A Study to Understand the Effect of Low-Fat and High-Fat Meals on the Medicine Called PF-07284890 in Healthy Adults: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Understand the Effect of Low-Fat and High-Fat Meals on the Medicine Called PF-07284890 in Healthy Adults (NCT05349864), a randomized phase 1 study. The primary outcome is Area Under the Plasma Concentration-Time Profile From Time Zero to the Time of the Last Quantifiable Concentration (AUClast) of PF-07284890 in Plasma, Comparison of Low-Fat Meal With Fasted Condition.

2. Clinical Question

3. Trial Design

This was a phase 1, open-label, randomized, single-dose, two-sequence, three-period crossover study. Twelve healthy participants were enrolled. In a crossover design, participants receive more than one treatment condition in separate periods, so the comparison can be made using measurements from the same participants under different conditions. The registry describes two randomized sequences and three treatment periods.

In Sequence 1, participants received PF-07284890 while fasting in Treatment A, then received it after a low-fat meal five days later in Treatment B, and after a high-fat meal five days later in Treatment C. In Sequence 2, the low-fat meal condition came first, followed five days later by the fasted condition, then five days later by the high-fat meal condition. Each treatment was described as a single dose. The registry does not report further details about the rationale for the sequences or the interval between periods beyond the stated five-day spacing.

The study was open-label, meaning masking was not used. That is a design feature, although the primary outcome was a pharmacokinetic measurement rather than a subjective rating. The registry record gives the study start as 2022-06-17, primary completion as 2022-09-16, and status as completed. Participants were required to stay in the clinical research unit for 15 days and 14 nights, according to the brief summary.

The primary comparison concerned low-fat meal versus fasted administration, using AUClast in plasma. The high-fat meal condition was part of the treatment sequence, but a result for a high-fat meal comparison is not reported in the registry record. The record lists 14 outcome measures and 6 analyses. It does not report a sample-size justification, detailed statistical analysis plan, or further design rationale.

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.

Period 1 (5 Days)

Period 2 (5 Days)

Reasons for not completing this period:

Period 3 (5 Days)

Analysis population for the primary outcome, as defined in the registry: The analysis population included all participants randomly assigned to study intervention and who took at least 1 dose of study intervention and in whom at least 1 of the PK parameters of interest could be reported.

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: Baseline analysis population included all participants assigned to the study intervention.

6. Primary Endpoint

The registry lists the primary outcome as Area Under the Plasma Concentration-Time Profile From Time Zero to the Time of the Last Quantifiable Concentration (AUClast) of PF-07284890 in Plasma, Comparison of Low-Fat Meal With Fasted Condition, measured over 0, 0.5, 1, 1.5, 2, 2.5, 3, 4, 6, 8, 12, 16, 24, 36, 48, and 72 hours post-dose of PF-07284890 in Periods 1 and 2.

The primary endpoint was AUClast, the area under the plasma concentration-time profile from time zero to the time of the last quantifiable concentration. It summarizes measured drug concentration in plasma over the observed period, using the concentration-time measurements listed at 0, 0.5, 1, 1.5, 2, 2.5, 3, 4, 6, 8, 12, 16, 24, 36, 48, and 72 hours after dosing in Periods 1 and 2. The registry says AUClast was determined using the linear/Log trapezoidal method.

Exposure measures such as AUClast are useful in pharmacokinetic studies because they capture the overall concentration profile over a defined observation period, rather than relying on a single concentration measurement. The registry reports geometric means and geometric coefficients of variation for each condition, and gives the primary comparison as a ratio of adjusted geometric means. A ratio expresses the low-fat meal condition relative to the fasted condition on that scale.

AUClast does not describe every feature of the concentration profile. Other listed outcomes include time to reach Cmax, terminal elimination half-life, apparent clearance, and apparent volume of distribution. The primary result is therefore about the specified exposure measure and observation window, not a complete account of every pharmacokinetic characteristic or a clinical outcome.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the geometric mean describe?

A geometric mean is a summary suited to measurements that are positive and may vary multiplicatively. Instead of treating equal absolute increments as the central feature of the data, the geometric scale emphasizes relative differences. For a pharmacokinetic quantity such as AUClast, the registry’s choice to report geometric means is consistent with presenting exposure on a scale where ratios are meaningful. The posted values are 28270.0 ng*hr/mL for the low-fat meal condition and 23510.0 ng*hr/mL for the fasted condition.

