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
PACIFIC-AMI was a randomized, parallel-group, quadruple-masked phase 2 trial evaluating BAY 2433334, an oral FXIa inhibitor, in patients following an acute myocardial infarction. The registry reports 1,601 enrolled participants, four arms, two primary endpoints, and formal time-to-event analyses for the primary efficacy and safety endpoints.
| Feature | PACIFIC-AMI |
|---|---|
| Trial phase | Phase 2 |
| Condition | Acute Myocardial Infarction |
| Design | Randomized, parallel-group, quadruple-masked |
| Allocation | Randomized |
| Primary purpose | Treatment |
| Arms | 4 |
| Enrollment | 1,601 |
| Interventions | BAY2433334 and BAY2433334 matching placebo |
| Primary endpoints | Composite cardiovascular efficacy endpoint and BARC bleeding safety endpoint |
| Primary endpoint type | Binary in the registered endpoint classification; formally analyzed as time-to-event outcomes in the posted statistical analyses |
| Lead sponsor | Bayer |
| Trial status | Completed |
2. Clinical Question
The clinical question was whether different doses of BAY 2433334 could be evaluated for cardiovascular efficacy and bleeding safety in participants following an acute myocardial infarction.
Population
Patients following an acute myocardial infarction, with the registry condition listed as Acute Myocardial Infarction.
Intervention
BAY2433334, identified in the registry as the drug intervention and analyzed at 10 mg, 20 mg, and 50 mg dose levels.
Comparator
BAY2433334 matching placebo.
Primary questions
How did the asundexian dose groups compare with placebo for the composite of cardiovascular death, myocardial infarction, stroke, and stent thrombosis, and for BARC bleeding type 2, 3, and 5?
3. Trial Design
Asundexian 10 mg
- Asundexian dose: 10 mg
- Serious adverse events: 74/395
- Primary safety comparison versus placebo: HR 0.867
Asundexian 20 mg
- Asundexian dose: 20 mg
- Serious adverse events: 82/397
- Primary efficacy and safety comparisons versus placebo were posted.
Asundexian 50 mg
- Asundexian dose: 50 mg
- Serious adverse events: 67/402
- Primary efficacy and safety comparisons versus placebo were posted.
Placebo
- BAY2433334 matching placebo
- Serious adverse events: 82/399
- Used as the comparator for the posted dose-specific analyses.
4. Analysis Populations
The posted primary efficacy analyses used the FAS, while the primary safety analyses used the SAF. The registry does not provide expanded definitions of these population abbreviations in the statistical-analysis records shown here.
| Population | Role in posted analyses |
|---|---|
| FAS | Analysis population for the primary efficacy endpoint and secondary cardiovascular efficacy endpoints. |
| SAF | Analysis population for the primary BARC bleeding safety endpoint. |
5. Primary Endpoints
| Endpoint | Registry definition | Time frame | Statistical treatment |
|---|---|---|---|
| Efficacy - Number of Participants With Composite of CV Death, MI, Stroke and Stent Thrombosis (ST) | Composite of cardiovascular death, myocardial infarction, stroke, and stent thrombosis. Cardiovascular death included death due to stroke, MI, heart failure or cardiogenic shock, sudden death, or other cardiovascular causes. Death due to non-traumatic hemorrhage was included. MI was defined using evidence of myocardial necrosis in a clinical setting consistent with acute myocardial ischemia. Stroke was defined as an acute episode of focal or global neurological dysfunction caused by injury of the brain, spinal cord, or retina from hemorrhage or infarction. Stent thrombosis incorporated diagnostic certainty and timing, including definite and probable ST definitions. | From baseline up to 52 weeks | Time-to-event analysis using log-rank testing and Cox proportional-hazards estimates |
| Safety - Number of Participants With BARC Bleeding Definition Type 2, 3 and 5 | BARC type 2, 3, and 5 bleeding. Type 2 includes overt, actionable hemorrhage requiring nonsurgical medical intervention, hospitalization or increased level of care, or prompt evaluation. Type 3 includes specified hemoglobin decreases, transfusion with overt bleeding, major bleeding requiring intervention, cardiac tamponade, intracranial hemorrhage, or intraocular bleeding compromising vision. Type 5 is fatal bleeding, including probable and definite fatal bleeding. | From baseline up to 52 weeks | Time-to-event analysis using log-rank testing and hazard ratios |
6. Statistical Methodology
Time-to-event analysis
Although the registered endpoint type is classified as binary, the posted statistical analyses treat the primary outcomes as time-to-event endpoints. This distinction is important. A participant is not simply classified according to whether an event occurred during a fixed period; the analysis incorporates the timing of the event and censoring information.
