Investigation of Ritonavir Effects on Afatinib Exposure in Healthy Subjects: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Investigation of Ritonavir Effects on Afatinib Exposure in Healthy Subjects (NCT01426958), a randomized phase 1 study. The primary outcome is Area Under Curve From 0 to tz (AUC0-tz).

2. Clinical Question

This is a pharmacokinetic question rather than a question about tumour response, symptom improvement, or other clinical outcomes. The outcome is an exposure measure derived from plasma concentrations over time. The available results therefore characterize afatinib exposure in the studied healthy volunteers under the recorded regimens; they do not establish what would happen in a patient population or under other dosing schedules.

3. Trial Design

The registry describes a completed phase 1, randomized, open-label, three-way crossover trial in healthy male volunteers. Enrollment was 24. The study began in 2011-08 and primary completion was recorded in 2011-12. In a crossover design, treatment conditions are compared within a participant across treatment periods, rather than assigning each participant to only one treatment for the entire study. In general, this can reduce the influence of between-person differences on treatment comparisons, although it depends on the period and sequence structure and on appropriate handling of within-person data.

The three conditions were afatinib alone, afatinib with concomitant ritonavir, and afatinib with timed ritonavir. The official title specifies a single oral 40 mg afatinib dose. The arm descriptions state that the ritonavir regimens involved multiple oral doses for 3 days and afatinib on the second treatment day. Although the registry labels the two regimens as concomitant and timed, their descriptions use the same dosing wording; the exact timing difference is not reported in the registry record. The washout interval, treatment sequence details, and period-specific schedule are also not reported in the registry record.

The trial was open label, so masking was not used. The primary endpoint set included evaluable subjects in the treated set who were assigned to final dose groups and had at least one observation for at least one primary pharmacokinetic endpoint without important protocol violations relevant to pharmacokinetic evaluation. The endpoint-specific counts were 21 for afatinib alone, 24 for afatinib with concomitant ritonavir, and 22 for afatinib with timed ritonavir. These are the reported counts for the endpoint groups and should not be read as a replacement for the registry's overall enrollment.

The design supports comparisons of pharmacokinetic exposure across the three treatment conditions. It does not, by itself, establish the clinical consequences of any exposure difference. The registry record does not report the randomization sequence, the handling of period or sequence effects, or the washout details.

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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: This subject set includes all evaluable subjects of the treated set who were assigned to the final dose groups and who provide at least one observation for at least one primary (PK) endpoint without important protocol violations relevant to the evaluation of PK.

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 Area Under Curve From 0 to tz (AUC0-tz), measured over 0, 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 24, 36, 48, 72, 96, 120 hours post.

The primary endpoint was area under the plasma concentration curve from time zero to tz, the time of the last quantifiable plasma concentration. AUC0-tz represents accumulated measured exposure over that interval. It uses the concentration measurements collected after dosing rather than relying on a single concentration at one time point. The registry lists measurements at 0, 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 24, 36, 48, 72, 96, and 120 hours post-dose.

The outcome is reported as a geometric mean, with geometric coefficient of variation as the dispersion measure. In general, pharmacokinetic concentrations and exposure measures can be positively skewed, so summaries on a multiplicative scale are often useful. A geometric mean describes a central value on that scale; the geometric coefficient of variation describes relative variability and is not a confidence interval or an individual range.

AUC is used because it captures exposure across the sampled concentration-time profile, making it relevant to a question about whether coadministration or administration timing changes overall afatinib exposure. It does not directly measure clinical benefit, toxicity, or a patient-centred outcome. The registry gives no separate secondary outcomes as posted results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary analysis method as ANOVA and the reported parameter as a geometric mean ratio. The endpoint itself is AUC0-tz, summarized at the group level by geometric means and geometric coefficients of variation. The analysis therefore links a concentration-based exposure outcome to pairwise comparisons among the three treatment conditions.

What does a geometric mean ratio convey?

In general, a ratio expresses one treatment condition relative to another on a multiplicative scale. A ratio is not an absolute difference in AUC units. Its interpretation depends on which group is placed in the numerator and which in the denominator. The registry lists the groups in each comparison and gives the resulting ratio, but does not further spell out the numerator and denominator convention. For that reason, the reported comparison values are best quoted with their group labels rather than assigning an unrecorded direction.

