This page separates reported registry data from statistical interpretation. The ClinicalTrials.gov record reports a primary time-to-event endpoint and posted outcome measures, but the ClinicalTrials.gov record contains no formal statistical analyses for that primary endpoint. No treatment-effect estimate, confidence interval, or p-value is therefore reported here.
1. Trial at a Glance
SURPASS-CVOT is a completed randomized, double-masked, parallel phase 3 trial comparing tirzepatide with dulaglutide in participants with type 2 diabetes. Its registered primary endpoint is a time-to-event composite of cardiovascular death, myocardial infarction, or stroke, evaluated from randomization through week 259.
| Feature | SURPASS-CVOT |
|---|---|
| Trial name | SURPASS-CVOT |
| NCT identifier | NCT04255433 |
| Phase | Phase 3 |
| Condition | Type 2 Diabetes Mellitus |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Primary purpose | Treatment |
| Enrollment | 13299.0 |
| Interventions | Tirzepatide; Dulaglutide |
| Primary endpoint type | Time-to-event |
| Results posted | Yes |
| Outcome measures posted | 14 |
| Lead sponsor | Eli Lilly and Company |
| Sponsor type | Industry |
2. Clinical Question
The trial addresses whether tirzepatide and dulaglutide can be compared with respect to the time to the first occurrence of a major cardiovascular composite endpoint in participants with type 2 diabetes.
Population
Participants with Type 2 Diabetes Mellitus enrolled in the phase 3 SURPASS-CVOT trial.
Intervention
Tirzepatide.
Comparator
Dulaglutide.
Primary question
How does the time from randomization to the first occurrence of MACE-3 compare between the tirzepatide and dulaglutide groups?
3. Trial Design
Tirzepatide
- Intervention: Tirzepatide
- Drug intervention
- Included in the randomized parallel comparison
Dulaglutide
- Intervention: Dulaglutide
- Drug intervention
- Included in the randomized parallel comparison
The ClinicalTrials.gov record identifies the two interventions and the overall enrollment, but do not provide randomized enrollment counts by arm. Those arm-specific sample sizes are therefore not inferred here.
4. Primary Endpoint
| Endpoint | Registry definition | Time frame | Endpoint type |
|---|---|---|---|
| Number of Participants From Randomization to First Occurrence of Death From MACE-3 | Composite endpoint of major adverse cardiovascular events: death from cardiovascular causes, myocardial infarction (MI) or stroke. The Clinical Endpoint Committee (CEC) definition combines these events, with time to event defined as the number of days from randomization to the first occurrence of any of these events during study follow-up. | From randomization (week 0) up to week 259 | Time-to-event |
The registry's primary endpoint is therefore not simply a count of cardiovascular events. The key statistical object is the time from randomization to the first qualifying event for each participant. Participants who do not experience a qualifying event during the relevant observation period contribute censored follow-up rather than an observed event time.
5. Results
Results are posted for the primary endpoint, but the ClinicalTrials.gov record contains no formal statistical analyses. Consequently, the ClinicalTrials.gov record does not report a formal between-group hazard ratio, confidence interval, p-value, Kaplan-Meier estimate, or other comparative statistical estimate for the primary endpoint.
What the registry does report
| Result element | Available in the ClinicalTrials.gov record |
|---|---|
| Primary endpoint definition | Yes |
| Primary endpoint time frame | Yes: from randomization (week 0) up to week 259 |
| Results posted | Yes |
| Outcome measures posted | 14 |
| Formal statistical analyses | No the posted statistical analyses reported |
| Hazard ratio | Not in the posted results |
| Confidence interval | Not in the posted results |
| P-value | Not in the posted results |
How this endpoint would normally be analyzed
For a randomized trial with a time-to-first-event primary endpoint, a conventional analysis would generally compare the event-time distributions between treatment groups using a Kaplan-Meier estimator, with a treatment-group comparison commonly performed using a log-rank test and a relative treatment-effect estimate commonly obtained from a Cox proportional-hazards model. The exact model, stratification factors, censoring rules, analysis population, multiplicity procedure, and hypothesis-testing framework should come from the prespecified statistical analysis plan.
Those methods describe how an analysis of this endpoint would ordinarily be constructed; they do not constitute a reported SURPASS-CVOT statistical result. The ClinicalTrials.gov record does not provide enough information to reproduce such an analysis independently.
6. Clinical Biostats Interpretation
The endpoint measures the time from randomization until the first occurrence of cardiovascular death, myocardial infarction, or stroke. A participant who remains free of all three components through the applicable observation period would not have an observed primary event and would ordinarily contribute censored follow-up.
If the prespecified analysis reported a hazard ratio below 1 for tirzepatide relative to dulaglutide, that would indicate a lower estimated instantaneous rate of the first MACE-3 event under the fitted time-to-event model. It would not mean that the same percentage of participants avoided an event, nor would it directly provide an absolute risk difference.
A confidence interval around a hazard ratio would describe statistical uncertainty around the estimated relative treatment effect under the specified model and sampling framework. It would not describe the range of treatment effects experienced by individual participants.
