Pharmacodynamic Effect of Prasugrel vs. Ticagrelor in Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pharmacodynamic Effect of Prasugrel vs. Ticagrelor in Diabetes (NCT01852214), a randomized na study. The primary outcome is P2Y12 Reaction Units.

2. Clinical Question

The clinical question was whether platelet reactivity, measured as P2Y12 Reaction Units (PRU), differed after treatment with prasugrel versus ticagrelor in people with type 2 diabetes mellitus and coronary artery disease. The registry identifies PRU measured by VerifyNow as the primary endpoint and gives a timeframe of 1 week. It describes treatment effects as comparisons of PRU after each treatment, regardless of the order in which participants received the treatments.

This is a pharmacodynamic question: it concerns a measured platelet-function response, not a direct comparison of clinical events. The registry background describes diabetes as associated with increased risk of adverse atherothrombotic events and notes reduced response to some oral antiplatelet medication, particularly clopidogrel. It also describes prasugrel and ticagrelor as P2Y12 receptor inhibitors and states that head-to-head comparisons were lacking. Those statements explain the motivation for comparing the two treatments; they are not results of this trial.

The posted treatment-specific least-squares mean PRU was 83.0 for prasugrel, with a 95% confidence interval from 63.0 to 103.0, and 52.0 for ticagrelor, with a 95% confidence interval from 32.0 to 72.0. The registry reports a mixed models analysis with p = 0.022. The question answered by these data is therefore about a measured laboratory endpoint under the study conditions. The registry does not establish whether any PRU difference translates into a difference in clinical outcomes.

3. Trial Design

The registry describes a randomized, double-masked, crossover trial comparing prasugrel with ticagrelor. Enrollment is listed as 50. In a crossover design, participants receive more than one treatment in sequence, so treatment comparisons can draw on measurements from the same participants under different treatments. That design differs from a parallel-group trial, in which participants remain in one assigned group. The registry lists sequences beginning with prasugrel or ticagrelor.

The treatment descriptions specify a 1-week treatment period (7±2 days) and a washout period of 2-4 weeks before the next treatment period. The descriptions in the registry are truncated, so details beyond the text that is recorded are not reported in the registry record. The record identifies prasugrel and ticagrelor as the study interventions. It does not provide enough information to reconstruct further dosing or sequence details.

The primary outcome is PRU, measured by VerifyNow, with a 1-week timeframe. The analysis population is defined as randomized subjects who received study drug, successfully completed at least one treatment period, and had valid data for the primary endpoint. The treatment-specific registry summaries list 45.0 observations for prasugrel and 46.0 for ticagrelor. These are endpoint summary counts, not a count of separate participants across the crossover trial.

The registry lists the trial as completed and identifies the University of Florida as sponsor. It does not report the phase beyond “NA.” Further information about concealment of allocation, how masking was maintained, or the conduct of the crossover periods is not reported in the registry record. The trial design gives a framework for interpreting the pharmacodynamic comparison, but the posted summaries do not provide individual participant measurements or sequence-specific estimates.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: 11 subjects were excluded before randomization: withdrawn of consent (n=4), screen failure (n=4), unable to draw blood (n=3).

Period 1

Reasons for not completing this period:

Washout Period

Reasons for not completing this period:

Period 2

Analysis population for the primary outcome, as defined in the registry: All analyses of platelet function conducted on all randomized subjects who received study drug, successfully completed at least one treatment period of the study and had valid data for the primary end point.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: All analyses of platelet function were conducted on the pharmacodynamic population, which was defined as all randomized subjects who received study drug, successfully completed at least one treatment period of the study and had valid data for the primary end point (n=46)

6. Primary Endpoint

The registry lists the primary outcome as P2Y12 Reaction Units, measured over 1 week.

The primary endpoint is P2Y12 Reaction Units (PRU), measured by the VerifyNow assay. It is a continuous measure of platelet reactivity. The registry describes the endpoint as the comparison of PRU values after ticagrelor and prasugrel, with treatment effects evaluated regardless of sequence. The listed timeframe is 1 week.

As a general statistical point, a continuous endpoint retains the magnitude of measured values rather than reducing them to categories such as above or below a threshold. A treatment comparison can therefore describe how the measured response differs between conditions. PRU is a pharmacodynamic measure: it captures a platelet-function response to treatment. It is not itself a count of ischemic events, bleeding events, or deaths.

