Bioequivalence (BE) Study of Test Griseofulvin 500 Milligram (mg) Tablets Versus Reference and Dose Proportionality Study of Test Griseofulvin 250 mg and 500 mg Tablets Under Fed Conditions: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Bioequivalence (BE) Study of Test Griseofulvin 500 Milligram (mg) Tablets Versus Reference and Dose Proportionality Study of Test Griseofulvin 250 mg and 500 mg Tablets Under Fed Conditions (NCT04318535), a randomized phase 1 study. The primary outcome is Maximum Plasma Concentration (Cmax) for Griseofulvin.

2. Clinical Question

In statistical terms, the study asks how the concentration peak observed after the test 500 mg product compares with the peak after the reference 500 mg product when the same participant receives treatments in a crossover. That design permits within-participant comparisons across treatment periods, while the sequence allocation varies the order of treatment. The study also collects pharmacokinetic data after the lower-dose test product, but its posted inferential analysis is the T1-versus-R Cmax comparison. This is a pharmacokinetic question in healthy participants under fed conditions, not a comparison of clinical outcomes in people receiving treatment for an infection.

3. Trial Design

The registry identifies this as a Phase 1, open-label, randomized, balanced, three-treatment, three-period, three-sequence, single-dose crossover study. Enrollment was 36, and the study status is completed. The study began and reached primary completion on 2020-01-13. GlaxoSmithKline is listed as the sponsor.

Each participant assigned to a sequence was to receive the test 500 mg tablet (T1), the test 250 mg tablet (T2), and the reference 500 mg tablet (R), in the order specified by that sequence. The listed sequences were T1T2R, T2RT1 and RT1T2, allocated according to a 1:1:1 ratio. Doses were single oral doses under fed conditions. The registry describes washout periods of at least 7 days between periods in the sequence descriptions, and a washout range of at least 7 days and not more than 14 days in the study summary.

A crossover design differs from a parallel-group design because participants receive multiple treatments over separate periods. This allows each participant’s observations under different products to contribute to treatment comparisons, while the randomized sequences distribute the order of treatments. Period and sequence effects can matter in such designs; the registry says the pharmacokinetic analysis used a mixed model, but further model specifications are not reported in the registry record.

The primary pharmacokinetic analysis set consisted of participants completing at least two periods with both 500 mg treatments. The reported arm summaries give 35 participants for each treatment, while the safety groups list 36 at risk for each treatment. The record does not explain the difference between these analysis populations. It also does not report a sample-size justification, detailed eligibility criteria, or further operational details about conduct beyond those.

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 36 participants were enrolled and received study treatment. The study had 3 periods: Period 1 (Day -1 to Day 5), Period 2 (Day 7 to Day 12) and Period 3 (Day 17 to Day 22) with a washout of atleast 7 days (not more than 14 days) between two subsequent dosing days.

Treatment Period 1 (Day -1 to Day 5)

Period1washout(Atleast 7 Days Post-dose)

Treatment Period 2 (Day 7 to Day 12)

Period2washout(Atleast 7 Days Post-dose)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Bioequivalence (BE) Analysis Set Population comprised of participants completing at least 2 periods with 500 mg test and 500 mg reference treatments of the study. Only those participants with data available at the specified time points were analyzed.

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) for Griseofulvin, measured over Pre-dose (within 1 hour prior to dosing) and at 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 8, 10, 12, 24, 48, 72 and 96 hours post-dose in each period.

Cmax is the maximum measured concentration of a drug in plasma after dosing. It describes the concentration peak, rather than the total amount of drug exposure over time or a clinical response. The registry reports Cmax in nanograms per milliliter, based on blood samples collected before dosing and at the specified post-dose times, from 0.5 through 96 hours.

In general, pharmacokinetic studies use Cmax to compare the concentration profiles associated with different products or doses. A comparison on the geometric-mean scale is useful when the analysis is performed after log transformation, as in this study. The ratio expresses the test product’s adjusted geometric mean relative to the reference product’s adjusted geometric mean. A ratio at 1 would indicate equal point estimates on that scale; a ratio above 1 indicates a higher test estimate, and a ratio below 1 indicates a lower test estimate.

The registry says pharmacokinetic parameters were estimated using standard non-compartmental methods and that the adjusted geometric mean is the antilog of the least-squares mean of log-transformed data. For T2, the reported Cmax is dose-normalized by multiplying the observed value by 2. That transformation allows the lower-dose estimate to be considered on a normalized basis, but the registry record does not report an inferential dose-proportionality comparison for T2 versus T1.

