confidence interval
The registry analysis notes specify a one-sided p-value from the stratified log-rank test, while the reported hazard-ratio confidence intervals are explicitly 95% two-sided intervals. These quantities should be read according to their stated constructions rather than treating the p-value and confidence interval as if they were necessarily generated from identical tail conventions.
7. Results: Primary Progression-Free Survival
The primary endpoint analysis compared investigator-assessed progression-free survival between belantamab mafodotin and pomalidomide plus dexamethasone over the registered time frame of up to 27 months.
Hazard ratio for progression or death
95% CI: 0.72–1.47 · P = 0.558
Analysis: stratified log-rank test · Analysis population: ITT
| Primary endpoint | Belantamab mafodotin vs Pom/dex |
|---|---|
| Progression-free survival | HR 1.03 |
| 95% confidence interval | 0.72–1.47 |
| Confidence interval | 95%, two-sided |
| P-value | 0.558 |
| Statistical method | Log-rank test |
| Analysis population | Intent-to-treat |
| Adjustment | Previous treatment with anti-CD38, ISS staging and number of prior lines of therapy |
The estimated hazard ratio of 1.03 means that the estimated instantaneous rate of progression or death was approximately 3% higher in the belantamab mafodotin group than in the Pom/dex group within the fitted comparison. Because the estimate is close to 1, the estimated relative difference in event hazard is small.
The HR does not mean that 3% more participants progressed, nor does it mean that the treatment caused a 3% increase in an individual's probability of progression or death. A hazard ratio is a relative time-to-event measure, not an absolute risk difference.
The 95% CI of 0.72–1.47 is relatively broad around the point estimate. It includes values below 1 and above 1, so the interval is compatible with a range of possible relative event hazards under the statistical model and sampling framework. It should not be interpreted as a range containing the individual treatment effects experienced by patients.
The P = 0.558 value describes the evidence against the statistical null within the stated testing framework; it does not measure the size or clinical importance of the observed effect. A p-value of 0.558 is not a probability that the null hypothesis is true, and it does not establish equivalence between treatments.
Because this is a time-to-event analysis using a hazard ratio, interpretation also depends on the behavior of hazards over follow-up. A single HR can be an incomplete summary if the relative hazards vary substantially with time. The ClinicalTrials.gov record does not provide enough information to independently assess that assumption.
8. Results: Overall Survival
Overall survival was reported as a secondary time-to-event endpoint with a time frame of up to approximately 263 weeks. The analysis used the ITT population and the same broad stratified log-rank framework described for the primary endpoint.
Hazard ratio for overall survival
95% CI: 0.73–1.30 · P = 0.440
Analysis: stratified log-rank test · Analysis population: ITT
| Secondary endpoint | Belantamab mafodotin vs Pom/dex |
|---|---|
| Overall survival | HR 0.98 |
| 95% confidence interval | 0.73–1.30 |
| Confidence interval | 95%, two-sided |
| P-value | 0.440 |
| Statistical method | Log-rank test |
| Analysis population | ITT Population |
| Adjustment | Previous treatment with anti-CD38, ISS staging and number of prior lines of therapy |
The estimated OS hazard ratio of 0.98 is very close to 1. Within the model-based framework, this corresponds to an estimated instantaneous rate of death that is approximately 2% lower in the belantamab mafodotin group than in the Pom/dex group.
That does not mean that mortality was reduced by exactly 2% for every participant, nor does it represent an absolute survival difference. The HR summarizes relative event rates over the analyzed follow-up.
The 95% CI of 0.73–1.30 spans both below and above 1. The interval therefore reflects uncertainty that encompasses both a lower and a higher estimated death hazard for belantamab mafodotin relative to the comparator.
The P = 0.440 value is evidence from the specified hypothesis-testing framework, not a measure of treatment effect size. A larger p-value does not quantify how clinically important an observed difference is, and a p-value alone cannot establish that two treatments are equivalent.
As with PFS, the interpretation of an HR requires attention to censoring and the time-varying behavior of hazards. The ClinicalTrials.gov record does not report enough additional information to independently reconstruct the survival curves or evaluate proportional-hazards behavior.
