This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.
1. Trial at a Glance
DREAMM-7 is a randomized phase 3 trial evaluating the efficacy and safety of belantamab mafodotin, bortezomib and dexamethasone versus daratumumab, bortezomib and dexamethasone in participants with relapsed/refractory multiple myeloma. The registry reports a China subpopulation enrollment of 72 participants and a primary progression-free survival analysis based on the Intent-to-Treat population.
| Feature | DREAMM-7 |
|---|---|
| Phase | Phase 3 |
| Population | Participants with relapsed/refractory multiple myeloma in the China subpopulation |
| Design | Randomized, parallel |
| Masking | None |
| Primary purpose | Treatment |
| Enrollment | 72 |
| Number of arms | 2 |
| Start date | 2021-07-28 |
| Primary completion date | 2024-04-03 |
| Status | Active, not recruiting |
| Lead sponsor | GlaxoSmithKline |
| Sponsor type | Industry |
| ClinicalTrials.gov | NCT06868654 |
2. Clinical Question
The primary statistical question is whether the time from randomization to disease progression or death differs between the belantamab mafodotin combination and the daratumumab combination in participants with relapsed/refractory multiple myeloma.
Population
Participants with relapsed/refractory multiple myeloma in the China subpopulation represented by the registry record.
Intervention
Belantamab mafodotin + bortezomib + dexamethasone.
Comparator
Daratumumab + bortezomib + dexamethasone.
Primary question
Does the belantamab mafodotin combination differ from the daratumumab combination in progression-free survival?
3. Trial Design
DREAMM-7 is registered as a phase 3, randomized, parallel-design trial with no masking. Its primary purpose is treatment. The registry reports two treatment arms and an enrollment of 72 participants.
Belantamab Mafodotin Combination
- Belantamab mafodotin
- Bortezomib
- Dexamethasone
Daratumumab Combination
- Daratumumab
- Bortezomib
- Dexamethasone
4. Endpoints
The registry identifies progression-free survival as the primary endpoint. Its time frame is up to approximately 32 months.
| Endpoint | Registry definition | Time frame |
|---|---|---|
| Progression-free Survival (PFS) | PFS is defined as time from randomization until earliest date of disease progression (PD), determined by Independent Review Committee (IRC), according to the International Myeloma Working Group (IMWG) Response Criteria, or death due to any cause. | Up to approximately 32 months |
Definition of disease progression
The registry defines PD as an increase of ≥25% from the lowest value in ≥1 of the specified disease measures. These include serum M-protein with an absolute increase ≥0.5 g/dL; serum M-protein increase ≥1 g/dL when the lowest M-protein is ≥5 g/dL; urine M-protein with an absolute increase ≥200 mg/24h; for participants without measurable serum and urine M-protein levels, a difference between involved and uninvolved serum free light chains with an absolute increase >10 mg/dL; or specified changes involving new lesions or lesion measurements.
The lesion-based criteria include appearance of a new lesion, ≥50% increase from nadir in the sum of the products of the maximal perpendicular diameters of measured lesions (SPD) of >1 lesion, or ≥50% increase in the longest diameter of a previous lesion >1 cm in short axis.
5. Analysis Population and Statistical Comparison
The posted primary analysis uses the Intent-to-Treat (ITT) Population. The registry defines this population as including all randomized participants whether or not randomized treatment was administered.
| Statistical feature | Registry specification |
|---|---|
| Analysis population | Intent-to-Treat Population |
| Population definition | All randomized participants whether or not randomized treatment was administered |
| Groups compared | Belantamab Mafodotin + Bortezomib + Dexamethasone vs Daratumumab + Bortezomib + Dexamethasone |
| Effect measure | Hazard Ratio |
| Model | Cox Proportional Hazards model |
| Model structure | Stratified by number of lines of prior therapy, prior bortezomib and revised international staging system (R-ISS) at screening |
| Treatment variable | Covariate of treatment |
| Confidence interval | 95%, two-sided |
The ITT approach keeps participants associated with the treatment group to which they were randomized. This preserves the randomized comparison even if a participant did not ultimately receive the assigned treatment.
6. Primary Result: Progression-Free Survival
The registry reports a posted statistical analysis for the primary PFS endpoint. The analysis covers up to approximately 32 months and uses the ITT population.