Cmax is a pharmacokinetic endpoint, not a direct measure of symptom improvement, infection resolution, or adverse clinical outcomes. Its interpretation is limited to the concentration peak observed under the study conditions and in the analyzed population.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary pharmacokinetic outcome was Cmax, the maximum plasma concentration measured after a single dose. Blood sampling was scheduled before dosing and at multiple post-dose times in each period. The registry says pharmacokinetic parameters were measured using standard non-compartmental methods. In general, non-compartmental methods summarize concentration-time data without requiring a fully specified compartment model for the drug’s movement through the body.

Why analyze geometric means?

The registry describes adjusted geometric means, calculated as the antilog of the least-squares mean of log-transformed data. In general, pharmacokinetic measurements can be analyzed on a logarithmic scale when their distribution is better represented in multiplicative than additive terms. After analysis on the log scale, taking the antilog returns estimates to the original measurement scale. A ratio of geometric means then expresses the relative size of the test estimate to the reference estimate.

The adjusted estimates are not simply arithmetic averages of raw concentrations. They come from a model-based analysis, with adjustment as specified by that model. The registry says a mixed model was used for evaluation of bioequivalence. A mixed model can account for features of a crossover design, including repeated observations from the same participant. The specific fixed and random effects, covariance structure, model diagnostics and handling of sequence or period terms are not reported in the registry record.

What does the ratio express?

The reported ratio is for the adjusted geometric least-squares mean of T1 divided by that of R. A value of 1 represents equal point estimates on the ratio scale. Values above 1 indicate a higher estimate for T1, while values below 1 indicate a lower estimate for T1. The reported ratio is 1.157, with a 90% confidence interval from 1.103 to 1.213. That interval conveys uncertainty in the estimated ratio under the analysis model.

In general, a confidence interval is not a statement that individual participant values lie within its limits. Nor does an interval alone determine bioequivalence. The interval must be considered against the acceptance boundaries defined for the analysis. Those boundaries are not reported in the registry record. The record also does not report a p-value for the Cmax comparison.

How does crossover affect the comparison?

In a crossover study, participants receive multiple treatments in different periods. This can make within-participant comparisons informative because each participant contributes observations under more than one treatment. Randomized sequences vary treatment order. Period effects, sequence effects and possible carryover are relevant general considerations for crossover analyses. The registry describes washout periods between treatments and says a mixed model was used, but it does not report details sufficient to assess each of these components independently.

The planned sequences were T1T2R, T2RT1 and RT1T2. The registry describes allocation among these sequences in a 1:1:1 ratio. This is the design allocation, not a statement that the final analyzed counts were identical across sequences. Sequence-specific analyzed counts are not reported in the registry record.

What is dose normalization?

The registry reports dose-normalized Cmax for T2, calculated by multiplying the observed value by 2. In general, dose normalization expresses an observed pharmacokinetic estimate on a common dose basis, which can support a comparison of dose-related patterns. It does not, by itself, establish proportionality. Assessing dose proportionality ordinarily requires an appropriate comparison and uncertainty estimate; a corresponding inferential result for T2 versus T1 is not reported in the registry record.

The arm summaries report adjusted geometric mean Cmax values of 2229.0 for T1, 2877.0 for T2 and 1927.0 for R, with standard errors of 0.033, 0.035 and 0.033, respectively. The T2 estimate is explicitly identified as dose-normalized. These descriptive summaries do not replace the reported model-based T1-versus-R comparison, and they do not provide a reported inferential test for dose proportionality.

How should planning illustrations be read?

Planning illustrations based on an observed effect are not observed power and are not evidence that a study had a particular probability of detecting an effect after the data were collected. In general, a planning calculation explores the participant count that might be needed under chosen assumptions. The registry describes assumptions of a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and effects defined as stated fractions of the observed effect. Such calculations are illustrative and are not the trial’s reported sample-size justification. The sample-size justification is not reported in the registry record.

The record does not report the detailed analysis plan, interim analyses, missing-data assumptions, model diagnostics or sensitivity analyses. These omissions limit independent assessment of how robust the reported inferential result is to alternative analytical choices. The posted endpoint estimate should therefore be read in the context of the specific analysis population and model description that the registry does provide.

10. Analyses Reported by the Sponsor

The registry’s reported inferential result for Cmax is the sponsor’s model-based comparison of T1 with R: a ratio of adjusted geometric least-squares means of 1.157, with a 90% confidence interval from 1.103 to 1.213. The analysis is described as using a mixed model on log-transformed pharmacokinetic data. That is the result to use when describing the sponsor-reported analysis.