Geometric summaries are common for pharmacokinetic measures because concentrations and exposure can vary multiplicatively and may not have a symmetric distribution on the original scale. In general, analysis on a logarithmic scale can make multiplicative comparisons easier to model, with results then expressed as ratios. The registry does not describe the transformation or model details, so these are general explanations of the approach, not claims about unreported implementation choices in this trial.

How should the confidence intervals be read?

Each reported pairwise ratio has a 90% confidence interval. In general, a confidence interval represents uncertainty in an estimated parameter under the model and assumptions used. It indicates which values are reasonably compatible with the observed data and analysis; it is not a guarantee about the true value, nor does it describe the spread of individual measurements. A narrower interval indicates greater precision than a wider interval in the same analysis context.

Whether an interval supports a non-inferiority or equivalence conclusion depends on a prespecified boundary defining the acceptable range. The registry labels these comparisons as non-inferiority or equivalence analyses, but the margin and decision rule are not reported in the registry record. The confidence intervals can therefore be described, but they cannot independently establish that a formal criterion was met. The p-values are reported in the registry, but the full testing hypotheses and any relationship between those values and a margin are not detailed.

What does ANOVA mean in this setting?

In general, analysis of variance compares modeled outcome levels across conditions while accounting for variation represented in the model. In a crossover study, observations from a participant under different treatments are related, so a suitable analysis normally needs to reflect that structure. The registry reports ANOVA but does not state the model terms, covariance assumptions, transformations, or how sequence and period were handled. No more specific description of the fitted model can be made from the registry record.

The design is randomized and crossover, but randomization does not remove the need to describe how the analysis handles within-participant comparisons. The record also does not report the washout interval or a carryover assessment. These omissions do not invalidate the posted estimates on their own, but they limit an independent assessment of whether the analysis fully accounts for the design.

How do the arm summaries differ from the comparison estimates?

The arm-level geometric means and geometric coefficients of variation describe exposure and relative dispersion within each treatment condition. The pairwise ratios describe estimated contrasts between conditions. Neither set of summaries provides all participant-level data, and the group summaries cannot reveal the distribution of individual responses or the correlation between repeated measurements within a participant.

In general, geometric coefficient of variation is a measure of relative spread, not a confidence interval for the geometric mean. A confidence interval instead expresses uncertainty about an estimated parameter. Keeping these quantities distinct matters: one describes variability in the measured values, while the other describes precision in a comparison estimate.

What can the planning illustrations establish?

Planning illustrations based on the observed effect are not observed power and are not the trial's original sample-size justification. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. These are planning assumptions, not details of the original crossover design. The registry record does not report the original sample-size justification, so it is not possible to determine whether the enrollment was selected using these assumptions or a different rationale.

In general, a sample-size calculation estimates the amount of information needed for a specified design, effect, and error criterion. A calculation that uses an effect observed after the trial is complete is a retrospective planning illustration. It does not measure the probability that this completed study detected an effect, and it should not be used as a substitute for confidence intervals or the reported analysis. The registry record does not report an interim analysis plan or a fuller statistical analysis plan.

10. Analyses Reported by the Sponsor

The registry reports the sponsor's pairwise analyses as ANOVA comparisons of geometric mean ratios, with 90% confidence intervals and p-values. The posted arm-level geometric means and geometric coefficients of variation are descriptive summaries, while the ANOVA ratios are the reported inferential comparisons. They answer related but different questions and should not be substituted for one another.

No discrepancy between the sponsor-reported results and a recalculation is identified in the registry record. The record does not give an additional recalculated estimate to compare with the sponsor's reported ratios. Accordingly, the numerical interpretation rests on the reported ANOVA estimates and intervals, rather than on a second estimate or a claim that the sponsor analysis has been reproduced independently.

The registry classifies the pairwise analyses as non-inferiority or equivalence analyses, but does not report the margin or complete testing rule. The sponsor's labels alone are not enough to determine whether a formal non-inferiority or equivalence criterion was satisfied. The 90% confidence intervals can be read as measures of uncertainty around the reported ratios, while the conclusions that depend on a margin remain unspecified in the registry record.

