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
CONCERTO was a randomized, quadruple-masked, parallel phase 3 trial evaluating laquinimod in participants with relapsing remitting multiple sclerosis. The registry reports 2,199 participants, six arms, and a primary endpoint based on time to confirmed disease progression sustained for at least three months.
| Feature | CONCERTO |
|---|---|
| Phase | Phase 3 |
| Condition | Multiple Sclerosis |
| Study population | Participants with relapsing remitting multiple sclerosis (RRMS) |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Primary purpose | Treatment |
| Enrollment | 2,199 |
| Primary endpoint type | Time-to-event |
| Primary hypothesis | Superiority |
| Primary analysis | Cox proportional-hazards model |
| Primary effect measure | Hazard ratio |
| Trial status | Completed |
| Start | February 20, 2013 |
| Primary completion | April 13, 2015 |
2. Clinical Question
The primary statistical question was whether treatment with laquinimod 0.6 mg differed from placebo with respect to the time to confirmed disease progression in participants with relapsing remitting multiple sclerosis over the period from baseline to Month 24.
Population
Participants with relapsing remitting multiple sclerosis (RRMS).
Intervention
Laquinimod. The posted primary analysis specifically compares laquinimod 0.6 mg with placebo.
Comparator
Placebo in the placebo-controlled phase.
Primary question
Does laquinimod 0.6 mg change the time to confirmed disease progression compared with placebo?
3. Trial Design
The registry reports six arms, while the posted primary statistical analysis concerns the placebo-controlled comparison of laquinimod 0.6 mg versus placebo. The safety information also identifies a laquinimod 1.2 mg group during the placebo-controlled phase and several groups during an active-treatment phase.
4. Trial Timeline
Trial start
The registry lists February 20, 2013 as the study start date.
Primary completion
The registry lists April 13, 2015 as the primary completion date.
Registry status
The ClinicalTrials.gov record identifies the study as completed and reports posted results, including one formal statistical analysis.
5. Primary Endpoint
| Endpoint | Registry definition | Time frame |
|---|---|---|
| Time to Confirmed Disease Progression (CDP) Confirmed After At Least 3 Months | Time to CDP was defined as the time to a sustained increase in Kurtzke's Expanded Disability Status Scale (EDSS) score of at least 1 point if baseline EDSS score was less than or equal to 5.0, or at least 0.5 point if the baseline EDSS score was 5.5, over a period of at least three months. EDSS assesses disability in 8 functional systems with an overall score ranging from 0 (normal) to 10 (death due to multiple sclerosis [MS]). Data is presented as distribution of CDP (number of participants with CDP) sustained for 3 months. | Baseline to Month 24 |
The endpoint is inherently time-to-event because the analysis is concerned not only with whether confirmed disease progression occurred, but also with the time from baseline to that event. Participants who do not experience the event during the relevant observation period can contribute information through the point at which their follow-up ends.
6. Statistical Methodology
Intention-to-treat analysis
The posted primary analysis used an intention-to-treat (ITT) analysis set that included all randomized participants. This preserves the randomized treatment comparison as the foundation for the efficacy analysis rather than redefining treatment groups according to subsequent treatment exposure.
Cox proportional-hazards model
The primary comparison between laquinimod 0.6 mg and placebo was conducted using a baseline-adjusted Cox proportional-hazards model. The reported effect measure was the hazard ratio.
The registry states that categorical EDSS at baseline, country/geographical region (CGR), categorical age at baseline, and T2 volume at baseline were included as covariates in the model.
Covariate adjustment
The model included categorical EDSS at baseline, defined as less than or equal to 4 versus greater than 4; country/geographical region; categorical age at baseline, defined as less than or equal to 38 versus greater than 38; and T2 volume at baseline. These variables were incorporated into the baseline-adjusted Cox model rather than relying only on an unadjusted treatment comparison.
Hazard ratio
The hazard ratio compares the estimated instantaneous event rates between the treatment groups under the Cox model. A value below 1 corresponds to a lower estimated hazard in the treatment group, whereas a value above 1 corresponds to a higher estimated hazard. The hazard ratio is a relative time-to-event measure rather than an absolute probability of progression.