A simple recalculation from arm-level summaries would be unadjusted and would not necessarily reproduce a model-based estimate from a crossover analysis. In general, the model-based comparison may account for design features and adjustments not represented by separate arm summaries. No alternative numerical recalculation is reported in the registry record. The registry record does not flag a discrepancy between the sponsor’s analysis and a recalculation, so there is no stated discrepancy requiring a different reading of the sponsor result.

The ratio and interval describe the Cmax comparison, but the registry record does not report the acceptance boundaries needed to judge the result against a bioequivalence criterion. The sponsor’s reported analysis therefore supports a precise description of the estimated ratio and its interval; it does not, on the information reported, permit a conclusion about whether a stated criterion was met. No sponsor-reported inferential result for dose proportionality between T2 and T1 is listed in the registry record.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 12 outcome measures, 7 posted secondary outcomes and 3 analyses. The primary outcome considered in the reported comparison is Cmax. The secondary outcomes include serious and non-serious adverse events, vital signs, hematology, clinical chemistry, alanine aminotransferase, serum beta-hCG results for female participants, and urinalysis dipstick results.

When a study evaluates multiple outcomes or makes multiple comparisons, the chance of at least one apparently unusual result can increase if each is interpreted separately without accounting for multiplicity. In general, multiplicity procedures can control that issue, but their use depends on the planned analysis and inferential goals. A multiplicity adjustment or hierarchical testing plan is not reported in the registry record. It is also not reported whether all listed outcomes were treated as confirmatory or how the three analyses relate to the full set of outcome measures.

The bioequivalence analysis population consisted of participants completing at least two periods with both 500 mg treatments, with data available at the specified time points. The record does not give a detailed account of missing observations, exclusions by reason, or any imputation method. It therefore does not report whether unavailable measurements were related to treatment, period, participant characteristics or other factors.

The pharmacokinetic summaries list 35 participants for each treatment, whereas the safety groups list 36 at risk for each treatment. Those are different analysis populations as reported; the registry record does not explain the difference. It would be inappropriate to infer a reason from those counts alone.

Further elements of the statistical plan—including a sample-size justification, interim analyses, sensitivity analyses, and detailed missing-data procedures—are not reported in the registry record. Without those details, the interpretation should remain focused on the outcome, population and model description that are explicitly reported, rather than assuming unreported analytical safeguards or choices.

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 0% reporting threshold.

Most frequent non-serious adverse event terms

The safety record lists 36 participants at risk for serious adverse events and 36 at risk for other adverse events in each treatment group. No serious adverse events were recorded for T1, T2 or R. For non-serious adverse events, the record lists none for T1 and T2, and 1 for R. No deaths were reported in any of the three treatment groups.

These are participant counts by treatment group as recorded in the safety results. The registry does not provide event terms for serious adverse events, and none are listed among the top serious-event terms. The reported counts do not describe the timing, duration, severity or relationship of the non-serious event to treatment; those details are not reported in the registry record.

In general, absence of recorded events in a study does not establish that an event cannot occur. Safety counts are also limited by the participants, treatment exposure and observation described in the study. This was a single-dose crossover study in healthy adult participants, and the safety figures should be understood within that setting. They are not a substitute for a broader assessment of safety in other populations or with other patterns of use.

The safety denominators are not the same as the pharmacokinetic analysis counts reported for each treatment. The record does not explain the distinction, so the safety figures should not be treated as if they came from the bioequivalence analysis set.

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

The posted result gives an estimate of relative Cmax for the 500 mg test product compared with the 500 mg reference product, in healthy adult participants under fed conditions. The adjusted geometric mean ratio was 1.157, and the 90% confidence interval was 1.103 to 1.213. This describes a higher estimated Cmax for T1 than R on the ratio scale, with uncertainty represented by the interval.

The result does not, by itself, establish that the products meet a bioequivalence standard. That determination requires the applicable acceptance range, which is not reported in the registry record. The confidence interval also does not describe the range of concentrations for individual participants. It characterizes uncertainty in the estimated ratio under the stated analysis.

The record gives dose-normalized descriptive Cmax summaries for the test 250 mg and 500 mg tablets, but it does not report an inferential dose-proportionality result. The arm summaries should not be treated as a formal demonstration of dose proportionality. The study’s pharmacokinetic estimates also do not directly establish a difference in clinical outcomes, including whether a treatment improves or fails to improve a health condition.

