Study of Changes in Hepatic Fat Following Administration of MK-4074 and Pioglitazone Hydrochloride (MK-4074-008): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Changes in Hepatic Fat Following Administration of MK-4074 and Pioglitazone Hydrochloride (MK-4074-008) (NCT01431521), a randomized phase 1 study. The primary outcome is Percent Change From Baseline in Hepatic Fat.

2. Clinical Question

3. Trial Design

NCT01431521 is a completed Phase 1 study sponsored by Merck Sharp & Dohme LLC. The registry describes it as randomized, double-masked, and parallel-group. Enrollment was 31. The study began on 2011-10-26, and the primary completion date was 2012-09-18. The registry identifies Non-alcoholic Fatty Liver Disease as the condition under study.

The intervention descriptions list oral MK-4074 200 mg twice daily for 4 weeks and a matching placebo, as well as oral pioglitazone hydrochloride 30 mg once daily for 4 weeks and a matching placebo. The primary outcome results list three groups: MK-4074, Pioglitazone, and Placebo. The registry record does not report how the matching-placebo assignments relate to those three outcome groups, so the design should not be characterized as a particular factorial or combined-treatment arrangement.

Hepatic fat was assessed by magnetic resonance imaging before the first dose and after 4 weeks of treatment. The endpoint was analyzed in the Per-Protocol population. The registry lists 10.0 participants for each of the three primary outcome groups. The disposition of all enrolled participants and the reason the endpoint population differs from overall enrollment are not reported in the registry record.

The registry lists two secondary outcomes: percent change from baseline in alanine transaminase (ALT) and aspartate transaminase (AST). It records five outcome measures and seven analyses. Further details about randomization procedures, masking procedures, allocation ratios, and the sample-size justification are not reported in the registry record.

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: Per-Protocol (PP) Population consists of those participants who comply with the protocol sufficiently to ensure that these data will be likely to exhibit the effects of treatment, according to the underlying scientific model.

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 Percent Change From Baseline in Hepatic Fat, measured over Baseline and Week 4.

The primary endpoint is percent change from baseline in hepatic fat, measured from baseline to Week 4. Magnetic resonance imaging was performed before the first dose and after 4 weeks of treatment. The registry says hepatic fat fraction was calculated separately for each of the 9 liver regions and then averaged to obtain an overall percent change for each participant.

This endpoint expresses the change relative to each participant’s baseline value, rather than reporting only an end-of-treatment measurement. A negative percent change indicates a decrease from baseline; a positive percent change indicates an increase. Group estimates are least-squares means with 95% confidence intervals, and the registry’s between-group analyses report least-squares mean differences.

As a general statistical point, a continuous imaging measure can retain information across the range of observed values rather than reducing outcomes to categories. MRI-based hepatic fat assessment is the measure selected for this study’s primary question about change in liver fat. The registry record does not report the MRI acquisition protocol, image-reading procedures, or any validation details for the measurement method.

The endpoint is defined for the Per-Protocol population, not explicitly for every enrolled participant. How missing measurements were handled and whether endpoint values were available for participants excluded from that population are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry reports least-squares mean estimates and a linear mixed effects model for the between-group analyses. Least-squares means are model-based group estimates. They can differ from simple arithmetic averages because the model may account for factors or structure specified in the analysis. The registry record does not describe the model’s covariates, repeated-measure structure, or other specification details, so those features cannot be reconstructed from the reported description alone.

What does a confidence interval communicate?

A confidence interval conveys uncertainty around an estimated quantity under the model and its assumptions. A narrower interval indicates greater precision on the scale of the estimate, while a wider interval indicates less precision. It does not describe the spread of individual outcomes, and it does not guarantee that the true value lies inside the particular interval. The reported intervals accompany both the arm-level least-squares means and the between-group mean differences.

For example, the MK-4074 arm estimate is -35.73, with a 95% confidence interval from -44.53 to -26.93. The Placebo estimate is 8.63, with a 95% confidence interval from -0.17 to 17.43. These intervals apply to the estimated group means. The comparison estimates have their own confidence intervals, which quantify uncertainty in the model-based difference between groups.

How should the mean differences be read?

A least-squares mean difference compares the model-estimated mean change in one group with that in another. The registry reports -44.37 for MK-4074 versus Placebo, -26.67 for Pioglitazone versus Placebo, and -17.69 for MK-4074 versus Pioglitazone. The sign depends on the order used to define the contrast. In these posted results, the negative contrasts indicate a lower percent change for the first-named group than for its comparator. These differences are in the endpoint’s percent-change units; they are not ratios or percentages of participants.

The reported p-values assess how compatible the observed contrast is with a null hypothesis of no difference under the model and its assumptions. A small p-value is not a measure of effect size, clinical importance, or the probability that the null hypothesis is true. The record reports p < 0.001 for each comparison with Placebo and p = 0.007 for the MK-4074 versus Pioglitazone comparison. Whether those tests were adjusted for multiple comparisons is not reported in the registry record.