The registry reports least-squares means and 95% confidence intervals for each treatment. A least-squares mean is a model-based mean; it may reflect the statistical model used rather than a simple arithmetic average. The registry does not provide enough model detail to explain which covariates or other terms contributed to these estimates. The intervals describe uncertainty around each treatment-specific mean, not a confidence interval for the treatment contrast.

The endpoint is relevant to the trial’s stated aim of comparing pharmacodynamic effects. Its interpretation should remain within that scope: the posted PRU summaries describe laboratory measurements after the study treatments. Whether a difference in PRU corresponds to a difference in patient-important outcomes is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the reported mean describe?

The registry identifies the endpoint parameter as a least-squares mean. In general, a least-squares mean is an estimate generated from a statistical model. It can differ from a raw arithmetic mean when the model accounts for features of the design or data. Here, the registry reports 83.0 PRU for prasugrel and 52.0 PRU for ticagrelor. It does not state which covariates, sequence terms, period terms, or other specifications were included in the model that produced those means. Those details are not reported in the registry record.

The attached 95% confidence intervals are 63.0 to 103.0 for prasugrel and 32.0 to 72.0 for ticagrelor. In general, a confidence interval conveys uncertainty in an estimate under the assumptions of its statistical procedure. It is not a range that guarantees the location of the true value, and it does not directly give the uncertainty around the difference between the two treatment means. The record does not report a confidence interval for the treatment contrast.

Why use a mixed model?

The registry names a mixed models analysis for the treatment comparison. In general, mixed models can represent both fixed effects, such as a treatment comparison, and random effects that account for clustering or repeated measurements. In a crossover study, measurements from the same participant can be correlated because each participant may receive both treatments. A model that accounts for this structure can be more appropriate than treating all measurements as independent.

The term “mixed models analysis” alone does not reveal exactly how the analysis was performed. The registry does not report the full model formula, covariance structure, adjustment variables, handling of period or sequence, or diagnostics. It also does not report whether the treatment comparison was adjusted for any other factors. These details should not be inferred from the method name.

How should the p-value be read?

The reported mixed-model comparison has p = 0.022. In general, a p-value is calculated relative to a null hypothesis and a statistical model. It summarizes how unusual data at least as inconsistent with that null model would be if the model assumptions and null hypothesis held. It does not provide the probability that the treatments are equivalent, nor does it communicate the size of a difference. Interpretation depends on the prespecified analysis and assumptions, details of which are not reported in the registry record.

The p-value is the registry’s inferential comparison, while the two treatment-specific intervals describe uncertainty around their respective means. One should not derive a contrast test from the mean intervals by visual inspection. In general, paired or repeated-measure designs can have a contrast precision that depends on within-participant association, not just on the separate uncertainty around each treatment mean.

What does crossover add to the interpretation?

In general, a crossover design allows within-participant comparison because a participant receives multiple treatments in sequence. This can account for stable differences between participants, but valid interpretation also depends on the conduct and analysis of periods and sequences. Washout periods are intended to separate treatment periods, but a registry description of a washout does not by itself establish that carryover effects were absent. The record describes a washout period of 2-4 weeks and treatment periods of 1 week (7±2 days); it does not report a carryover assessment or sequence-specific results.

The primary treatment effect is described as comparing PRU after prasugrel and ticagrelor regardless of sequence. A mixed model is named, but whether sequence, period, or carryover was included cannot be determined from the available registry description. General concerns about crossover analysis should therefore be distinguished from claims about this particular analysis: the relevant model details are not reported in the registry record.

What does the analysis population mean?

The stated primary analysis population consists of randomized subjects who received study drug, successfully completed at least one treatment period, and had valid primary-endpoint data. This is a defined subset based on treatment receipt, period completion, and endpoint availability. It is not described as an analysis of every randomized subject regardless of treatment received or missing outcome data. The registry does not report how many randomized subjects did not meet these criteria or why any measurements were unavailable.

In general, exclusions based on post-randomization events or outcome availability can affect interpretation if those exclusions are related to treatment or endpoint values. The registry does not provide enough information to assess that possibility. It also does not describe imputation, sensitivity analyses, or a separate intention-to-treat analysis.

10. Analyses Reported by the Sponsor

The registry reports the sponsor’s comparison as a mixed models analysis with p = 0.022. It also reports treatment-specific least-squares means and their 95% confidence intervals. The treatment-specific summaries provide context for the direction of the observed values, while the mixed-model p-value is the inferential comparison recorded by the sponsor.