The comparison was laquinimod 0.6 mg versus placebo, using the ITT analysis set and the baseline-adjusted Cox proportional-hazards model.
7. Results: Confirmed Disease Progression
The posted primary analysis compared placebo with laquinimod 0.6 mg for time to confirmed disease progression after at least three months, from baseline to Month 24. The analysis used all randomized participants in the ITT analysis set.
Primary efficacy estimate
Hazard ratio for confirmed disease progression
95% CI: 0.668–1.313 · P = 0.7057
Baseline-adjusted Cox proportional-hazards model; laquinimod 0.6 mg vs placebo.
| Element | Posted primary analysis |
|---|---|
| Endpoint | Time to Confirmed Disease Progression (CDP) Confirmed After At Least 3 Months |
| Time frame | Baseline to Month 24 |
| Analysis population | ITT analysis set including all randomized participants |
| Comparison | Laquinimod 0.6 mg vs placebo |
| Method | Cox proportional-hazards model |
| Effect measure | Hazard ratio |
| Estimate | 0.937 |
| 95% CI | 0.668–1.313 |
| P-value | = 0.7057 |
| Hypothesis | Superiority |
An HR of 0.937 means that the fitted model estimated the instantaneous rate of confirmed disease progression in the laquinimod 0.6 mg group at approximately 93.7% of the corresponding rate in the placebo group over the analyzed period. Expressed as a simple relative transformation, 1 − 0.937 = 0.063, so the point estimate corresponds to an approximately 6.3% lower estimated hazard.
That point estimate should not be interpreted as saying that 6.3% of participants avoided progression, that each participant experienced a 6.3% reduction in risk, or that the cumulative probability of progression was reduced by 6.3%. A hazard ratio is a model-based relative measure of event rates over time.
The 95% confidence interval of 0.668–1.313 describes the statistical uncertainty around the estimated hazard ratio. It spans values below and above 1, so the interval is compatible with a lower hazard, little difference, or a higher hazard under the model and sampling framework. The width of the interval also shows that the point estimate alone is not an adequate description of the uncertainty.
The P-value of 0.7057 addresses the statistical evidence against the null hypothesis under the specified model; it is not a measure of the size or clinical importance of the treatment effect. A p-value should therefore be read together with the hazard ratio and its confidence interval rather than treated as a standalone measure of benefit.
Because the analysis is based on a Cox proportional-hazards model, interpretation of a single hazard ratio also depends on the proportional-hazards framework. If the relative hazards change materially over time, a single HR may summarize a complex time-varying pattern imperfectly. The posted analysis does not provide a separate assessment of that assumption.
8. Understanding the Primary Endpoint
Why confirmed disease progression is a time-to-event endpoint
The registry defines confirmed disease progression using sustained change in EDSS rather than a single transient observation. The requirement for persistence over at least three months is part of the endpoint definition, so the event is determined only after the qualifying disability increase has been sustained for the required period.
Why EDSS thresholds depend on baseline status
The progression criterion is baseline-dependent. Participants with a baseline EDSS score less than or equal to 5.0 require an increase of at least 1 point, whereas participants whose baseline EDSS score is 5.5 require an increase of at least 0.5 point. The statistical analysis therefore operates on an endpoint whose event definition already incorporates baseline disability.
Why time matters
Two participants can both experience confirmed disease progression while having very different times to the event. A time-to-event analysis retains this timing information and also accommodates participants who have not experienced the event by the end of their observed follow-up.
9. Statistical Methods Explained
Why was a Cox proportional-hazards model used?
The primary endpoint is time to confirmed disease progression, making survival-analysis methodology appropriate. The Cox model estimates a relative hazard while allowing the underlying baseline hazard to remain unspecified. It also permits adjustment for baseline covariates, which was explicitly incorporated into this analysis.
What does an HR of 0.937 mean?
Under the fitted model, the estimated instantaneous rate of confirmed disease progression for laquinimod 0.6 mg was 0.937 times the corresponding rate for placebo. The point estimate alone should not be translated into a cumulative probability or interpreted as an individual-level risk reduction.
Why was covariate adjustment used?