Why use a mixed effects model?

In general, mixed effects models can represent structured observations and account for modeled sources of variation. They may be useful when observations are related or when the study design includes grouping or repeated measurements. The registry names a linear mixed effects model but does not report enough detail to determine which model features were used in this analysis. No particular adjustment, covariance structure, or random effect should therefore be assumed.

The imaging endpoint was formed by calculating percent change for each of the 9 liver regions separately and averaging those values to create an overall participant-level measure. The registry then reports model-based least-squares means for the Per-Protocol population. The record does not say whether any additional transformations, outlier handling, or missing-data procedures were applied.

What do the planning illustrations mean?

General sample-size illustrations use the observed effect as a planning reference, not as observed power. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. The illustrations are not the trial’s own sample-size justification. The registry record does not report the trial’s sample-size justification, and no numerical planning requirements are reported in the registry record. Such illustrations describe hypothetical future planning conditions and do not change the uncertainty in the results observed in this trial.

Because the registry’s outcome is continuous, sample-size illustrations of this type concern participants per arm. They should not be interpreted as evidence about the trial’s realized power. The trial’s enrollment and analyzed group sizes are separate details from any planning requirement.

10. Analyses Reported by the Sponsor

The registry’s posted between-group results use a linear mixed effects model and report least-squares mean differences with 95% confidence intervals and p-values. Those are the sponsor-reported analysis results available in the registry: -44.37 for MK-4074 versus Placebo, -26.67 for Pioglitazone versus Placebo, and -17.69 for MK-4074 versus Pioglitazone.

The arm-level least-squares means are also reported: -35.73 for MK-4074, -18.04 for Pioglitazone, and 8.63 for Placebo. A simple unadjusted recalculation from arm-level summaries would not necessarily reproduce a model-based contrast, because the model may incorporate analysis features not reflected in a direct comparison of group estimates. The registry does not provide enough model detail to reconstruct that analysis independently.

The registry record does not report an additional recalculated comparison to set alongside these model estimates. The reported analysis should therefore be read as the posted inferential result, with its stated model, confidence intervals, and p-values. The record does not identify a discrepancy between the posted analysis and a recalculation.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures and seven analyses, including the primary hepatic-fat outcome and two secondary outcomes: percent change from baseline in ALT and AST. It reports three pairwise comparisons for the primary endpoint. This creates a context in which multiplicity may matter: when several outcomes or comparisons are tested, the chance of at least one apparently notable result can increase if each test is treated independently.

Whether the analyses were prespecified in a hierarchy, whether any adjustment for multiple testing was used, and which comparisons were designated as confirmatory are not reported in the registry record. The reported p-values should therefore be interpreted as posted, without assuming a multiplicity adjustment or an unreported testing strategy.

The primary endpoint is defined for the Per-Protocol population. The record describes that population as participants whose protocol compliance was sufficient for their data to be likely to exhibit treatment effects according to the underlying scientific model. The registry does not report the criteria used to determine sufficient compliance, the number excluded from this population, or the reasons for exclusion.

Enrollment was 31, while the primary outcome groups each list 10.0 participants. The registry record does not report the disposition of all enrolled participants or explain the relationship between total enrollment and the group counts in the endpoint results. Although the record describes the primary endpoint timeframe as Baseline and Week 4, the amount or pattern of missing endpoint data is not reported.

Missing-data assumptions, imputation methods, sensitivity analyses, and handling of participants with incomplete measurements are not reported in the registry record. These omissions limit assessment of how robust the estimated results are to missingness or the choice of analysis population. The registry also does not report interim analyses or stopping rules.

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 serious adverse event figures report zero affected participants in each listed group: MK-4074, Placebo for MK-4074, Pioglitazone, and Placebo for Pioglitazone. The numbers at risk for serious adverse events are 10, 5, 10, and 5, respectively. These counts describe the registry’s reported safety groups; they are distinct from the primary endpoint group labels.

Zero serious adverse events reported in these groups means that none were recorded among the participants counted in the registry’s serious-event figures. It does not establish that serious events could not occur, particularly given the limited number of participants. The registry lists no top serious adverse event terms. Death counts are not reported in the registry record.

For other adverse events, the reported affected-participant counts are 4 among 10 at risk for MK-4074, 1 among 5 at risk for Placebo for MK-4074, 4 among 10 at risk for Pioglitazone, and 4 among 5 at risk for Placebo for Pioglitazone. These are counts, not evidence by themselves of a causal relationship or a comparative safety conclusion. The registry record does not provide event descriptions or further details needed to characterize these other events.

Safety counts should be interpreted separately from the hepatic-fat estimates. They do not establish the frequency of events beyond the reported groups or provide information about longer-term safety.