Log-rank test
The registry reports the log-rank test for the formal primary comparisons. The log-rank test compares the observed pattern of event occurrence between treatment groups over follow-up. It is especially natural when the endpoint is a time-to-event outcome and participants can have different amounts of observed follow-up.
Cox proportional-hazards effect measure
The reported effect measure for the primary analyses is a hazard ratio obtained from a Cox proportional-hazards framework. A hazard ratio compares the estimated instantaneous event rates between groups within the model.
An HR of 1 corresponds to equal modeled hazards. An HR below 1 corresponds to a lower estimated instantaneous event rate in the numerator group, while an HR above 1 corresponds to a higher estimated instantaneous event rate.
Competing event specification
For the primary efficacy composite, the registry states that non-CV death was the competing event for events that include CV death. This matters because a participant who dies from a non-cardiovascular cause cannot subsequently experience one of the cardiovascular events included in the composite.
Event-count requirement for hazard ratios
The registry states that hazard ratios were only calculated if at least three events occurred in one of the compared groups and at least one event occurred in each of the compared treatment groups. This is a useful reminder that a hazard ratio is not automatically generated for every comparison when event counts are sparse.
Confidence intervals
All posted primary analyses use two-sided 90% confidence intervals. The confidence interval describes statistical uncertainty around the estimated hazard ratio under the corresponding analysis framework. It does not describe the range of effects that individual patients will experience.
7. Primary Efficacy Results
The primary efficacy endpoint was the number of participants with the composite of cardiovascular death, myocardial infarction, stroke, and stent thrombosis from baseline up to 52 weeks. The posted analyses compared the 20 mg and 50 mg dose groups with placebo.
Asundexian 20 mg and 50 mg vs Placebo
Composite cardiovascular efficacy endpoint
90% CI: 0.687–1.612 · P = 0.8439
FAS · Log-rank test · Cox proportional-hazards estimate
The estimated hazard ratio of 1.052 indicates that the modeled hazard for the combined Asundexian 20 mg and 50 mg comparison relative to placebo was estimated at approximately 1.052 times the placebo hazard under the reported Cox analysis.
The estimate does not mean that 1.052 times as many participants experienced the composite, nor does it represent an absolute increase in event probability. A hazard ratio is a relative time-to-event measure.
The two-sided 90% confidence interval of 0.687–1.612 is fairly broad and spans 1.0. It therefore includes values corresponding to a lower modeled hazard and values corresponding to a higher modeled hazard. The interval is more informative about statistical precision than the point estimate alone.
The reported P = 0.8439 addresses the statistical evidence against the null comparison used for the test; it does not measure the size or clinical importance of the estimated effect. The registry does not state a specific hypothesis type or non-inferiority margin for this comparison.
Asundexian 20 mg vs Placebo
Composite cardiovascular efficacy endpoint
90% CI: 0.674–1.781 · P = 0.7562
FAS · Log-rank test · Cox proportional-hazards estimate
The hazard ratio of 1.096 estimates the event hazard in the Asundexian 20 mg group relative to placebo at approximately 1.096 under the reported Cox model. The point estimate alone should not be interpreted as an absolute difference in the probability of experiencing the composite.
The 90% confidence interval of 0.674–1.781 includes 1.0 and extends in both directions. Thus, the reported estimate is compatible with a lower hazard as well as a higher hazard relative to placebo within the uncertainty represented by this interval.
The reported P = 0.7562 is a measure used for the hypothesis test; it is not a measure of how large the effect is. The confidence interval is needed to understand the precision of the estimate.
Because this is a Cox hazard ratio, interpretation also depends on the suitability of the proportional-hazards framework. The registry does not report a separate assessment of that assumption.
Asundexian 50 mg vs Placebo
Composite cardiovascular efficacy endpoint
90% CI: 0.614–1.656 · P = 0.978
FAS · Log-rank test · Cox proportional-hazards estimate
The estimated hazard ratio of 1.008 is close to 1.0, indicating that the modeled event hazard in the Asundexian 50 mg group was estimated to be approximately the same as placebo at the point estimate.
That point estimate should not be mistaken for proof of identical event risk. The 90% confidence interval of 0.614–1.656 is substantially wider than the point estimate and includes both lower and higher hazards relative to placebo.
