Drug-Drug Interaction Study to Estimate the Effect of PF-07321332/Ritonavir and Ritonavir on Midazolam in Healthy Participants: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Drug-Drug Interaction Study to Estimate the Effect of PF-07321332/Ritonavir and Ritonavir on Midazolam in Healthy Participants (NCT05032950), a randomized phase 1 study. The primary outcome is Maximum Plasma Concentration (Cmax) of Midazolam When Administered Alone and With PF-07321332/Ritonavir.

2. Clinical Question

The study asks how the maximum plasma concentration of midazolam differs when it is administered alone and when it is administered with PF-07321332/ritonavir. Midazolam is described in the registry as a CYP3A4 substrate. The registry also lists a treatment with ritonavir plus midazolam, and secondary outcomes include pharmacokinetic measures under the different treatment conditions.

The population was healthy participants. The study title identifies a Phase 1, open-label, randomized, three-treatment, six-sequence, three-period crossover design. The registry reports enrollment of 12 participants. The primary endpoint’s analysis population was defined as randomized and treated participants with at least one primary-interest pharmacokinetic parameter in at least one treatment period; the primary-arm summaries report 10 participants in each of the two treatment groups shown for that endpoint.

The primary comparison is midazolam with PF-07321332/ritonavir versus midazolam alone. The registry’s endpoint description identifies the combination as the test treatment and midazolam alone as the reference. The posted analysis gives a model-based ratio and 90% confidence interval. Its analysis label says Reference/Test Ratio, while the endpoint description says the ratio is test/reference. That difference in labeling is relevant when reading the direction of the reported ratio.

This is a pharmacokinetic question, not a comparison of clinical benefit. The record does not report a patient-disease outcome or an efficacy question. It also does not report a rationale for the enrollment target, a washout schedule, or further details about the timing and conduct of the treatment periods. Those details are not reported in the registry record.

3. Trial Design

The registry describes a Phase 1, randomized, open-label, three-treatment, six-sequence, three-period crossover study in healthy participants. In a crossover design, participants receive treatments in periods rather than being assigned to remain in a single treatment group throughout. This design can support within-participant comparisons because measurements under different treatments come from the same participant, subject to the study’s actual sequence and period procedures.

The listed treatments were oral midazolam alone, oral PF-07321332/ritonavir with oral midazolam, and oral ritonavir with oral midazolam. The primary endpoint comparison concerns the first two conditions. The study title identifies the crossover structure, but the registry record does not describe the sequence allocation details, any washout interval, or how possible carryover effects were handled beyond the stated analysis model.

The study was unmasked, meaning masking was listed as none. The record reports enrollment of 12 and status as completed. It lists a start date of 2021-09-17 and a primary completion date of 2021-12-09. The primary endpoint’s two posted arm summaries each have a participant count of 10. The record does not explain why that count differs from the enrollment figure, or how many participants contributed data to each period or each of the other outcomes.

The primary analysis population was defined as randomized and treated participants with at least one primary-interest pharmacokinetic parameter in at least one treatment period. This is a parameter-specific definition rather than a statement that every randomized participant contributed every measurement. Details on exclusions, missing observations, and the handling of incomplete treatment periods are not reported in the registry record.

The registry lists other pharmacokinetic and safety outcomes, including measures of midazolam concentration, exposure, clearance, and adverse events. It lists 14 outcome measures and six analyses. The details available for those outcomes are not uniform, so the primary comparison should be distinguished from the other posted measures.

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: A total of 12 healthy participants were randomized to receive midazolam, PF-07321332/ritonavir + midazolam, and ritonavir + midazolam in a total of 6 sequences.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The PK parameter analysis population was defined as all participants randomized and treated who had at least 1 of the PK parameters of primary interest in at least 1 treatment period.

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.

6. Primary Endpoint

The registry lists the primary outcome as Maximum Plasma Concentration (Cmax) of Midazolam When Administered Alone and With PF-07321332/Ritonavir, measured over Midazolam: Day 1 Predose, 0.5, 1, 2, 3, 4, 6, 8, 10, 12, 16, 24, 36 hours postdose; Midazolam+PF-07321332/ritonavir: Day 5 predose, 0.5, 1, 2, 3, 4, 6, 8, 10, 12, 16, 24, 36, 48 and 72 hours postdose.

The primary endpoint is maximum plasma concentration, or Cmax, of midazolam after administration alone and with PF-07321332/ritonavir. Cmax is the highest measured concentration in the sampling schedule. In pharmacokinetic studies, it describes the peak concentration observed over the measured time course; it is not a measure of clinical benefit or a direct measure of how a participant feels.