9. Comparing the Two Time-to-Event Results
| Endpoint | HR | 95% CI | P-value | Method |
|---|---|---|---|---|
| Primary PFS | 1.03 | 0.72–1.47 | 0.558 | Stratified log-rank |
| Secondary OS | 0.98 | 0.73–1.30 | 0.440 | Stratified log-rank |
The two estimates are close to the null value of 1, but they describe different event processes. The PFS analysis counts either investigator-assessed progression or death, whereas OS counts death. It is therefore not statistically contradictory for the two hazard ratios to differ.
The confidence intervals also illustrate why point estimates should not be interpreted in isolation. The PFS interval extends from 0.72 to 1.47, while the OS interval extends from 0.73 to 1.30. Both intervals contain 1, and both indicate uncertainty around the respective point estimates.
10. Secondary Endpoint Results
The ClinicalTrials.gov record contains one posted secondary statistical analysis: overall survival. No additional secondary endpoint statistical analyses are included in the posted results.
| Endpoint | Role | HR | 95% CI | P-value |
|---|---|---|---|---|
| Overall Survival (OS) | Secondary | 0.98 | 0.73–1.30 | 0.440 |
The registry reports 25 outcome measures overall, but the ClinicalTrials.gov record contains only the formal primary PFS analysis and the secondary OS analysis. The page therefore does not infer additional numerical efficacy results from the existence of other registered outcome measures.
11. Safety
The ClinicalTrials.gov record reports serious adverse events by treatment arm as affected participants divided by participants at risk. These figures should be kept separate from the efficacy analyses because the safety denominator and analysis context can differ from the ITT efficacy population.
| Safety measure | Belantamab Mafodotin | Pomalidomide Plus Dexamethasone |
|---|---|---|
| Serious adverse events | 101 / 217 affected / at risk | 45 / 102 affected / at risk |
The ClinicalTrials.gov record does not provide a statistical comparison, confidence interval, or p-value for serious adverse events. Accordingly, the figures are reported descriptively rather than treated as evidence of a formal between-group hypothesis test.
12. Statistical Methods Explained
Why was a log-rank test used?
PFS and OS are time-to-event endpoints, so participants can contribute different amounts of follow-up and can be censored before an event occurs. The log-rank test is designed to compare the event experience of randomized groups across follow-up while incorporating this censored time-to-event structure.
What does a hazard ratio of 1.03 mean?
An HR of 1.03 indicates an estimated instantaneous event hazard approximately 1.03 times that of the comparator under the analysis framework. Put differently, the estimated relative hazard is about 3% higher. It does not represent a 3% absolute increase in the probability of progression or death.
What does an HR of 0.98 mean?
An HR of 0.98 corresponds to an estimated instantaneous death hazard approximately 0.98 times that of the comparator. This is an estimated relative difference of about 2% lower in the belantamab mafodotin group. Again, this is not an absolute survival difference.
Why does the confidence interval matter?
A point estimate summarizes the observed comparison, but the confidence interval describes uncertainty around that estimate. For PFS, the 95% CI is 0.72–1.47; for OS, it is 0.73–1.30. The widths of these intervals show that the point estimates should not be treated as exact measurements.
Why doesn't the p-value measure effect size?
The p-value measures how compatible the observed data are with a specified null hypothesis under the stated statistical framework. It is affected by the effect estimate, variability, sample information, censoring, and other features of the analysis. It therefore cannot be read as a direct measure of clinical magnitude.
Why does ITT analysis matter?
The ITT principle keeps randomized participants in their assigned groups for efficacy analysis. This preserves the comparison generated by randomization and avoids redefining treatment groups based on events that occur after randomization.
Why was stratification used?
The registry analysis notes identify previous treatment with anti-CD38, ISS staging, and number of prior lines of therapy as adjustment factors. Stratified analysis allows the comparison to account for these prespecified factors rather than treating the randomized population as completely unstructured with respect to them.