The reported P = 0.978 does not quantify the size of the observed difference. It is a test statistic expressed as a probability under the relevant null framework.
The registry reports no stated non-inferiority or equivalence hypothesis for this comparison, so the confidence interval should not be interpreted as establishing equivalence merely because it contains 1.0.
| Primary efficacy comparison | HR | 90% CI | P-value | Population |
|---|---|---|---|---|
| Asundexian 20 mg + 50 mg vs Placebo | 1.052 | 0.687–1.612 | = 0.8439 | FAS |
| Asundexian 20 mg vs Placebo | 1.096 | 0.674–1.781 | = 0.7562 | FAS |
| Asundexian 50 mg vs Placebo | 1.008 | 0.614–1.656 | = 0.978 | FAS |
8. Primary Safety Results
The primary safety endpoint was the number of participants with BARC bleeding definition type 2, 3, and 5 from baseline up to 52 weeks. The registry provides pooled and dose-specific comparisons against placebo.
Pooled Asundexian vs Placebo
BARC type 2, 3 and 5 bleeding
90% CI: 0.713–1.347 · P = 0.9158
SAF · Log-rank test · Hazard ratio
The pooled Asundexian hazard ratio of 0.98 estimates the instantaneous hazard of BARC type 2, 3, or 5 bleeding at approximately 0.98 times the placebo hazard under the reported analysis.
The estimate is close to 1.0, but the 90% confidence interval of 0.713–1.347 includes both lower and higher hazards. The interval therefore communicates uncertainty that is not captured by the point estimate alone.
The reported P = 0.9158 is evidence about the statistical comparison, not a measure of bleeding risk magnitude or clinical importance.
No non-inferiority margin is stated in the registry analysis record. Consequently, this comparison should not be described as demonstrating non-inferiority or equivalence.
Asundexian 10 mg vs Placebo
BARC type 2, 3 and 5 bleeding
90% CI: 0.577–1.302 · P = 0.5633
SAF · Log-rank test · Hazard ratio
The point estimate of 0.867 corresponds to a modeled bleeding hazard approximately 13.3% lower than placebo at the point estimate. This is a relative hazard interpretation, not a statement that 13.3% fewer participants bled.
The 90% confidence interval of 0.577–1.302 includes 1.0 and spans a range from a potentially lower hazard to a potentially higher hazard.
The P = 0.5633 does not tell us that the estimated difference is clinically small or large. The point estimate and confidence interval provide the effect-size information, while the p-value addresses the statistical test.
Asundexian 20 mg vs Placebo
BARC type 2, 3 and 5 bleeding
90% CI: 0.587–1.306 · P = 0.584
SAF · Log-rank test · Hazard ratio
The estimated hazard ratio of 0.875 corresponds to a modeled bleeding hazard approximately 12.5% lower than placebo at the point estimate.
The 90% confidence interval of 0.587–1.306 includes 1.0, so the observed point estimate does not establish a directional difference by itself. The width of the interval also shows that the estimate has meaningful statistical uncertainty.
The reported P = 0.584 should not be interpreted as a 58.4% probability that the treatment is ineffective or as a probability that the hazard ratio equals 1.0.
Asundexian 50 mg vs Placebo
BARC type 2, 3 and 5 bleeding
90% CI: 0.828–1.747 · P = 0.417
SAF · Log-rank test · Hazard ratio
The point estimate of 1.202 corresponds to an estimated bleeding hazard approximately 20.2% higher than placebo under the reported Cox model. This is a relative hazard estimate and is not an absolute increase in bleeding probability.
The 90% confidence interval of 0.828–1.747 includes 1.0 and extends in both directions. Therefore, the uncertainty interval encompasses both a lower and a higher hazard relative to placebo.
The reported P = 0.417 is not an effect-size measure. The confidence interval and hazard ratio together provide the more informative description of the estimated magnitude and precision.
| Primary safety comparison | HR | 90% CI | P-value | Population |
|---|---|---|---|---|
| Pooled Asundexian vs Placebo | 0.98 | 0.713–1.347 | = 0.9158 | SAF |
| Asundexian 10 mg vs Placebo | 0.867 | 0.577–1.302 | = 0.5633 | SAF |
| Asundexian 20 mg vs Placebo | 0.875 | 0.587–1.306 | = 0.584 | SAF |
| Asundexian 50 mg vs Placebo | 1.202 | 0.828–1.747 | = 0.417 | SAF |
9. Secondary Efficacy Results
The registry also posts analyses of cardiovascular death, myocardial infarction, and all-cause mortality. The posted records use hazard ratios with two-sided 90% confidence intervals. Where the registry does not report a formal analysis method, that omission is preserved below.