Hazard ratio for progression or death
95% CI: 0.11–0.56 · Two-sided confidence interval
Cox proportional-hazards model stratified by prior-therapy and disease-status factors.
| Endpoint | Belantamab Mafodotin + Bor + Dex | Daratumumab + Bor + Dex | Effect estimate |
|---|---|---|---|
| Progression-free survival | ITT population | ITT population | HR 0.24 (95% CI 0.11–0.56) |
A hazard ratio of 0.24 means that, under the fitted Cox model, the estimated instantaneous hazard of progression or death in the belantamab mafodotin combination group was 0.24 times that in the daratumumab combination group over the analyzed follow-up. Expressed as a relative model-based comparison, this corresponds to an estimated 76% lower hazard of progression or death.
The HR does not mean that 76% of participants avoided progression or death, nor does it mean that an individual participant's probability of progression or death was reduced by exactly 76%. A hazard ratio is a relative time-to-event measure, not an absolute risk difference.
The 95% confidence interval of 0.11–0.56 describes uncertainty around the estimated hazard ratio under the statistical model and sampling framework. It does not describe the range of outcomes that individual participants could experience.
The confidence interval also shows that the estimated effect is not a single known number. The interval spans a substantial range of possible relative hazard values, which is particularly relevant when interpreting a relatively small randomized population of 72 participants.
The posted analysis does not report a p-value. A p-value, when available, would address evidence against a specified null hypothesis; it would not measure the size or clinical importance of the observed effect. The HR and its confidence interval are therefore the central reported measures for this analysis.
The Cox model also carries the usual proportional-hazards interpretation: the single HR is most naturally understood when the relative hazards can reasonably be summarized by a common ratio over time. The registry's posted analysis does not provide a separate assessment of that assumption.
7. Understanding the Cox Proportional-Hazards Analysis
The primary analysis estimates the PFS hazard ratio with a Cox proportional-hazards model. The model is stratified by the number of lines of prior therapy, prior bortezomib, and R-ISS at screening, with treatment included as a covariate.
The hazard ratio associated with treatment is represented conceptually by exp(β). The stratified model allows the baseline hazard to differ across the specified strata while estimating the treatment effect within the model.
Stratification is important because the model does not simply treat all participants as if they had the same underlying baseline hazard. Instead, the analysis accounts for the registered stratification factors when estimating the treatment comparison.
The treatment covariate then represents the central randomized comparison: belantamab mafodotin + bortezomib + dexamethasone versus daratumumab + bortezomib + dexamethasone.
8. Statistical Methods Explained
Why was an ITT population used?
The posted analysis includes all randomized participants whether or not randomized treatment was administered. This is the intention-to-treat principle. It preserves the original treatment assignment created by randomization and therefore avoids redefining the comparison according to treatment actually received.
Why is PFS analyzed as a time-to-event endpoint?
PFS measures time from randomization until the first qualifying event: disease progression or death. Participants who have not experienced the event by the relevant follow-up point contribute information through their available follow-up. This makes time-to-event methods more appropriate than simply comparing the proportion of participants who have progressed at one arbitrary time point.
What does an HR of 0.24 mean?
An HR of 0.24 is a relative comparison of the estimated instantaneous event rates under the Cox model. It indicates a substantially lower estimated hazard for progression or death in the belantamab mafodotin combination group than in the daratumumab combination group. It is not equivalent to a 76-percentage-point difference in PFS.
Why is the confidence interval important?
The 95% CI of 0.11–0.56 communicates statistical precision around the point estimate of 0.24. The interval is wide enough that the exact magnitude of the relative treatment effect remains uncertain. Reporting the interval alongside the HR is therefore more informative than presenting 0.24 alone.
What does stratification accomplish?
The Cox analysis was stratified by number of lines of prior therapy, prior bortezomib and R-ISS at screening. Stratification allows the analysis to account for these registered factors without requiring the same baseline hazard function across their levels. The resulting treatment comparison is therefore based on a model that incorporates the trial's specified stratification structure.
Does the HR prove that every participant benefits?
No. A population-level hazard ratio summarizes the randomized comparison. It does not imply that every individual participant experiences the same treatment effect, and it does not provide an absolute probability of remaining progression-free for an individual participant.
9. Confidence Interval and Effect Size
Point estimate
The HR of 0.24 is the single treatment-effect estimate produced by the posted Cox model.
Uncertainty
The 95% CI extends from 0.11 to 0.56, showing that the magnitude of the estimated relative effect is not known with exact precision.
Relative measure
The HR compares hazards rather than directly comparing median PFS, PFS probabilities, or absolute numbers of events.
No p-value reported
The posted statistical analysis gives the HR and two-sided 95% CI but does not report a p-value.