11. Multiplicity, Missing Data and Analysis Populations

The primary endpoint has three reported pairwise comparisons: afatinib alone versus concomitant ritonavir, afatinib alone versus timed ritonavir, and concomitant versus timed ritonavir. The registry records 9 analyses overall and 3 outcome measures, while no secondary results are posted. The record does not explain how those analyses relate to one another or whether a multiplicity adjustment was planned or applied.

In general, when several hypotheses are tested, the chance of at least one false-positive result can increase if each test is interpreted separately without an adjustment or a prespecified testing hierarchy. Whether adjustment is appropriate depends on the aims and analysis plan. Because the registry does not report the multiplicity strategy, the pairwise p-values should not be treated as though a particular familywise error procedure was documented. The confidence intervals and ratios remain useful descriptions of the posted analyses, but the record does not establish how confirmatory conclusions were controlled across comparisons.

The primary analysis population is described as evaluable subjects from the treated set, assigned to final dose groups, with at least one observation for at least one primary pharmacokinetic endpoint and no important protocol violations relevant to pharmacokinetic evaluation. Endpoint group counts were 21, 24, and 22 for the three conditions. These figures are endpoint-specific; they do not establish that every participant contributed a complete set of measurements under every condition.

Missing-data handling, reasons for missing observations, and any imputation method are not reported in the registry record. The record also does not provide a detailed account of exclusions or protocol deviations. In a crossover trial, incomplete treatment-period data can affect which within-participant comparisons are available; without more information, the extent or direction of any such effect cannot be assessed.

The trial is randomized, but the registry does not report an interim analysis plan, a full statistical analysis plan, or details of the treatment of sequence and period effects. Those omissions leave uncertainty about prespecification and the complete analysis population, not about the values explicitly posted for the primary endpoint comparisons.

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 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events for the three treatment conditions, with 0 affected in the afatinib group, 0 in the afatinib plus concomitant ritonavir group, and 0 in the afatinib plus timed ritonavir group. The corresponding numbers at risk were 23, 24, and 22. These are the recorded serious-event counts for the listed groups; they do not establish that no adverse events of any kind occurred.

Other adverse events were recorded for 2 participants in the afatinib group, 7 in the concomitant ritonavir group, and 5 in the timed ritonavir group. These counts are distinct from the serious adverse event figures. The registry does not provide event-term details in the listed serious-event terms, and deaths are not reported in the registry record.

In general, adverse-event counts describe events observed and recorded during the specified trial conditions. They do not by themselves show whether an event was caused by a treatment, and a small phase 1 study is limited in its ability to characterize uncommon or delayed events. The registry record does not provide sufficient detail to assess event severity beyond the serious-event classification, timing, recurrence, or individual-level overlap across treatment periods. These figures should be read as descriptive safety reporting, not as a comparative estimate of risk.

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

The posted results describe afatinib exposure measured by AUC0-tz in healthy male volunteers under three treatment conditions: afatinib alone, afatinib with concomitant ritonavir, and afatinib with timed ritonavir. The arm summaries and pairwise geometric mean ratios provide information about pharmacokinetic exposure and the precision of the reported comparisons. They address the trial's stated question about ritonavir and afatinib pharmacokinetics.

They do not establish whether a treatment improves or worsens a clinical outcome. A pharmacokinetic endpoint measures drug exposure, not tumour response, symptom change, or survival. Nor do the results establish an individual participant's expected exposure. The arm summaries describe group-level values, and the confidence intervals quantify uncertainty in the modelled comparisons rather than individual variation.

The registry labels the pairwise analyses as non-inferiority or equivalence analyses. However, the margin and complete decision criteria are not reported in the registry record. Therefore, the reported ratios and intervals should not be converted into a definitive equivalence or non-inferiority conclusion without those criteria. The registry's p-values are part of the reported analysis, but they do not replace a stated margin or full analysis plan.

The population and regimen also limit generalization. Participants were healthy male volunteers, and the official title describes a single oral dose of 40 mg afatinib with multiple oral doses of ritonavir. These results do not directly describe patients, repeated afatinib treatment, other populations, or unrecorded dosing schedules. The exact timing distinction between the concomitant and timed ritonavir regimens is not reported in the registry record, so that comparison cannot be tied to a more specific timing interval.