The registry gives the unit as ng/mL and describes Cmax as observed directly from the data. Sampling for midazolam alone was scheduled from predose through 36 hours after dosing on Day 1. For midazolam with PF-07321332/ritonavir, the listed schedule was on Day 5 and extended through 72 hours postdose. The different schedules are part of the record and should be kept in mind when interpreting the measured values.

The endpoint is reported using geometric means and geometric coefficients of variation. A geometric mean is commonly used for positive pharmacokinetic measurements that are analyzed on a logarithmic scale; it describes a central value on the original scale after analysis on the log scale. The coefficient of variation describes relative spread rather than spread in the original measurement units. These summaries do not show each participant’s concentration-time profile.

The endpoint is useful for characterizing how measured midazolam exposure changes under the listed treatment conditions. The registry analysis models natural log-transformed Cmax and reports a treatment ratio with a 90% confidence interval. That interval communicates uncertainty in the estimated relative difference, not the range of individual participant responses.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint was analyzed on the natural logarithm scale using a mixed-effects model. Log transformation is commonly used for positive pharmacokinetic measurements because differences on the log scale correspond to relative changes on the original scale. A treatment contrast from such a model can therefore be expressed as a ratio after transforming the contrast back to the original scale. This is a general explanation of the method; the registry does not give further details about model diagnostics or the transformation back to the reported percentage.

Why include sequence, period, and treatment?

In a crossover trial, sequence identifies the order in which treatments are assigned, while period identifies the treatment period in which a measurement is collected. Treatment is the comparison of scientific interest. Including these terms as fixed effects allows the model to account for systematic differences associated with the assigned order, period, and treatment, rather than attributing every observed difference to treatment alone. The registry reports these fixed effects but does not state whether additional terms were considered.

What does the participant random effect do?

The model includes participant within sequence as a random effect. Repeated measurements from the same participant are generally correlated rather than independent. A participant-level random effect is one way to account for that dependence and for participant-to-participant differences in the underlying measurement. The registry does not report the estimated variance components or the model’s covariance structure, so those aspects cannot be assessed from the record.

How should the confidence interval be read?

A confidence interval gives a range of values compatible with the data and the model under the method’s assumptions. It expresses uncertainty about the estimated treatment contrast; it is not a range expected to contain most individual participants’ values. The primary result is reported with a 90% confidence interval. The record does not report a primary p-value. The posted ratio and interval should be interpreted with the ratio-direction labeling caveat described in the endpoint information.

What do geometric summaries mean?

A geometric mean is formed using logarithms and is appropriate for positive measurements that may have an asymmetric distribution on the original scale. It is not the ordinary arithmetic average. The geometric coefficient of variation expresses variability relative to the geometric mean, rather than in ng/mL. Because the measure is a summary, it does not reveal the distribution of individual measurements or whether a small number of observations had substantial influence.

What is different about the secondary re-calculations?

The detailed secondary endpoint is terminal half-life. The posted comparison estimates for this endpoint are simple, unadjusted re-calculations from the posted summary statistics. They are not the same as the primary mixed-model analysis, and they should not be presented as if they were analyses performed by the sponsor. The re-calculated p-values are unadjusted for the multiple secondary comparisons. The registry does not post sponsor analyses for this secondary endpoint.

For planning purposes, the sample-size illustrations use the observed effect or a stated fraction of that effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation. Such calculations illustrate what sample sizes might be needed under those stated assumptions; they are not observed power and are not the trial’s original sample-size calculation. The registry does not report the protocol’s sample-size assumptions or justification.

Why does multiplicity matter?

When multiple outcomes or comparisons are examined, the chance of at least one apparently notable result can increase if each is assessed without adjustment. Whether adjustment is appropriate depends on the study’s prespecified goals and analysis plan. The registry lists 14 outcome measures and six analyses, but it does not report a multiplicity strategy or identify a hierarchy for the secondary outcomes. The reported secondary re-calculations should therefore be treated as exploratory rather than as a collection of independently confirmatory findings.

The registry defines the primary pharmacokinetic analysis population as randomized and treated participants with at least one primary-interest parameter in at least one treatment period. That definition does not itself explain how missing measurements were handled, whether participants with incomplete periods were included in a particular model, or whether sensitivity analyses were performed. These details are not reported in the registry record.