The registry does not report a contrast estimate or confidence interval for the difference between treatments. It also does not provide a separately reported, estimable two-arm recalculation for this continuous endpoint. Accordingly, there is no recalculated treatment-comparison estimate to set alongside the sponsor’s model result. The registry’s stated method and p-value should be described as the reported analysis, not replaced by a comparison inferred from the treatment-specific means or their intervals.

The p-value alone does not quantify the size or clinical importance of the contrast. The model specification and the assumptions underlying it are not reported in the registry record, which limits independent assessment of the analysis. The registry does not flag a discrepancy between the sponsor result and a recalculated result.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists three secondary outcomes: P2Y12 Reaction Units and Platelet Reactivity Index, with Platelet Reactivity Index appearing twice in the listing. It records four outcome measures and two analyses. The registry does not explain how these entries relate to distinct analyses or specify a multiplicity adjustment. Whether the primary comparison was treated as the sole confirmatory test, or how secondary comparisons were interpreted in relation to it, is not reported in the registry record.

In general, testing multiple outcomes or comparisons can increase the chance of at least one apparently notable result when each is assessed without accounting for the set of tests. Methods such as adjusted significance thresholds or hierarchical testing can address multiplicity, but no such approach should be attributed to this trial without documentation. The reported p = 0.022 is the posted result; the registry does not state whether it was adjusted for other tests.

The primary analysis population is defined as randomized subjects who received study drug, completed at least one treatment period, and had valid primary-endpoint data. The treatment summaries list 45.0 observations for prasugrel and 46.0 for ticagrelor. These counts do not explain which participants or measurements were missing, whether participants contributed data to both treatment conditions, or why the counts differ. A detailed participant flow and reasons for missing endpoint data are not reported in the registry record.

Missing-data methods, imputation procedures, and sensitivity analyses are not reported. The registry also does not state whether a full intention-to-treat analysis was conducted. In general, analysis restricted to participants with observed outcomes can be affected if the likelihood of having valid data is related to treatment or outcome. The available record does not permit assessment of that possibility.

The registry does not report a statistical analysis plan, interim analyses, or a sample-size justification. Those elements should not be inferred from enrollment or from the observed endpoint summaries. The record’s planning-illustration assumptions are not a substitute for a documented trial design justification.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent non-serious adverse event terms

For serious adverse events, the safety population figures report zero affected in the prasugrel group, with 46 listed at risk, and zero affected in the ticagrelor group, with 49 listed at risk. The registry’s serious-event term list is empty. Death counts are not reported in the registry record.

The registry also reports other adverse events: 4 affected in the prasugrel safety population, with 46 listed at risk, and 10 affected in the ticagrelor safety population, with 49 listed at risk. These are counts as recorded; the registry does not provide a detailed event breakdown in the posted safety information. The serious and other adverse-event figures describe different categories and should not be combined.

In general, a zero count in a trial does not establish that an event cannot occur, and safety comparisons require attention to event definitions, follow-up, exposure, and the amount of information available. The record does not provide details needed to interpret the counts as comparative estimates of risk. In particular, the figures do not establish that one treatment is safer than the other. They are descriptive safety results for the reported populations and should be read separately from the primary pharmacodynamic endpoint.

13. What the Result Does, and Does Not, Mean

The registry supports a focused statistical statement: treatment-specific mean PRU was 83.0 after prasugrel and 52.0 after ticagrelor, and the sponsor reported a mixed-model comparison with p = 0.022. The corresponding 95% confidence intervals for the means were 63.0 to 103.0 and 32.0 to 72.0. These results describe the recorded pharmacodynamic endpoint under the trial’s study conditions.

They do not establish the size of the model-based treatment contrast with the same clarity as a contrast estimate and its confidence interval would. Neither is reported in the registry record. The p-value gives evidence from the reported model but is not an effect-size measure, and it cannot by itself establish clinical importance. The registry does not state a threshold for interpreting PRU differences as clinically meaningful.

The primary endpoint is a laboratory measure, not a clinical outcome. The posted results do not establish whether the observed PRU pattern corresponds to differences in ischemic events, bleeding, or other outcomes important to patients. The safety counts are descriptive and do not provide a definitive comparative safety assessment.

Interpretation is also bounded by the available analysis information. The study used a crossover design, but the registry does not report the complete mixed-model specification, sequence-specific findings, or details of missing-data handling. The primary analysis population required valid endpoint data and at least one completed treatment period. The registry does not report a full account of participants or measurements not included in that population.