The results are best understood as controlled pharmacokinetic comparisons under the study conditions. They do not provide a complete assessment of safety or clinical consequences, and the serious adverse event counts alone cannot establish comparative safety. No medical treatment recommendation follows from these exposure summaries.

14. Why This Trial Matters Statistically

This trial illustrates why pharmacokinetic studies often focus on exposure rather than clinical outcomes. When a study asks whether a coadministered drug or its timing changes drug exposure, AUC provides a summary of concentration over time, while geometric mean ratios express contrasts on a multiplicative scale. Those measures are suited to the stated pharmacokinetic question, but they must not be mistaken for measures of efficacy.

The crossover structure is also statistically important. In general, comparing treatments within the same participants can reduce the effect of stable between-person differences. It also creates correlated observations that should be reflected in the analysis. The registry identifies an ANOVA method but does not describe its model terms, treatment of periods, or sequence effects. This makes the design informative while leaving important analytic details unavailable to readers.

The record also demonstrates the difference between a statistical label and a fully interpretable decision rule. A comparison can be tagged as non-inferiority or equivalence, but evaluating that claim requires the prespecified margin and analysis details. Without the margin, the geometric mean ratios and confidence intervals remain reportable estimates, yet the threshold for a formal conclusion cannot be checked from the registry record.

Finally, endpoint-specific evaluable counts, missing-data details, and multiplicity strategy matter when interpreting precision and the set of comparisons. The registry gives the primary endpoint population definition and group counts, but not a detailed account of missing observations or multiplicity handling. These omissions show why reading estimates alongside design and analysis information is essential: the numerical summaries describe the reported data, while the protocol and analysis plan determine how strongly they support the intended claims.

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.

This is a small, early-phase pharmacokinetic study in healthy male volunteers, with registry enrollment of 24. Its population and single-dose afatinib question limit how directly the exposure comparisons can be applied to patients or other treatment settings. The registry does not report results for clinical efficacy outcomes, and the primary endpoint is a pharmacokinetic measure rather than a patient outcome.

The crossover design requires attention to within-participant dependence, treatment sequence, and period effects. The registry identifies ANOVA but does not provide the model specification or say how sequence and period were handled. Washout details and any carryover assessment are not reported in the registry record. These gaps prevent a full assessment of whether the analysis reflects all relevant features of the design.

The descriptions of the concomitant and timed ritonavir regimens do not explain the exact timing difference. Both arm descriptions state that ritonavir was given over 3 days and afatinib on the second treatment day. As a result, the timing comparison cannot be translated into a more precise administration schedule from the registry record alone.

The endpoint population is defined as evaluable subjects, and the group counts differ across treatment conditions. The record does not describe missing observations, reasons for exclusions, or how incomplete periods were handled. It also does not report a multiplicity adjustment, a non-inferiority or equivalence margin, or a complete statistical analysis plan. Those details are important for judging the scope of formal claims based on the comparisons.

Safety reporting is limited in the available summary. Serious adverse event counts are reported, but deaths are not reported in the registry record, and the listed information does not give detailed event terms or participant-level patterns across periods. The results therefore provide a bounded account of exposure and recorded events, not a comprehensive appraisal of longer-term safety or clinical effects.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Investigation of Ritonavir Effects on Afatinib Exposure in Healthy Subjects — NCT01426958.
• Registry record last updated 2014-06-09; status last verified 2013-10.
• Results first posted 2013-12-19.
• 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, randomized, open-label, three-way crossover phase 1 trial in healthy male volunteers, with enrollment of 24. It compared a single oral dose of afatinib alone with afatinib given with concomitant or timed ritonavir. The primary endpoint was AUC0-tz, summarized by geometric means and geometric coefficients of variation.

Reported pairwise ANOVA analyses give geometric mean ratios and 90% confidence intervals for all three treatment comparisons. The registry labels these as non-inferiority or equivalence analyses but does not report the margin or full decision rule. The record also does not detail the timing distinction between ritonavir schedules, the analysis model specification, missing-data handling, or multiplicity strategy. The results are pharmacokinetic findings in the studied population and do not directly establish clinical outcomes.