10. Analyses Reported by the Sponsor

The registry reports a sponsor-style mixed-model analysis for the primary Cmax comparison. It gives a Reference/Test Ratio of 368.33% and a 90% confidence interval from 318.91% to 425.41%. The endpoint description specifies the ratio as PF-07321332/ritonavir plus midazolam divided by midazolam alone, while the analysis label uses Reference/Test wording. The record does not explain the difference, so the labels should be kept in view when interpreting direction.

The primary result is not a simple ratio recalculated from the arm summaries in this analysis. It is reported as the output of the specified mixed-effects model, which accounts for sequence, period, treatment, and participant within sequence. The registry does not report a primary p-value or enough participant-level information to reproduce that model independently.

The detailed secondary half-life comparisons are different: they are unadjusted re-calculations from posted summary statistics, not sponsor analyses. They do not replace or revise the sponsor’s primary mixed-model result. The registry lists no sponsor-reported analyses for that detailed secondary endpoint. As the discrepancy flag is false, no discrepancy between the posted result and a re-calculation is identified in the record.

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies Cmax as the primary endpoint and describes a model for comparing midazolam alone with midazolam plus PF-07321332/ritonavir. It also lists 14 outcome measures and six analyses. The record does not report a statistical analysis plan, a hierarchy for secondary outcomes, or a procedure for adjusting for multiple comparisons. Without those details, the secondary findings should be interpreted cautiously, especially where results are unadjusted re-calculations.

Multiplicity is a general statistical concern: as more outcomes or comparisons are examined, the chance of finding at least one apparently notable result can increase when each test is considered in isolation. This does not mean that a particular result is necessarily spurious. It means that interpretation depends on prespecification, the relationships among outcomes, and any adjustment or prioritization strategy. Those details are not reported in the registry record.

The primary analysis population is defined as randomized and treated participants with at least one primary-interest pharmacokinetic parameter in at least one treatment period. This definition gives a broad eligibility rule for the analysis but does not describe the number contributing to every analysis or how incomplete data were handled. The record does not report reasons for missing measurements, imputation methods, sensitivity analyses, or whether missingness differed by treatment or period.

The registry reports enrollment of 12, while the two primary arm summaries each report 10 participants. The record does not explain this difference, so no reason should be inferred. It also does not report how participants with incomplete treatment sequences were handled, or whether all enrolled participants contributed to safety and pharmacokinetic summaries. The absence of those details limits assessment of attrition and potential selection in the analyzed data.

The registry does not report interim analyses, stopping rules, or a sample-size justification. The posted sample-size illustrations are planning calculations based on observed effects and stated assumptions; they are not a report of the original design process and do not constitute observed power. These missing design and analysis details constrain how confidently the posted estimates can be generalized beyond their stated model and analysis population.

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 reports serious adverse event figures separately for the three treatment conditions. It records 0 participants with serious adverse events in the midazolam-alone group, 0 in the PF-07321332/ritonavir plus midazolam group, and 0 in the ritonavir plus midazolam group. The corresponding numbers at risk were 10, 11, and 11. No deaths were reported in any of these groups.

These are counts in the registry record, not proof that serious events cannot occur. The number of participants at risk is limited, and the record does not provide event narratives or further detail about exposure time, monitoring, or adjudication. The counts should be read as the observed serious-event information posted for these groups, not as a precise estimate of the frequency of rare harms.

The record also reports non-serious adverse event counts, but these are distinct from the serious-event figures. A count of participants with an event does not describe severity, duration, causality, or whether the same participant experienced more than one event. Safety information should not be conflated with the primary pharmacokinetic endpoint, which measures midazolam concentration rather than adverse effects.

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

The primary result describes a model-based comparison of midazolam maximum plasma concentration when midazolam was administered alone and with PF-07321332/ritonavir in healthy participants. The posted geometric means are 9.812 ng/mL and 36.18 ng/mL, respectively, and the registry reports a ratio of 368.33% with a 90% confidence interval from 318.91% to 425.41%. The endpoint description defines the ratio in test/reference terms, while the analysis label says Reference/Test Ratio; that difference should remain part of the interpretation.

The result is about measured pharmacokinetics under the study conditions. It does not establish clinical efficacy, demonstrate a clinical benefit, or answer whether a particular treatment should be used. The registry record does not report patient outcomes or a disease-treatment comparison. It also does not establish how the result applies to populations other than the healthy participants studied.

The confidence interval expresses uncertainty in the estimated ratio under the model. It does not describe the range of values among individuals, and it does not show that every participant had a similar response. The geometric mean and geometric coefficient of variation are group summaries and cannot replace participant-level concentration profiles or an assessment of the full distribution.