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

The posted data show that the model-estimated mean percent change in hepatic fat differed between the listed groups over the study’s 4-week assessment. The registry reports negative least-squares mean differences for MK-4074 versus Placebo and Pioglitazone versus Placebo, with p < 0.001 for both comparisons. It also reports a difference between MK-4074 and Pioglitazone, with p = 0.007. The estimates and confidence intervals describe the analyzed participants and the reported model.

These results do not establish that the same changes would occur in every participant. A group mean can conceal substantial variation among individuals, and the registry does not report the distribution of individual endpoint values. The confidence intervals describe uncertainty in estimated group quantities, not individual response ranges.

The results do not establish long-term changes, because the primary endpoint was assessed at Week 4 and longer-term hepatic-fat measurements are not reported in the registry record. Nor do the results establish whether the measured changes correspond to clinical outcomes or have a particular clinical importance; that interpretation is not reported in the registry record.

The study is described as Phase 1 and enrolled 31 participants. The endpoint results list 10.0 participants in each of three groups, and the analysis population was Per-Protocol. The size and population limit how broadly these estimates can be generalized. Further details about participant characteristics, missing measurements, protocol deviations, and the broader study population are not reported in the registry record.

The study’s stated expectation of an approximately 50% reduction in hepatic fat with MK-4074 was a hypothesis. The reported endpoint is mean percent change from baseline, and the reported comparisons are model-based mean differences. These quantities should be kept distinct. The findings describe the posted short-term data and do not, on their own, establish a treatment’s general clinical effect.

14. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates how a continuous imaging endpoint can be summarized in several related ways: group-level least-squares means, confidence intervals around those means, and model-based differences between groups. Keeping those quantities distinct helps avoid reading a between-group contrast as though it were the change within a single group, or interpreting a mean as an individual participant’s result.

The endpoint also has a structured measurement process. Hepatic fat fraction was assessed in each of the 9 liver regions, and the regional percent changes were averaged to create an overall participant-level outcome. That construction matters when interpreting what the final mean represents: it is based on a specified aggregation of measurements rather than on a single region.

The study also highlights the importance of the analysis population. The primary outcome is reported for a Per-Protocol population, while total enrollment was 31 and the three endpoint groups each list 10.0 participants. Without further disposition and missing-data details, readers cannot fully trace how enrolled participants contributed to the primary analysis.

Finally, the record illustrates why model details and multiplicity plans matter. It reports a linear mixed effects model, three primary comparisons, and two secondary outcomes, but does not report the full model specification or whether multiplicity was addressed. These details affect how much can be inferred from p-values and how readily an independent analyst can reproduce the reported estimates.

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.

Several trial-specific limitations constrain interpretation. The study enrolled 31 participants, and the primary outcome results list 10.0 participants in each of three groups. The registry does not report how all enrolled participants were allocated, whether any did not contribute to the endpoint analysis, or why the Per-Protocol population has the reported group counts. The group results therefore cannot be fully reconciled with overall enrollment from the registry record alone.

The primary endpoint is assessed at Baseline and Week 4. Longer-term measurements are not reported in the registry record, so persistence of the measured changes cannot be assessed from these results. The study is Phase 1, and the registry does not report details that would support a broader assessment of how representative the analyzed population is of other populations with fatty liver disease.

The registry identifies a linear mixed effects model but does not report its detailed specification, covariates, or handling of missing data. It also does not report the multiplicity strategy for the three primary comparisons and secondary outcomes. These gaps make it difficult to independently reproduce the inferential analysis or assess sensitivity to alternative model choices.

The results are reported as least-squares means and confidence intervals, but the registry does not report individual-level distributions or additional measures of variability. The record also does not describe MRI acquisition or reading procedures. The protocol compliance criteria for the Per-Protocol population are not reported in detail. These limitations are specific to what the registry record documents and should not be taken as evidence that unreported procedures were or were not conducted.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study of Changes in Hepatic Fat Following Administration of MK-4074 and Pioglitazone Hydrochloride (MK-4074-008) — NCT01431521.
• Registry record last updated 2018-09-10; status last verified 2018-08.
• Results first posted 2016-03-25.
• 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

NCT01431521 is a completed, randomized, double-masked, parallel-group Phase 1 study sponsored by Merck Sharp & Dohme LLC. It enrolled 31 participants with Non-alcoholic Fatty Liver Disease. The study assessed change in hepatic fat by MRI from baseline to Week 4 after multiple-dose administration of MK-4074, pioglitazone hydrochloride, or placebo.

The primary endpoint was percent change from baseline in hepatic fat in the Per-Protocol population. The registry reports least-squares mean estimates of -35.73 for MK-4074, -18.04 for Pioglitazone, and 8.63 for Placebo. Reported model-based comparisons were -44.37 for MK-4074 versus Placebo, -26.67 for Pioglitazone versus Placebo, and -17.69 for MK-4074 versus Pioglitazone.

ALT and AST percent change were listed as secondary outcomes. The registry does not report detailed model specifications, multiplicity procedures, missing-data methods, or participant disposition beyond the posted outcome group counts.