13. Interpreting the Hazard Ratio Carefully
The PFS HR of 1.03 is a relative time-to-event estimate comparing the instantaneous rate of progression or death between the two randomized treatment groups under the reported analysis.
It does not mean that 3% of participants had progression, that 3% more participants progressed, or that each participant experienced a 3% increase in individual risk. Those would require different measures.
The OS HR of 0.98 indicates an estimated death hazard approximately 0.98 times that of the comparator in the reported analysis.
Neither HR directly gives the probability that an individual participant will survive to a specified time, the absolute number of additional months lived, or the proportion of patients who benefit. Those quantities require absolute survival estimates, median event times, or other clinically interpretable summaries that are not contained in the ClinicalTrials.gov record.
14. Confidence Intervals and Statistical Uncertainty
The PFS and OS confidence intervals provide an important complement to their point estimates.
| Endpoint | Point estimate | 95% two-sided CI | Includes HR = 1? |
|---|---|---|---|
| PFS | 1.03 | 0.72–1.47 | Yes |
| OS | 0.98 | 0.73–1.30 | Yes |
For a hazard ratio, 1 represents equal estimated hazard between the groups. Because both confidence intervals include 1, neither interval excludes the null hazard ratio at the corresponding 95% confidence level.
The confidence interval should not be interpreted as saying that there is a 95% probability that the true hazard ratio lies inside this particular interval. Rather, it is an interval produced by a procedure with a stated long-run coverage property under its statistical assumptions.
15. One-Sided P-Values and Two-Sided Confidence Intervals
The registry analysis notes explicitly state that the p-value was a one-sided p-value from a stratified log-rank test, adjusted for previous treatment with anti-CD38, ISS staging, and number of prior lines of therapy. The reported confidence intervals for the hazard ratios are specified as 95% two-sided.
This distinction matters because one-sided and two-sided procedures allocate their reference probability differently. A reader should therefore preserve the registry's reported test convention rather than silently converting the p-value into a different testing framework.
16. Censoring and Time-to-Event Data
PFS and OS require methods that can accommodate incomplete follow-up. If a participant has not experienced the relevant event by the last observation, the participant's follow-up is censored rather than treated as if the event had occurred at that time.
This is one reason a time-to-event analysis cannot generally be replaced by a simple comparison of the proportions of participants who experienced events. Participants contribute information for the amount of time they remain under observation, and the statistical analysis uses the timing of events and censoring.
For PFS, the event is progressive disease or death. For OS, the event is death. The registry's primary PFS definition begins the clock at randomization.
17. Trial Timeline
Trial start
The registered DREAMM-3 study began on April 2, 2020.
Primary completion
The registry lists September 12, 2022 as the primary completion date.
Active, not recruiting
The ClinicalTrials.gov record identifies the trial status as ACTIVE_NOT_RECRUITING.
18. What the Results Do and Do Not Establish
What is directly reported
The registry reports a PFS HR of 1.03 with a 95% CI of 0.72–1.47 and P = 0.558, and an OS HR of 0.98 with a 95% CI of 0.73–1.30 and P = 0.440.
What is not directly reported here
The ClinicalTrials.gov record does not provide median PFS, median OS, Kaplan-Meier survival probabilities, subgroup estimates, or additional efficacy estimates.
What the p-values do not establish
The p-values do not establish that the treatments are equivalent, nor do they quantify the magnitude of any treatment difference.
What the confidence intervals add
The confidence intervals show the statistical uncertainty surrounding the reported hazard-ratio estimates and include the null value of 1 for both endpoints.