The confidence interval is especially important in a randomized population of 72 participants. A point estimate alone can make an effect appear more precise than the available information supports. The interval gives readers a direct indication of the uncertainty surrounding the estimated hazard ratio.
10. Safety Results
The registry reports serious adverse events by randomized treatment arm. These data are presented as affected participants divided by participants at risk.
| Safety measure | Belantamab Mafodotin + Bortezomib | Daratumumab + Bor + Dex |
|---|---|---|
| Serious adverse events | 25/30 | 27/40 |
The reported figures correspond to 25 affected participants among 30 at risk in the belantamab mafodotin + bortezomib group and 27 affected participants among 40 at risk in the daratumumab + bortezomib + dexamethasone group.
These are safety counts, not time-to-event efficacy estimates. They should therefore be interpreted separately from the PFS hazard ratio. The denominators also differ between the two reported safety groups, so the raw affected counts alone do not constitute a direct comparison of incidence.
11. Trial Timeline
Study start
The registry lists 2021-07-28 as the study start date.
Primary completion
The registry lists 2024-04-03 as the primary completion date.
Primary PFS time frame
The primary progression-free survival endpoint is defined over a time frame of up to approximately 32 months.
Active, not recruiting
The registry classifies DREAMM-7 as active, not recruiting.
12. What the Hazard Ratio Does — and Does Not — Mean
The reported PFS hazard ratio of 0.24 means that the estimated instantaneous hazard of progression or death was 0.24 times that of the comparator group under the fitted stratified Cox model.
Equivalently, 1 − 0.24 = 0.76, so the estimated relative reduction in hazard is 76%. This is a simple interpretation of the reported HR; it is not an absolute reduction in the probability of progression or death.
The 95% CI of 0.11–0.56 indicates uncertainty around the HR estimate. It does not mean that individual participants have a 95% probability of experiencing an HR somewhere within this interval.
The posted analysis does not provide a median PFS, time-specific PFS probabilities, or an absolute risk difference. Those quantities answer different questions and should not be inferred from the HR alone.
13. Statistical Methodology
Time-to-event analysis
PFS is inherently a time-to-event endpoint because it incorporates both the timing of disease progression and death. The analysis therefore summarizes the relative rate at which the defined PFS event occurs over follow-up rather than reducing the endpoint to a single binary outcome.
Kaplan-Meier framework
For a time-to-event endpoint such as PFS, Kaplan-Meier estimation is commonly used to describe the event-time distribution while accounting for right censoring. The registry's posted statistical analysis, however, specifically identifies the Cox proportional-hazards model as the method used to estimate the hazard ratio. No Kaplan-Meier estimates are reported in the posted analysis summarized here.
Here, di represents events at a particular event time and ni represents participants at risk immediately before that time. The framework allows participants without an observed event by the end of available follow-up to contribute censored information.
Stratified Cox proportional-hazards model
The posted analysis estimates the HR using a Cox proportional-hazards model stratified by number of lines of prior therapy, prior bortezomib and R-ISS at screening, with treatment as a covariate.
The HR is a model-based relative measure. It should not be interpreted as an absolute PFS probability, an absolute risk difference, or a proportion of participants who benefit.
Two-sided confidence interval
The posted analysis specifies a 95% two-sided confidence interval. The interval is reported as 0.11–0.56. A two-sided interval expresses uncertainty on both sides of the point estimate rather than restricting the reported uncertainty to a single direction.
Intention-to-treat analysis
The analysis population includes all randomized participants whether or not randomized treatment was administered. This keeps the primary efficacy comparison tied to the randomized allocation rather than redefining treatment groups after randomization.
14. Stratification in the Primary Model
Three factors are explicitly incorporated into the stratified Cox model: number of lines of prior therapy, prior bortezomib and R-ISS at screening.
| Stratification factor | Role in the analysis |
|---|---|
| Number of lines of prior therapy | Stratification factor in the Cox proportional-hazards model |
| Prior bortezomib | Stratification factor in the Cox proportional-hazards model |
| R-ISS at screening | Stratification factor in the Cox proportional-hazards model |
| Treatment | Covariate used to estimate the treatment hazard ratio |
Stratification is different from simply adding a variable to an ordinary regression model. In a stratified Cox analysis, each stratum can have its own baseline hazard while the treatment effect is estimated across the stratified structure.
This approach is particularly useful in a randomized trial when important baseline factors are incorporated into the planned analysis. It can improve alignment between the statistical model and the trial's design without requiring the model to estimate a separate coefficient for every stratification factor.