Accordingly, the result should be understood as a statistical comparison of PRU in the registry-defined population, not as a conclusion about all people with diabetes or coronary artery disease, all treatment settings, or clinical outcomes. The posted data do not provide a basis for treatment advice or for claims beyond the pharmacodynamic comparison.

14. Why This Trial Matters Statistically

This trial is statistically notable because it compares two treatments using a crossover design and a continuous pharmacodynamic endpoint. In general, crossover designs can support within-participant comparisons, which may reduce the influence of stable differences between participants. They also create analytic requirements: repeated measurements from one participant are related, and sequence and period can matter. The registry identifies a mixed models analysis, but does not provide enough detail to assess how those features were handled.

The registry reports treatment-specific least-squares means and 95% confidence intervals alongside a model-based p-value. This illustrates an important distinction in statistical reporting. Separate intervals around two means are not the same as an interval around their difference, and a p-value does not convey the magnitude or precision of that difference. The absence of a reported contrast estimate and confidence interval limits a fuller account of the treatment comparison.

The endpoint also illustrates the boundary between a biological or pharmacodynamic measure and a clinical outcome. PRU is the measured primary endpoint, while the registry background discusses atherothrombotic risk and antiplatelet response. The trial’s posted primary result answers a question about measured platelet reactivity; it does not, by itself, answer whether clinical events differ.

Finally, the analysis population and safety summaries show why denominators and outcome definitions matter. The primary analysis required study-drug receipt, at least one completed treatment period, and valid endpoint data. Safety counts are reported for treatment-specific safety populations. The registry does not provide enough detail to reconstruct all missingness, model assumptions, or event-level safety information. These limits are part of the statistical interpretation, not reasons to treat the reported endpoint and p-value as interchangeable with broader clinical conclusions.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry provides treatment-specific least-squares means and confidence intervals, but not a model-based estimate and confidence interval for the treatment contrast. The sponsor’s mixed-model p-value is reported without the model specification. Covariates, period and sequence terms, covariance structure, and diagnostic checks are not reported in the registry record. This limits independent assessment of the comparison and its precision.

The primary analysis population is restricted to randomized subjects who received study drug, completed at least one treatment period, and had valid primary-endpoint data. The record does not report a detailed participant flow, reasons for unavailable measurements, missing-data methods, or sensitivity analyses. It is therefore not possible to determine from the registry how missing observations may have influenced the result.

The crossover treatment descriptions are truncated. Although they describe 1-week treatment periods (7±2 days) and a 2-4 week washout period, the registry does not report sequence-specific outcome estimates or a carryover assessment. It also does not provide detail sufficient to assess all aspects of treatment-period conduct and masking.

The primary endpoint is PRU, a pharmacodynamic measure. The registry does not report a primary clinical-outcome comparison or establish how a PRU difference relates to clinical outcomes. The safety information is limited to reported counts of serious and other adverse events; event definitions and a detailed breakdown are not reported. Death counts are not reported in the registry record.

Finally, multiplicity procedures, the statistical analysis plan, interim analyses, and sample-size justification are not reported. The registry’s enrollment entry and observed results do not provide those missing methodological details. These limitations constrain conclusions to the posted endpoint summaries and the analysis as described.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Pharmacodynamic Effect of Prasugrel vs. Ticagrelor in Diabetes — NCT01852214.
• Registry record last updated 2016-10-17; status last verified 2016-06.
• Results first posted 2016-10-17.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry describes NCT01852214 as a completed, randomized, double-masked crossover trial sponsored by the University of Florida. It compared prasugrel and ticagrelor in people with type 2 diabetes mellitus and coronary artery disease. Enrollment is listed as 50.

The primary endpoint was PRU measured by VerifyNow at 1 week. The registry reports least-squares means of 83.0 for prasugrel and 52.0 for ticagrelor, with respective 95% confidence intervals of 63.0 to 103.0 and 32.0 to 72.0. The reported mixed models analysis has p = 0.022. The treatment summaries list 45.0 observations for prasugrel and 46.0 for ticagrelor.

The primary analysis population required study-drug receipt, completion of at least one treatment period, and valid primary-endpoint data. The record also reports safety counts, but does not provide a detailed event breakdown or death counts. Model specifications, missing-data methods, multiplicity procedures, and a contrast confidence interval are not reported in the registry record.