Cardiovascular Death
| Comparison | HR | 90% CI | Formal method reported |
|---|---|---|---|
| Asundexian 20 mg + 50 mg vs Placebo | 1.275 | 0.322–5.050 | Not reported |
| Asundexian 50 mg vs Placebo | 1.991 | 0.479–8.273 | Not reported |
The cardiovascular-death estimates have notably wide confidence intervals. That is important statistically because the point estimates alone provide little information about precision. The registry again states that hazard ratios were calculated only when at least three events occurred in one compared group and at least one event occurred in each compared treatment group. Non-CV death was treated as the competing event for events that include cardiovascular death.
Myocardial Infarction
| Comparison | HR | 90% CI | Formal method reported |
|---|---|---|---|
| Asundexian 20 mg + 50 mg vs Placebo | 1.125 | 0.696–1.819 | Not reported |
| Asundexian 20 mg vs Placebo | 1.181 | 0.686–2.031 | Not reported |
| Asundexian 50 mg vs Placebo | 1.070 | 0.613–1.866 | Not reported |
Each myocardial-infarction confidence interval includes 1.0. The intervals also extend considerably on either side of the point estimates, illustrating why a hazard ratio should be interpreted together with its confidence interval rather than in isolation.
All-Cause Mortality
| Comparison | HR | 90% CI | P-value | Method |
|---|---|---|---|---|
| Asundexian 20 mg + 50 mg vs Placebo | 1.266 | 0.603–2.658 | = 0.6016 | Log-rank |
| Asundexian 20 mg vs Placebo | 0.996 | 0.414–2.401 | = 0.9945 | Log-rank |
| Asundexian 50 mg vs Placebo | 1.506 | 0.667–3.405 | = 0.4085 | Log-rank |
For all-cause mortality, the registry explicitly states that there is no competing event. Death from any cause is itself the event of interest, so a participant cannot experience a separate competing cause of death after death has occurred.
10. Safety: Serious Adverse Events by Arm
The registry reports serious adverse events by treatment arm as affected participants divided by participants at risk.
| Arm | Serious adverse events | Participants at risk |
|---|---|---|
| Asundexian 10 mg | 74 | 395 |
| Asundexian 20 mg | 82 | 397 |
| Asundexian 50 mg | 67 | 402 |
| Placebo | 82 | 399 |
These serious-adverse-event counts are descriptive arm-level safety information. They should not be substituted for the primary BARC bleeding time-to-event analysis because the endpoint definition, follow-up structure, and statistical method differ.
11. Statistical Methods Explained
Why was a log-rank test used?
The primary endpoints were analyzed as time-to-event outcomes. A log-rank test compares the observed event-time experience between treatment groups across follow-up rather than reducing every participant to a simple yes/no classification at week 52. This makes it appropriate for outcomes where the timing of an event is informative.
What does a hazard ratio of 1.008 mean?
For the primary efficacy comparison of Asundexian 50 mg versus placebo, the estimated hazard ratio was 1.008. In the Cox model, this means the estimated instantaneous event hazard was approximately 1.008 times that of placebo. It does not mean that 1.008 times as many participants experienced an event, and it does not provide an absolute risk difference.
Why is the confidence interval important?
A point estimate is only one possible estimate under repeated sampling. The 90% confidence interval shows the uncertainty around that estimate under the specified statistical framework. For example, the primary 50 mg efficacy estimate of 1.008 has a 90% CI of 0.614–1.656, showing considerably more uncertainty than the point estimate alone would suggest.
Why does the p-value not measure effect size?
A p-value summarizes the evidence against a specified null hypothesis under the statistical model and testing framework. It does not say how large an effect is. Two analyses can have similar p-values with different effect sizes, and a large effect estimate can have substantial uncertainty when relatively few events are observed.
What does the competing-event statement mean?
For the cardiovascular composite and cardiovascular-death analyses, the registry states that non-CV death is the competing event. A participant who dies from a non-cardiovascular cause can no longer experience a later cardiovascular event. Recognizing that event structure is important when interpreting survival analyses.
Why does the analysis population matter?
The primary efficacy analyses use the FAS, whereas the primary safety analyses use the SAF. An analysis population defines which participants contribute to the statistical estimate. This is one reason that efficacy and safety results should not automatically be interpreted as though they came from an identical population.