19. Multiplicity, Interim Analysis, and Other Design Features
The ClinicalTrials.gov record does not report a non-inferiority margin, factorial design, crossover scheme, formal multiplicity hierarchy, or interim-analysis alpha-spending procedure. Those design features are therefore not inferred from the reported PFS and OS analyses.
| Design topic | Information supported by the ClinicalTrials.gov record |
|---|---|
| Non-inferiority margin | Not reported in the ClinicalTrials.gov record |
| Crossover | Not reported in the ClinicalTrials.gov record |
| Factorial design | Not reported; the registered design model is parallel |
| Multiplicity hierarchy | Not reported in the ClinicalTrials.gov record |
| Interim analysis | Not reported in the ClinicalTrials.gov record |
| Missing-data / imputation method | Not reported in the ClinicalTrials.gov record |
| Bayesian methods | Not reported in the ClinicalTrials.gov record |
20. Important Limitations and Interpretation Issues
- Point estimates are not exact quantities: both reported hazard ratios are accompanied by confidence intervals that show substantial uncertainty around the estimates.
- Confidence intervals include 1: the 95% two-sided intervals for both PFS and OS include the null hazard ratio.
- PFS is composite: the primary endpoint treats progressive disease and death as alternative event pathways, so it does not represent progression alone.
- Hazard-ratio interpretation: an HR is a relative time-to-event measure and should not be translated directly into an absolute risk difference or individual treatment benefit.
- Proportional-hazards caution: the registry reports hazard ratios, but the ClinicalTrials.gov record does not provide sufficient information to independently evaluate whether relative hazards remain approximately proportional over follow-up.
- Analysis-population distinction: efficacy analyses use the ITT population, while the ClinicalTrials.gov record is expressed using affected and at-risk counts by arm.
- Limited posted statistical analyses: although 25 outcome measures are reported as posted, the registry-reported formal statistical-analysis records contain one primary endpoint analysis and one secondary endpoint analysis.
- No non-inferiority conclusion: the ClinicalTrials.gov record does not identify a non-inferiority margin, so the reported PFS and OS results should not be reframed as a non-inferiority analysis.
21. Why This Trial Matters Statistically
DREAMM-3 is a useful teaching case because its registry results illustrate how a randomized phase 3 comparison can be summarized through time-to-event methods without reducing the analysis to a single p-value.
| Concept | How it appears in DREAMM-3 |
|---|---|
| Randomization | Participants were assigned using a randomized parallel design. |
| ITT analysis | The primary PFS analysis included all randomized participants whether or not randomized treatment was administered. |
| Time-to-event endpoint | PFS was defined from randomization to progression or death. |
| Log-rank test | Used for the reported PFS and OS comparisons. |
| Hazard ratio | Used as the effect measure for PFS and OS. |
| Confidence interval | 95% two-sided intervals quantify uncertainty around both HR estimates. |
| Stratified analysis | Analyses were adjusted for previous treatment with anti-CD38, ISS staging and number of prior lines of therapy. |
| One-sided testing | The analysis notes identify a one-sided p-value from the stratified log-rank test. |
| Composite endpoint | PFS combines progressive disease and death as events. |
| Safety analysis | Serious adverse events are reported separately by treatment arm. |
22. Statistical Concepts in This Trial
Learn more about the methods used in this trial:
23. Related Statistical Calculators
Use related tools to explore the statistical concepts underlying this trial:
24. Sources
- ClinicalTrials.gov: DREAMM-3, NCT04162210.
- Linked publication: PubMed record for PMID 37793771: 37793771.
Continue through the Clinical Biostats statistical pathway
Connect this trial's time-to-event endpoints and statistical methods to deeper tutorials, calculators, and clinical-trial analysis workflows.
25. Record Summary
DREAMM-3 provides a focused example of randomized time-to-event analysis. The trial used a randomized parallel phase 3 design with 325 participants enrolled and compared belantamab mafodotin with pomalidomide plus low-dose dexamethasone. Its primary endpoint was investigator-assessed progression-free survival according to IMWG criteria, analyzed in the ITT population using a stratified log-rank test and hazard ratio.
The reported primary PFS analysis produced an HR of 1.03 with a 95% two-sided CI of 0.72–1.47 and P = 0.558. The reported secondary OS analysis produced an HR of 0.98 with a 95% two-sided CI of 0.73–1.30 and P = 0.440. Both intervals include the null hazard ratio of 1, emphasizing why the point estimate, confidence interval, p-value, analysis population, and endpoint definition should be interpreted together.