A p-value addresses evidence against a specified null hypothesis under the statistical testing framework. It does not measure the magnitude of a treatment effect. A clinically important analysis therefore requires the effect estimate and its confidence interval in addition to any hypothesis-test result.
7. Statistical Methodology
Time-to-event analysis
The registered primary endpoint is explicitly a time-to-event outcome. This matters because participants can have different lengths of observed follow-up, and some participants may remain event-free when observation ends. The analysis therefore needs to account for both event times and censoring.
The primary endpoint definition specifies that time to event is measured as the number of days from the date of randomization to the first occurrence of cardiovascular death, myocardial infarction, or stroke.
Kaplan-Meier estimation
Kaplan-Meier estimation is a standard nonparametric approach for describing time-to-event distributions. It uses the observed event times while retaining information from participants who are censored before experiencing the event.
Here, di represents the number of events at an event time and ni represents the number at risk immediately before that time.
Log-rank comparison
A log-rank test is commonly used to compare survival or event-free distributions between randomized groups. It evaluates the accumulated difference in observed versus expected events over follow-up rather than reducing the entire time course to a single fixed time point.
Cox proportional-hazards model
A Cox model is commonly used to estimate a hazard ratio comparing the event rate between treatment groups. The model is attractive for clinical-trial time-to-event analyses because it does not require specification of the baseline hazard function.
This is a methodological interpretation only. The registry-reported SURPASS-CVOT data do not report a hazard ratio.
Composite endpoint analysis
MACE-3 is a composite endpoint. The first occurrence of cardiovascular death, myocardial infarction, or stroke becomes the participant's primary event. This can increase the number of observed primary events relative to a single-component endpoint, but the interpretation remains tied to the combined definition.
Censoring
Time-to-event methods generally distinguish observed events from censored observations. Censoring rules matter because the statistical analysis assumes that the censoring mechanism is appropriately handled by the prespecified method. The ClinicalTrials.gov record does not provide the detailed censoring rules or missing-data procedures.
8. Statistical Methods Explained
Why is this endpoint analyzed as time-to-event rather than as a simple proportion?
Because the registered endpoint measures the time from randomization to the first occurrence of cardiovascular death, myocardial infarction, or stroke. A simple proportion would discard information about when events occurred and how long event-free follow-up lasted.
Why is the first event important in a composite endpoint?
The registry defines time to event as the time from randomization to the first occurrence of any component of MACE-3. Once a qualifying first event occurs, the participant has experienced the primary endpoint regardless of which component occurred first.
Why would Kaplan-Meier methods be useful?
Kaplan-Meier methods allow investigators to estimate the probability of remaining free of the primary event over time while accommodating participants whose follow-up ends before an event is observed.
What would a hazard ratio add?
A hazard ratio would provide a relative comparison of the instantaneous event rate between the randomized groups under a specified survival model. It would complement, rather than replace, absolute event-free probabilities at clinically relevant time points.
Why does the proportional-hazards assumption matter?
A conventional Cox hazard ratio is most straightforward to interpret as a single relative effect when the treatment hazards are reasonably proportional over time. If hazards vary substantially in relative magnitude over follow-up, a single hazard ratio can compress a more complicated time-varying treatment effect.
Why can the registry results not be used to calculate the treatment effect here?
The ClinicalTrials.gov record provides the endpoint definition, trial design, enrollment, and safety counts, but no statistical-analysis record containing event counts, effect estimates, confidence intervals, or p-values. Without those data, calculating a comparative treatment effect would require assumptions that are not supported by the ClinicalTrials.gov record.
9. Safety
The ClinicalTrials.gov record reports serious adverse events by arm as affected participants among those at risk. The counts are reported exactly as provided and are not converted into percentages.
| Safety measure | Affected | At risk |
|---|---|---|
| Serious adverse events — Tirzepatide (MTD) | 2117 | 6647 |
| Serious adverse events — 1.5 mg Dulaglutide | 2121 | 6647 |
These are serious-adverse-event counts rather than the primary cardiovascular efficacy endpoint. They should therefore be interpreted separately from MACE-3. The ClinicalTrials.gov record does not provide a formal statistical comparison of serious adverse events between the two arms.
10. Trial Timeline
Study start
The trial began on May 29, 2020 according to the ClinicalTrials.gov record.
Randomized clinical-trial population
The trial enrolled 13299.0 participants across two randomized parallel intervention groups.
MACE-3 through week 259
The registered primary endpoint follows participants from randomization at week 0 up to week 259 for the first occurrence of cardiovascular death, myocardial infarction, or stroke.
Primary completion
The ClinicalTrials.gov record lists June 12, 2025 as the primary completion date.
11. Design Topics That Matter Statistically
Randomization
Randomization creates the basis for comparing the two intervention groups without assigning treatment according to investigators' or participants' preferences.
Double masking
The registry describes the study as double masked. Masking can reduce the influence of treatment knowledge on participant behavior, assessment, and other aspects of trial conduct.
Parallel design
Participants are assigned to one of the two interventions and remain part of that randomized treatment comparison rather than crossing between treatment sequences as a feature of the design.
Time-to-event endpoint
The primary endpoint incorporates both whether a first cardiovascular event occurred and when it occurred during follow-up.