The posted model adjusted for categorical baseline EDSS, country/geographical region, categorical baseline age, and baseline T2 volume. Adjustment incorporates these prespecified baseline characteristics into the estimated treatment comparison and can account for prognostic information contained in those variables.
Why is the confidence interval as important as the hazard ratio?
The hazard ratio is a point estimate, whereas the confidence interval communicates the uncertainty around that estimate. Here, the 95% CI extends from 0.668 to 1.313, illustrating that the observed point estimate is compatible with a substantially different range of underlying hazard ratios.
What does the p-value of 0.7057 tell us?
The p-value quantifies the compatibility of the observed result with the null hypothesis under the specified statistical model and testing framework. It does not quantify the magnitude of the treatment effect, the probability that the treatment is effective, or the probability that the null hypothesis is true.
Why is ITT analysis important in this trial?
The ITT analysis set included all randomized participants. Analyzing participants according to randomized assignment maintains the comparison created by randomization and avoids changing the primary efficacy population based on subsequent treatment behavior.
10. Safety Results
The registry reports serious adverse events separately for the placebo-controlled and active-treatment phases. These counts are presented as affected participants over the corresponding number at risk.
| Phase / group | Serious adverse events | Affected / at risk |
|---|---|---|
| Placebo-Controlled Phase: Laquinimod 0.6 | Serious adverse events | 57/727 |
| Placebo-Controlled Phase: Laquinimod 1.2 | Serious adverse events | 49/732 |
| Placebo-Controlled Phase: Placebo | Serious adverse events | 43/740 |
| Active Treatment Phase: Early Laquinimod | Serious adverse events | 24/580 |
| Active Treatment Phase: Early Laquinimod | Serious adverse events | 10/268 |
| Active Treatment Phase: Off Drug | Serious adverse events | 5/215 |
| Active Treatment Phase: From Placebo to | Serious adverse events | 5/311 |
| Active Treatment Phase: From Placebo to | Serious adverse events | 9/236 |
For the placebo-controlled phase, the registry therefore reports serious adverse events in 57 of 727 participants receiving laquinimod 0.6, 49 of 732 receiving laquinimod 1.2, and 43 of 740 receiving placebo. These safety counts should be interpreted separately from the primary time-to-event efficacy analysis because the safety populations and exposure histories can differ from the primary ITT efficacy framework.
11. Safety Interpretation
Numerator and denominator matter
A serious-adverse-event count such as 57/727 contains both the number of affected participants and the number at risk. Reporting only the numerator would obscure the population in which the events were observed.
Phase matters
The registry separates placebo-controlled and active-treatment phases. These phases should not be treated as interchangeable exposure periods.
Safety is distinct from efficacy
The primary efficacy analysis addresses time to confirmed disease progression. Serious adverse events represent a separate safety dimension and should not be combined with the efficacy hazard ratio.
Exposure history matters
Active-treatment groups include different treatment histories, including an off-drug group and groups identified as transitioning from placebo. Those histories are relevant when interpreting safety counts.
12. Limitations and Interpretation Issues
- Single posted primary analysis: the registry reports one formal statistical analysis for the primary endpoint, focused on the laquinimod 0.6 mg versus placebo comparison.
- Confidence interval width: the 95% CI of 0.668–1.313 spans a substantial range of possible hazard ratios, so the point estimate should not be considered independently of its uncertainty.
- Proportional-hazards framework: the Cox model expresses the treatment effect through a hazard ratio and therefore relies on the proportional-hazards framework for a straightforward single-HR interpretation. The posted analysis does not report a separate assessment of that assumption.
- Time-to-event censoring: time-to-event analyses can include participants whose event status is not observed through the entire study period by censoring their follow-up according to the analysis rules. The posted statistical analysis does not provide sufficient detail here to characterize every censoring rule.
- Covariate adjustment: the estimate is from a baseline-adjusted Cox model, so it should not be interpreted as an unadjusted treatment hazard ratio.
- Analysis population: the primary analysis used the ITT analysis set, including all randomized participants. Safety summaries are presented by phase and treatment history and should not be assumed to use the same analytical population.