Applicability is bounded by the study setting: healthy adult participants, single-dose administration, fed conditions, and a randomized crossover design. The registry record does not establish what the estimates would be under different dosing conditions, in other populations, or with repeated dosing. The sample size and safety observations likewise do not resolve questions outside the study’s reported scope.

In short, the record supports a description of the reported Cmax comparison and its precision. It does not support a definitive bioequivalence conclusion without the acceptance boundaries, nor a conclusion about dose proportionality without a reported inferential comparison. It is a pharmacokinetic result, not a clinical effectiveness result.

14. Why This Trial Matters Statistically

This study is statistically interesting because it combines a test-versus-reference bioequivalence objective with a dose-proportionality objective in a randomized crossover design. The design uses different treatment sequences, allowing pharmacokinetic observations for several products to be collected across periods. The registry’s primary reported comparison is specifically the 500 mg test product against the 500 mg reference product.

The analysis illustrates why the scale of measurement and the model matter. Cmax is analyzed using log-transformed data, and the reported adjusted geometric means are compared as a ratio. This is a standard way to express relative pharmacokinetic estimates, but the ratio and confidence interval need to be interpreted against the appropriate acceptance boundaries before making a bioequivalence determination. Those boundaries are not reported in the registry record.

The record also illustrates the difference between descriptive summaries and inferential analyses. It reports arm-level adjusted geometric means for T1, T2 and R, while the inferential result is given for T1 versus R. Although the T2 summary is dose-normalized, a formal dose-proportionality comparison is not reported. A descriptive pattern alone is not the same as an estimate with an associated uncertainty interval and a defined inferential criterion.

Finally, the reported analysis and safety counts use different populations, and the registry does not explain that difference. Clear identification of analysis populations is important when interpreting trial results, especially when the main endpoint requires data from multiple periods. The record’s concise model description and lack of detail on missing data, multiplicity and acceptance boundaries also show why registry results should be read narrowly: they communicate key estimates, but do not always provide every element needed for a full independent assessment.

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 does not report the bioequivalence acceptance boundaries. As a result, the reported ratio and 90% confidence interval cannot be used alone to determine whether a prespecified bioequivalence criterion was met. The record also does not report a p-value for the Cmax comparison.

The study summary describes a dose-proportionality objective involving the test 250 mg and 500 mg tablets. The posted results give adjusted geometric mean Cmax summaries and identify the T2 estimate as dose-normalized, but an inferential comparison for dose proportionality is not reported in the registry record. The descriptive estimates therefore do not answer that objective on their own.

The model description is limited. The registry identifies a mixed model and log-transformed pharmacokinetic analysis, but does not report the full model specification, terms included, covariance assumptions, diagnostics or sensitivity analyses. It does not report the protocol’s sample-size justification or interim analyses. Multiplicity procedures are also not reported.

The bioequivalence analysis set required completion of at least two periods with both 500 mg treatments and data at the specified time points. Treatment summaries list 35 participants each, whereas safety groups list 36 at risk each. The registry record does not explain the difference, provide reasons for unavailable data, or describe a missing-data method. Those gaps make it difficult to assess potential selection effects or the sensitivity of results to missing observations.

Finally, the study was conducted in healthy adult participants under fed conditions using single doses. The record does not establish how the pharmacokinetic estimates apply outside that setting. The limited safety counts likewise describe only the reported study exposure and observation, not a comprehensive safety profile.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Bioequivalence (BE) Study of Test Griseofulvin 500 Milligram (mg) Tablets Versus Reference and Dose Proportionality Study of Test Griseofulvin 250 mg and 500 mg Tablets Under Fed Conditions — NCT04318535.
• Registry record last updated 2020-12-28; status last verified 2020-11.
• Results first posted 2020-12-28.
• 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

NCT04318535 is a completed Phase 1 randomized, open-label, three-treatment crossover study sponsored by GlaxoSmithKline. The study enrolled 36 healthy adult participants and evaluated Griseofulvin tablets under fed conditions, including a 500 mg test product, a 250 mg test product and a 500 mg reference product.

The primary reported pharmacokinetic comparison was Cmax for the 500 mg test product versus the 500 mg reference product. The adjusted geometric mean ratio was 1.157, with a 90% confidence interval of 1.103 to 1.213. A dose-proportionality result for the test products is not reported in the registry record. No serious adverse events or deaths were recorded; the registry lists 1 non-serious adverse event for the reference product and none for the test products.

The record does not report the acceptance boundaries required to judge the Cmax result against a bioequivalence criterion. It also does not provide a detailed missing-data account, full model specification or multiplicity plan.