The ratio of adjusted geometric means compares the conditions on that relative scale. A ratio of 100 would indicate equal adjusted geometric means; a ratio above 100 indicates a larger estimate for the first-named condition relative to the reference, while a ratio below 100 indicates a smaller estimate. The reported ratio is 120.25, with low-fat meal as the test condition and fasting as the reference. This description concerns the scale and direction of the estimate, not the practical or clinical importance of its size.

How should the confidence interval be read?

The 90% confidence interval runs from 109.76 to 131.75. In general, a confidence interval represents uncertainty in an estimate under the statistical procedure used: repeated use of the same procedure in comparable studies would produce intervals that cover the underlying quantity at the stated rate. It does not mean that there is a 90% probability that this particular interval contains the true value. The interval concerns the adjusted ratio, not the range of observed values among participants.

The interval does not include 100, the value corresponding to equal adjusted geometric means. That gives information about the estimate’s precision relative to equality on the ratio scale. The registry does not state a prespecified equivalence margin or another decision threshold, so the interval should not be used to infer a formal equivalence conclusion. No p-value is reported for this analysis.

What does the geometric coefficient of variation add?

The registry reports geometric coefficients of variation of 25.0 for the low-fat meal condition and 38.0 for the fasted condition. These values describe the spread of participant measurements on a geometric scale. They are not confidence limits and do not directly give the uncertainty of the ratio. The ratio’s confidence interval and the condition-specific coefficients of variation answer different questions: one concerns uncertainty in the comparison, and the others describe variation within conditions.

Pharmacokinetic measurements can show multiplicative variability. In general, analyzing data on a logarithmic scale and reporting geometric summaries can make a relative comparison more interpretable. The registry identifies the summary measure and the adjusted ratio, but does not report the full statistical model, transformations, covariance structure, or all estimation details. Those specific methods are not reported in the registry record.

Why does the crossover structure matter?

In a crossover design, the same participant receives more than one condition, allowing comparisons to account for stable differences between participants. Randomizing treatment order can help separate treatment effects from order effects, while the sequences distribute the order of conditions across participants. Here, the registry describes two sequences and three periods, with the low-fat meal, fasted, and high-fat meal conditions arranged as specified for each sequence.

A crossover analysis must also account for the repeated measurements within participants and may need to consider period or sequence effects. Those are general statistical considerations, not claims about what was included in this study’s model. The registry does not report enough model detail to establish which terms or assumptions were used. It also does not report a carryover assessment or specific handling of the five-day intervals.

What can be said about planning illustrations?

Planning illustrations based on an observed effect are not observed power and are not the study’s original sample-size justification. In general, such illustrations depend on assumptions about significance level, allocation, the effect size to detect, and the calculation method. The registry record describes the illustrations’ assumptions as a two-sided significance level of 0.05, equal allocation, and a normal approximation, with effects expressed as fractions of the observed effect. It does not report the trial’s own sample-size justification.

The study enrolled 12 participants, but enrollment alone does not establish whether a design was adequately powered. A planning calculation can illustrate how assumptions affect a hypothetical future design; it cannot change what this completed study observed. Any planning numbers should therefore be read as design illustrations rather than evidence that the trial had a particular probability of detecting an effect.

What is not specified?

The registry does not report the complete statistical analysis plan, any interim analyses, the model specification for the adjusted ratio, or the method used to handle missing pharmacokinetic observations. It also does not give a primary result for the high-fat meal condition. These gaps do not invalidate the posted result, but they limit independent assessment of all analytic choices and the scope of comparisons.

10. Analyses Reported by the Sponsor

The registry reports the sponsor analysis as a ratio of adjusted geometric means of 120.25, with a 90% confidence interval from 109.76 to 131.75. The corresponding posted geometric means are 28270.0 ng*hr/mL for the low-fat meal condition and 23510.0 ng*hr/mL for the fasted condition. The record does not report a p-value for this comparison.