- Multiple arms: the registry describes six arms, but the posted primary statistical analysis concerns one specific comparison: laquinimod 0.6 mg versus placebo. The primary result should not be generalized to every arm without a corresponding analysis.
13. Why This Trial Matters Statistically
CONCERTO is a useful teaching case for understanding how a randomized clinical trial can combine an objective disability endpoint with survival-analysis methodology. The statistical structure is especially instructive because the endpoint is based on sustained EDSS progression, the analysis is conducted in an ITT population, and the primary treatment effect is estimated through a covariate-adjusted Cox proportional-hazards model.
| Concept | How it appears in CONCERTO |
|---|---|
| Randomization | The registry describes the allocation as randomized. |
| Quadruple masking | The registry identifies the study as quadruple-masked. |
| Parallel design | The trial uses a parallel design with six arms. |
| ITT analysis | The primary analysis includes all randomized participants. |
| Time-to-event endpoint | The primary endpoint is time to confirmed disease progression sustained for at least three months. |
| Hazard ratio | The treatment effect is reported as an HR of 0.937. |
| Cox model | The primary analysis uses a Cox proportional-hazards model. |
| Covariate adjustment | Baseline EDSS, country/geographical region, baseline age, and T2 volume are included as covariates. |
| Confidence interval | The primary estimate is accompanied by a two-sided 95% CI of 0.668–1.313. |
| P-value | The posted primary analysis reports P = 0.7057. |
14. Primary Result: Statistical Interpretation in Context
The estimated hazard ratio of 0.937 is below 1. In the fitted Cox model, this corresponds to an estimated hazard of confirmed disease progression that was approximately 6.3% lower for laquinimod 0.6 mg than placebo at the point estimate.
The 95% CI of 0.668–1.313 is considerably wider than the single point estimate and crosses 1. It therefore encompasses both lower and higher estimated hazards relative to placebo. The interval is the more informative description of precision than the point estimate alone.
The reported P = 0.7057 is large relative to conventional significance thresholds, but the more important statistical interpretation is that the observed result does not provide strong evidence against the null hypothesis under the specified superiority analysis. The p-value itself is not an effect-size measure.
The HR of 0.937 does not establish a 6.3% reduction in the cumulative probability of confirmed disease progression, does not describe an individual participant's risk, and does not imply that the treatment effect is constant at every point in time. Those interpretations require information beyond a single Cox hazard ratio.
15. Statistical Reporting Checklist
| Question | CONCERTO primary analysis |
|---|---|
| What was analyzed? | Time to confirmed disease progression after at least three months. |
| When? | Baseline to Month 24. |
| Who? | All randomized participants in the ITT analysis set. |
| What comparison? | Laquinimod 0.6 mg versus placebo. |
| What model? | Cox proportional-hazards model. |
| Was the model adjusted? | Yes; baseline EDSS category, country/geographical region, age category, and T2 volume were included as covariates. |
| What effect measure? | Hazard ratio. |
| What estimate? | 0.937. |
| What uncertainty? | Two-sided 95% CI 0.668–1.313. |
| What p-value? | = 0.7057. |
| What hypothesis? | Superiority. |
16. Related Tutorials
Learn more about the methods used in this trial:
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: NCT01707992.
- PubMed: PMID 34378456.
Continue through Clinical Biostats
Explore statistical tutorials, calculators, and additional clinical trial analyses covering study design, survival analysis, regression, and clinical-trial methodology.
19. Record Summary
CONCERTO provides a focused example of time-to-event methodology in a randomized phase 3 trial. The primary endpoint was time to confirmed disease progression sustained for at least three months, assessed from baseline to Month 24. The posted analysis used an ITT population and a baseline-adjusted Cox proportional-hazards model comparing laquinimod 0.6 mg with placebo. The resulting hazard ratio was 0.937, with a two-sided 95% CI of 0.668–1.313 and P = 0.7057.
The statistical interpretation depends on considering all three components together: the point estimate describes the fitted relative hazard, the confidence interval describes uncertainty around that estimate, and the p-value describes the statistical evidence under the specified hypothesis-testing framework. The registry also reports serious adverse events across placebo-controlled and active-treatment phases, providing a separate safety dimension to the randomized efficacy analysis.