Non-inferiority margin
No non-inferiority margin is included in the ClinicalTrials.gov record. It would therefore be inappropriate to assign a margin or describe the primary analysis as a non-inferiority analysis from the available information.
Crossover
No crossover procedure is reported in the ClinicalTrials.gov record. No crossover effect on interpretation is therefore inferred.
Factorial design
The registry identifies a two-arm parallel design. No factorial design is reported.
Interim analysis
No interim-analysis procedure is included in the ClinicalTrials.gov record. No alpha-spending rule, stopping boundary, or interim-testing schedule is therefore described.
Missing data and imputation
The ClinicalTrials.gov record does not report a missing-data or imputation strategy. For a time-to-event primary endpoint, censoring and follow-up rules would ordinarily be prespecified in the statistical analysis plan.
Stratification
The ClinicalTrials.gov record does not identify randomization strata or stratification variables. No stratification factors are therefore added to the analysis description.
Bayesian methods
No Bayesian analysis is reported in the ClinicalTrials.gov record.
12. Limitations
- Incomplete statistical-analysis detail: the ClinicalTrials.gov record contains no formal statistical analyses, so the formal treatment-effect estimates for the primary endpoint cannot be reproduced or reported here.
- No primary event counts: the ClinicalTrials.gov record does not provide the number of MACE-3 events by randomized treatment group.
- No confidence intervals: the ClinicalTrials.gov record does not provide uncertainty intervals for the primary endpoint.
- No p-values: the ClinicalTrials.gov record does not report a formal hypothesis-test result for the primary endpoint.
- No analysis population details: the ClinicalTrials.gov record does not specify the exact efficacy or safety analysis populations beyond the reported serious-adverse-event denominators.
- No stratification information: the ClinicalTrials.gov record does not identify whether the time-to-event model was stratified or which variables, if any, were used for stratification.
- No multiplicity procedure: although 14 outcome measures are reported as posted, the ClinicalTrials.gov record does not describe a multiplicity-control strategy.
- No interim-analysis information: the ClinicalTrials.gov record does not describe interim looks, stopping boundaries, or alpha spending.
- Composite interpretation: MACE-3 combines cardiovascular death, myocardial infarction, and stroke. An overall composite result, if reported, would not necessarily imply identical effects on each component.
13. Why This Trial Matters Statistically
SURPASS-CVOT provides a useful teaching example of how a large randomized clinical trial can be structured around a composite cardiovascular time-to-event endpoint. The statistical questions begin with the design—randomization, parallel treatment groups, and double masking—and continue through endpoint definition, censoring, event-time estimation, and comparative modeling.
| Concept | How it appears in SURPASS-CVOT |
|---|---|
| Randomization | The trial uses randomized allocation. |
| Parallel design | The design model is parallel with two intervention arms. |
| Double masking | The registry identifies the masking as double. |
| Time-to-event analysis | The primary endpoint measures time from randomization to first MACE-3 occurrence. |
| Composite endpoint | MACE-3 combines cardiovascular death, myocardial infarction, and stroke. |
| Kaplan-Meier methodology | A conventional method for estimating event-free survival in a censored time-to-event analysis. |
| Cox modeling | A conventional framework for estimating a relative hazard between randomized groups. |
| Confidence intervals | Would quantify uncertainty around an estimated treatment effect if reported. |
| Multiplicity | Relevant because the registry reports 14 outcome measures, although no multiplicity procedure is reported. |
| Analysis transparency | The ClinicalTrials.gov record demonstrates why endpoint definitions alone are insufficient to reconstruct a formal statistical comparison. |
14. Related Tutorials
Learn more about the methods used in this trial:
15. Related Calculators
16. Sources
- ClinicalTrials.gov: SURPASS-CVOT, NCT04255433.
- Linked publication: PubMed PMID 41903177.
- Linked publication: PubMed PMID 41406444.
- Linked publication: PubMed PMID 40746703.
- Linked publication: PubMed PMID 35949358.
Continue through the Clinical Biostats learning pathway
Explore clinical-trial methods, statistical concepts, and calculators that build on the design and endpoint principles illustrated by randomized trials such as SURPASS-CVOT.
17. Record Summary
SURPASS-CVOT is a completed phase 3, randomized, double-masked, parallel trial with 13299.0 enrolled participants and two drug interventions: tirzepatide and dulaglutide. Its registered primary endpoint is a time-to-event composite of cardiovascular death, myocardial infarction, or stroke, measured from randomization through week 259. The ClinicalTrials.gov record reports 14 posted outcome measures and serious-adverse-event counts by arm, but it does not provide formal statistical-analysis entries for the primary endpoint.
From a statistical perspective, the trial is therefore most useful here as an example of the structure required for a cardiovascular time-to-event analysis: randomized comparison, a prespecified composite endpoint, event-time measurement, censoring, and the potential use of Kaplan-Meier estimation, log-rank testing, and Cox modeling. Those methods explain how such an endpoint is ordinarily analyzed, while the absence of the ClinicalTrials.gov record sets a clear boundary on what can responsibly be concluded from the available registry data.