A simple recalculation from condition-level summary statistics would be unadjusted and would not necessarily reproduce an adjusted analysis. In general, adjusted analyses can use model information that is not contained in the two displayed geometric means and their dispersion summaries. The registry does not report a separate recalculated estimate for this comparison, so no numerical comparison between a recalculation and the sponsor’s adjusted result can be made from the registry record.

The record does not flag a discrepancy. The sponsor’s reported estimate and confidence interval are therefore the analysis results to interpret, while recognizing that the full model specification is not reported. A simple unadjusted recalculation, if available, would be a check on the posted summaries rather than a replacement for the sponsor’s analysis.

11. Multiplicity, Missing Data and Analysis Populations

The record lists 14 outcome measures, including the primary AUClast comparison and secondary pharmacokinetic and safety outcomes. Eight secondary outcomes are posted. The registry also lists 6 analyses. It does not report a multiplicity adjustment, a hierarchy of secondary hypotheses, or a detailed testing strategy. Therefore, the record does not establish whether or how the chance of false-positive findings across multiple outcomes or comparisons was controlled.

Multiplicity is a general statistical concern when several outcomes or comparisons are examined. If each is assessed separately, the chance of at least one apparently notable result can increase. A prespecified testing sequence or adjustment can address that issue, but the registry record does not report whether either was used. The primary result should be distinguished from secondary outcomes, and any unreported results should not be inferred from the list of planned measures.

The primary analysis population included participants randomly assigned to intervention who took at least one dose and had at least one of the pharmacokinetic parameters of interest reportable. This definition describes who could contribute to the analysis; it does not provide the number included in the primary analysis or a full accounting of exclusions. The primary outcome arms each list 12 participants, but the record does not give participant-level missingness or explain whether every participant contributed to every time point.

Missing data can matter in a concentration-time analysis because an incomplete profile may affect whether AUClast can be estimated. The registry defines AUClast using the last quantifiable concentration and lists sampling times, but does not report how missing samples, below-quantification values, or incomplete profiles were handled. Those details are not reported in the registry record.

The record also does not report interim analyses, a complete statistical analysis plan, or the trial’s sample-size justification. The listed planning illustrations use general assumptions and are not a substitute for the original design rationale. Without these details, readers cannot determine from the registry alone how multiplicity, missing observations, or prospective power considerations were addressed.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry presents serious and other adverse event counts separately for each treatment period and sequence. The serious-event figures show zero affected participants in every listed group, and the death count is zero in every group. The number at risk is 6 in each fasted and low-fat meal period group, and 5 in each high-fat meal period group. These are period-specific figures, not a combined participant-level summary.

For other adverse events, the registry reports 2 affected participants in the fasted condition in Period 1 of Sequence 1, 1 in the low-fat meal condition in Period 1 of Sequence 2, 2 in the low-fat meal condition in Period 2 of Sequence 1, and 3 in the fasted condition in Period 2 of Sequence 2. In Period 3, the high-fat meal condition has 1 affected participant in each sequence. The at-risk counts for those groups are listed alongside the figures.

These counts should be read with the period and sequence labels intact. Because participants received different conditions across periods, the same person could contribute to more than one period-specific group; the record does not provide participant-level identities or a deduplicated total across conditions. The figures do not establish causation. The registry lists no serious adverse event terms, and the listed counts alone do not describe event type, timing, severity, or clinical context.

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

The posted primary result describes plasma AUClast after a low-fat meal relative to fasting in a small phase 1 crossover study of healthy adults. The adjusted geometric-mean ratio was 120.25, with a 90% confidence interval from 109.76 to 131.75. The estimate and interval indicate the direction and precision of this particular pharmacokinetic comparison under the reported analysis.

The result does not by itself establish whether the observed difference is clinically important. The registry does not report a prespecified threshold for importance or a clinical outcome linked to AUClast. Nor does it support a conclusion about every pharmacokinetic property: the record lists other measures, but their results are not included in the primary comparison. A high-fat meal condition was administered, but its comparative primary result is not reported in the registry record.

The findings are specific to the study conditions and population: healthy participants receiving single doses in a randomized, open-label, three-period crossover design. They should not automatically be generalized to other populations, repeated dosing, or different circumstances. The record does not supply evidence for those settings.

The confidence interval conveys uncertainty in the adjusted comparison, while the geometric coefficients of variation describe dispersion within treatment conditions. Neither measure describes all sources of uncertainty, including uncertainty about model assumptions or missing observations. The model specification and detailed missing-data methods are not reported in the registry record.