What should be inferred from a confidence interval that crosses 1?
For a hazard ratio, 1 represents equal hazards. A confidence interval that crosses 1 means that the interval includes both directions of relative effect under the corresponding confidence framework. It does not prove that the true hazard ratio is exactly 1, nor does it by itself establish equivalence or non-inferiority.
12. Multiplicity and Multiple Comparisons
The registry posts multiple primary analyses: three efficacy comparisons and four safety comparisons. It also posts multiple secondary endpoint analyses. The statistical-analysis records identify the hypothesis type as Other / not stated and do not provide a multiplicity-adjustment procedure in the registry-reported analysis fields.
13. Non-Inferiority and Equivalence Considerations
The registry's statistical-analysis records do not state a non-inferiority or equivalence hypothesis type for the primary efficacy or safety comparisons, and no non-inferiority margin is provided in the analysis records.
This distinction matters because an estimate whose confidence interval includes 1 is not automatically evidence of equivalence. A non-inferiority conclusion requires a prespecified margin defining how much loss of efficacy or increase in harm would be considered acceptable, together with an analysis appropriate to that design.
14. Proportional-Hazards Interpretation
The primary analyses use Cox proportional-hazards effect measures. Under that framework, the hazard ratio is most naturally interpreted as a relative hazard that is assumed to be adequately summarized by the model over the analyzed period.
HR = 1 → equal modeled hazards
HR > 1 → higher modeled event hazard
The registry does not report a separate diagnostic or formal test of the proportional-hazards assumption in the posted statistical-analysis records.
A single hazard ratio also does not tell the reader whether the absolute event probabilities differ by a particular amount at week 52. Absolute risk measures and Kaplan-Meier estimates would provide a different perspective, but the registry information shown here does not report Kaplan-Meier estimates or median event times for these endpoints.
15. Kaplan-Meier Estimation and Censoring
For a time-to-event endpoint such as the PACIFIC-AMI primary composite, Kaplan-Meier estimation is a standard way to describe the event-free survival function over time. The registry's posted analysis fields identify log-rank testing and Cox proportional-hazards estimates but do not report Kaplan-Meier estimates or median event-free survival.
Here, di represents the number of events at event time ti, while ni represents the number at risk immediately before that time.
Censoring allows participants who have not experienced the endpoint by their last observed follow-up to contribute information before their censoring time. The quality of a survival analysis therefore depends not only on the number of events but also on how follow-up and censoring are handled.
16. Endpoint Construction and Clinical Meaning
Composite efficacy endpoint
The primary efficacy endpoint combines cardiovascular death, MI, stroke, and stent thrombosis. A participant can therefore meet the endpoint through different clinical event types.
Stent thrombosis
The registered definition distinguishes definite and probable stent thrombosis and incorporates diagnostic certainty and timing.
BARC bleeding
The primary safety endpoint includes BARC type 2, 3, and 5 bleeding, ranging from actionable overt bleeding through major and fatal bleeding categories.
Time frame
Both primary endpoints use the same registry time frame: from baseline up to 52 weeks.
Composite endpoints can improve statistical efficiency by accumulating more events, but they also require careful clinical interpretation because the component events can differ substantially in severity and frequency. The registry does not provide component-specific event counts for the primary composite in the statistical-analysis records shown here.
17. Secondary Endpoint Interpretation
The secondary analyses illustrate why the uncertainty interval is particularly important for relatively uncommon events. The cardiovascular-death analysis for Asundexian 50 mg versus placebo has an HR of 1.991 with a 90% CI of 0.479–8.273. The point estimate is above 1, but the interval is sufficiently wide to encompass substantially lower as well as substantially higher hazards.
Similarly, the all-cause mortality estimate for Asundexian 20 mg versus placebo is 0.996, with a 90% CI of 0.414–2.401. The point estimate is very close to 1, but the interval demonstrates that considerable uncertainty remains around that estimate.
18. Trial Timeline
Trial start
The registry lists June 17, 2020 as the study start date.
Primary completion
The registry lists February 21, 2022 as the primary completion date.
Registry status
The registry lists the trial status as completed.
19. What the Primary Hazard Ratios Do — and Do Not — Mean
A hazard ratio compares modeled instantaneous event rates between two groups. For example, the primary efficacy HR of 1.096 for Asundexian 20 mg versus placebo corresponds to an estimated hazard approximately 1.096 times that of placebo under the reported Cox analysis.
It does not mean that 9.6% more participants experienced the composite, nor does it mean that each participant had exactly a 9.6% higher risk.