15. Limitations and Interpretation Issues
- Small randomized population: the registry reports enrollment of 72 participants. A relatively small population can lead to substantial statistical uncertainty, reflected in part by the 95% CI of 0.11–0.56.
- Limited reported efficacy detail: the posted statistical analysis reports the PFS HR and confidence interval but does not provide a median PFS or time-specific PFS probabilities.
- No p-value in the posted analysis: the registry's posted statistical analysis gives the HR and two-sided 95% CI but does not report a p-value.
- Proportional-hazards interpretation: the HR is based on a Cox proportional-hazards model. A single HR is most straightforward to interpret when the relative hazards can reasonably be summarized over time by a common ratio. The posted analysis does not provide a separate assessment of that assumption.
- Unmasked design: the registry classifies masking as none. The primary endpoint is nevertheless defined using progression determined by an Independent Review Committee.
- Endpoint specificity: PFS combines disease progression and death into a single endpoint, so the HR does not separately quantify the treatment effect on progression and overall survival.
- Safety denominator: serious adverse events are reported as affected / at risk, with 25/30 and 27/40. Raw counts should not be interpreted without considering their denominators.
- Generalizability: the record represents a China subpopulation, so the reported estimates describe this registered study population rather than automatically representing every patient with relapsed/refractory multiple myeloma.
16. Why This Trial Matters Statistically
DREAMM-7 is a useful teaching example because its primary analysis shows how a randomized clinical trial can combine an ITT population, a clinically defined time-to-event endpoint, prespecified stratification factors and a stratified Cox model into a single treatment-effect estimate.
| Concept | How it appears in DREAMM-7 |
|---|---|
| Randomization | The study is registered as randomized with two parallel arms. |
| ITT analysis | The primary analysis population includes all randomized participants whether or not randomized treatment was administered. |
| Time-to-event endpoint | PFS measures time from randomization to disease progression or death. |
| Independent review | Disease progression is determined by an Independent Review Committee according to IMWG Response Criteria. |
| Hazard ratio | The primary treatment effect is reported as an HR of 0.24. |
| Confidence interval | The HR is accompanied by a two-sided 95% CI of 0.11–0.56. |
| Cox model | A Cox proportional-hazards model is used to estimate the HR. |
| Stratification | The Cox model is stratified by prior-therapy burden, prior bortezomib and R-ISS at screening. |
| Safety analysis | Serious adverse events are reported by arm as affected participants / participants at risk. |
17. A Practical Reading of the Primary Result
There are several layers to the reported PFS result. The first is the direction of the treatment effect: the HR is below 1. The second is the magnitude: 0.24 is substantially below 1. The third is precision: the 95% CI extends from 0.11 to 0.56. The fourth is the population and model: the estimate comes from the ITT population and a stratified Cox model.
These layers should be kept separate. The HR describes the relative event hazard. The confidence interval describes uncertainty around that estimate. The ITT definition describes who contributed to the primary randomized comparison. The stratification variables describe how the Cox model accounted for specified baseline factors.
The absence of a reported p-value does not prevent interpretation of the HR and confidence interval. It simply means that the posted statistical analysis does not supply that additional hypothesis-testing quantity. The confidence interval provides information about the precision and range of the estimated treatment effect.
18. Related Tutorials
Learn more about the methods used in this trial:
19. Related Calculators
20. Sources
- ClinicalTrials.gov: DREAMM-7, NCT06868654.
Continue through the Clinical Biostats statistical library
Explore related survival-analysis methods and statistical calculators for clinical-trial analysis.
21. Record Summary
DREAMM-7 provides a focused example of randomized time-to-event analysis in a phase 3 multiple-myeloma trial. The primary endpoint is progression-free survival, defined from randomization to the earliest date of Independent Review Committee-determined disease progression according to IMWG Response Criteria or death from any cause, over a time frame of up to approximately 32 months. The primary analysis uses the ITT population and a Cox proportional-hazards model stratified by number of lines of prior therapy, prior bortezomib and R-ISS at screening.
The reported PFS hazard ratio is 0.24, with a 95% two-sided confidence interval of 0.11–0.56. Statistically, the key lesson is that this estimate should be interpreted together with its confidence interval, analysis population and model specification. Clinically, the HR describes a relative difference in the hazard of progression or death; it does not by itself provide an absolute probability of remaining progression-free or a median PFS.
The safety information reported in the registry shows serious adverse events in 25/30 participants in the belantamab mafodotin + bortezomib group and 27/40 participants in the daratumumab + bortezomib + dexamethasone group. These safety data are appropriately considered separately from the primary PFS treatment-effect estimate.