The safety counts are also limited. The registry reports zero serious adverse event cases in each group and gives other adverse event counts by period and sequence, but these summaries do not establish cause, provide a full account of event characteristics, or resolve outcomes outside the recorded study. The statistical result is therefore best understood as a focused estimate of exposure under the studied conditions, rather than a broad conclusion about effects or safety.

14. Why This Trial Matters Statistically

This study illustrates how a crossover design can address a pharmacokinetic question when participants can receive multiple treatment conditions. Comparing low-fat meal and fasted exposure within a randomized sequence can reduce the influence of stable differences between participants. The design also requires attention to period, sequence, and repeated-measure structure, making the analysis details important to interpretation.

The primary outcome is reported on a geometric scale and compared as a ratio of adjusted geometric means. That makes the contrast directly interpretable as a relative exposure comparison. The confidence interval communicates uncertainty around that ratio, while the geometric coefficients of variation describe dispersion in each condition. Keeping those summaries distinct helps prevent a spread measure from being mistaken for uncertainty in the treatment comparison.

The record also illustrates the value of separating a primary comparison from a broader set of outcomes. It lists 14 outcome measures and 6 analyses, but does not report a complete multiplicity strategy or a detailed statistical analysis plan. Those omissions matter when readers consider how results across outcomes were selected and interpreted. Likewise, a list of planning assumptions should not be mistaken for the trial’s original sample-size justification or for observed power.

Finally, the difference between an adjusted sponsor estimate and a simple calculation from summary values is methodologically important. A recalculation can serve as a transparent check, but it may not reproduce a model-based estimate. The registry reports the sponsor’s adjusted ratio and interval, while the full model details and a separate recalculated result are not reported. The study is thus a useful example of how the reported effect, its uncertainty, design structure, and reporting gaps all shape statistical interpretation.

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 enrolled sample was 12 healthy participants. The primary endpoint arms each list 12 participants, but the registry does not report a detailed participant flow or explain whether all enrolled participants contributed complete concentration-time profiles. The results therefore need to be read in the context of this study population and the analysis population definition, rather than assumed to represent broader populations.

The study was open-label, and the registry does not report masking. The primary outcome is a plasma concentration measure, but the record does not report additional procedures for laboratory blinding or other steps that might affect measurement. Those details are not reported in the registry record and should not be assumed.

The registry describes the treatment sequences and five-day spacing between periods, but does not provide a full account of how period effects, sequence effects, or potential carryover were assessed. It also does not report the statistical model, the treatment of missing concentration samples, or the approach to values below quantification. These are important analysis details for a crossover pharmacokinetic study.

The posted result is for low-fat meal versus fasting. Although the protocol description includes a high-fat meal condition, its comparative result is not reported in the registry record. The record also does not report a multiplicity plan for the 14 outcome measures or the trial’s original sample-size justification. Finally, safety summaries are reported by period and sequence, without participant-level detail or event descriptions in the record; those counts cannot establish causation or provide a complete safety characterization.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study to Understand the Effect of Low-Fat and High-Fat Meals on the Medicine Called PF-07284890 in Healthy Adults — NCT05349864.
• Registry record last updated 2024-03-15; status last verified 2023-08.
• Results first posted 2024-03-15.
• 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

NCT05349864 is a completed phase 1, open-label, randomized, single-dose, three-period crossover study sponsored by Pfizer. It enrolled 12 healthy participants and compared PF-07284890 exposure after a low-fat meal with exposure while fasting. The primary endpoint was plasma AUClast, measured over the stated post-dose sampling times.

The registry reports geometric means of 28270.0 ng*hr/mL for the low-fat meal condition and 23510.0 ng*hr/mL for the fasted condition. The adjusted geometric-mean ratio was 120.25, with a 90% confidence interval from 109.76 to 131.75. A high-fat meal condition was included in the sequence descriptions, but a comparative result for it is not reported in the registry record.

The record lists secondary pharmacokinetic and safety outcomes, but does not report the full statistical analysis plan, model specification, multiplicity approach, or detailed missing-data methods. Safety counts are presented by period and sequence, with zero serious adverse events reported in the listed groups.