The 90% CI of 0.674–1.781 surrounding the 20 mg efficacy estimate shows that the point estimate has substantial uncertainty. A confidence interval should therefore be interpreted together with the point estimate rather than used as an afterthought.
The p-value provides information about the statistical test associated with the comparison. It does not measure effect magnitude, clinical importance, or the probability that the null hypothesis is true.
The primary safety estimates should be interpreted separately from efficacy. For example, the pooled Asundexian safety HR is 0.98, but that estimate concerns BARC type 2, 3, and 5 bleeding rather than the cardiovascular efficacy composite.
20. Important Limitations
- Registry-level reporting: the registry reports hazard ratios and confidence intervals but does not provide Kaplan-Meier estimates or median event times for the primary endpoints in the statistical-analysis records shown here.
- Analysis populations: primary efficacy analyses use FAS, whereas primary safety analyses use SAF. These are not necessarily identical populations.
- Multiple comparisons: several primary and secondary comparisons are posted, while the analysis records do not state a detailed multiplicity-adjustment strategy.
- Uncertainty for secondary endpoints: cardiovascular-death and mortality estimates have wide confidence intervals, indicating limited precision for some event-specific comparisons.
- Hazard-ratio assumptions: Cox proportional-hazards estimates depend on the adequacy of the model. The registry does not report a separate assessment of proportional hazards.
- Competing events: non-CV death is identified as a competing event for events that include cardiovascular death.
- Non-inferiority interpretation: no non-inferiority margin is reported in the statistical-analysis records, so confidence intervals crossing 1 should not be interpreted as establishing equivalence.
- Composite endpoint: the primary efficacy endpoint combines several clinically distinct event types, so an overall hazard ratio does not describe the effect on each component separately.
21. Why This Trial Matters Statistically
PACIFIC-AMI is a useful statistical teaching case because the registry combines randomized dose comparisons with time-to-event efficacy and safety endpoints. It also illustrates how the same general survival-analysis framework can be applied to cardiovascular efficacy and bleeding outcomes while using different analysis populations and endpoint definitions.
| Concept | How it appears in PACIFIC-AMI |
|---|---|
| Randomization | Randomized allocation across four parallel treatment arms |
| Blinding | Quadruple-masked design |
| Time-to-event endpoints | Primary efficacy and safety outcomes analyzed using survival-analysis methods |
| Log-rank test | Formal method reported for the primary comparisons and some secondary analyses |
| Hazard ratio | Primary and secondary treatment-effect measure |
| Confidence interval | Two-sided 90% intervals reported for the posted estimates |
| Competing events | Non-CV death identified as the competing event for events that include CV death |
| Analysis populations | FAS for efficacy and SAF for primary safety analyses |
| Composite endpoint | CV death, MI, stroke, and stent thrombosis combined into one primary efficacy endpoint |
| Safety endpoint | BARC type 2, 3, and 5 bleeding |
22. Related Tutorials
Learn more about the methods used in this trial:
23. Related Calculators
24. Sources
- ClinicalTrials.gov: PACIFIC-AMI, NCT04304534.
- PubMed: Publication indexed under PMID 38038437.
- PubMed: Publication indexed under PMID 36030390.
Continue through the Clinical Biostats statistical library
Explore deeper explanations of survival analysis, clinical-trial methods, and statistical calculations related to randomized time-to-event studies.
25. Record Summary
PACIFIC-AMI provides a detailed example of randomized cardiovascular trial analysis using time-to-event methods. The registry reports a four-arm, quadruple-masked phase 2 design with 1,601 participants and primary endpoints covering both cardiovascular efficacy and BARC bleeding safety through 52 weeks. The primary efficacy comparisons used FAS and produced hazard ratios of 1.052 for Asundexian 20 mg and 50 mg versus placebo, 1.096 for Asundexian 20 mg versus placebo, and 1.008 for Asundexian 50 mg versus placebo. The primary safety comparisons used SAF, with hazard ratios of 0.98 for pooled Asundexian, 0.867 for 10 mg, 0.875 for 20 mg, and 1.202 for 50 mg versus placebo.
The central statistical lesson is that these hazard ratios must be interpreted together with their confidence intervals, p-values, endpoint definitions, analysis populations, and time-to-event structure. The secondary cardiovascular-death, myocardial-infarction, and all-cause-mortality analyses further demonstrate how sparse or relatively infrequent events can generate wide uncertainty intervals even when a point estimate appears numerically distinctive.