Interpretation is also bounded by what the record reports. The primary analysis population is defined, but handling of missing data, reasons for the difference between enrollment and primary arm-summary counts, and sensitivity analyses are not reported. The registry does not report the original sample-size justification, a multiplicity approach, or detailed model diagnostics.

Secondary outcomes and safety counts can provide context, but they answer different questions and are not all reported with the same level of detail. Unadjusted secondary re-calculations should not be treated as equivalent to the primary model-based analysis. Together, the posted findings characterize this study’s pharmacokinetic comparison; they do not settle questions beyond its population, endpoint, and reported analysis.

14. Why This Trial Matters Statistically

This study illustrates why the design and scale of a pharmacokinetic analysis matter. A crossover design collects measurements under different treatment conditions within participants, and the primary model accounts for participant within sequence as well as period and sequence. That structure is important when measurements from the same participant are related rather than independent.

The primary endpoint uses a logarithmic analysis and is reported as a ratio, a natural way to express relative changes in positive concentration measurements. Reading such a result requires attention to the endpoint definition and to the reported analysis label. In this record, the description specifies test/reference, while the analysis label says Reference/Test. That is a useful reminder that interpretation depends not only on a numerical estimate and interval but also on clear documentation of the contrast’s direction.

The record also distinguishes a prespecified primary analysis from secondary re-calculations. A model-based primary result and unadjusted comparisons derived from summary statistics do not have the same evidentiary role. Secondary analyses can be informative, but a broad set of outcomes without a reported multiplicity plan makes isolated results less definitive.

Finally, the record shows why enrollment, analysis populations, and missing-data reporting need to be read separately. Enrollment is 12, while primary arm summaries report 10 participants each, and the registry does not explain the difference. That does not invalidate the reported estimate, but it limits assessment of how the analyzed observations relate to the enrolled group. These are practical statistical considerations when translating a compact registry result into an appropriately bounded 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 registry record describes the study design and primary model but leaves several details unavailable. It does not report the washout schedule, sequence-allocation details, or how possible carryover effects were assessed. Those details are particularly relevant to a crossover study, where measurements across treatment periods may be related.

The enrollment figure is 12, while each of the two posted primary arm summaries reports 10 participants. The record does not explain the difference or give participant flow by treatment period. It also does not report the number contributing to each secondary outcome, reasons for missing measurements, the handling of incomplete profiles, or sensitivity analyses. These omissions limit evaluation of missingness and the representativeness of the analyzed data.

The primary endpoint description and analysis label use different wording for the ratio direction. The endpoint description defines PF-07321332/ritonavir plus midazolam divided by midazolam alone; the result is labeled Reference/Test Ratio. The record does not clarify that difference. Although the arm summaries provide context, the label should be considered when interpreting the reported ratio.

Further limitations include the absence of a reported sample-size justification, multiplicity strategy, detailed model diagnostics, and interim-analysis information. The registry does not provide participant-level data that would allow independent reproduction of the mixed-effects analysis. The detailed secondary comparisons are unadjusted re-calculations from summary statistics, and the registry does not report sponsor analyses for that endpoint.

The study population was healthy participants, and the primary outcome was pharmacokinetic. The record therefore does not directly answer clinical outcome questions or establish applicability to other populations. These limits concern the scope of the evidence reported; they do not by themselves determine whether the primary estimate is valid.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Drug-Drug Interaction Study to Estimate the Effect of PF-07321332/Ritonavir and Ritonavir on Midazolam in Healthy Participants — NCT05032950.
• Registry record last updated 2023-10-04; status last verified 2022-12.
• Results first posted 2023-10-04.
• 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 a completed Phase 1, randomized, open-label crossover study in healthy participants. It lists three oral treatment conditions: midazolam alone, PF-07321332/ritonavir with midazolam, and ritonavir with midazolam. Enrollment was 12.

The primary endpoint was midazolam Cmax for treatment with PF-07321332/ritonavir compared with midazolam alone. The mixed-model result was reported as a ratio of 368.33% with a 90% confidence interval from 318.91% to 425.41%. The endpoint description specifies a test/reference ratio, while the result label says Reference/Test Ratio.

The registry lists 14 outcome measures and six analyses. It also reports safety counts and a detailed secondary half-life endpoint with unadjusted re-calculations. The record does not report a sample-size justification, multiplicity strategy, or full missing-data details.