Dose Exploration Study of GSK4532990 in. Participants With NASH or Suspected NASH: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Dose Exploration Study of GSK4532990 in. Participants With NASH or Suspected NASH (NCT06104319), a randomized phase 2 study. The primary outcome is Predicted Mean Percent Change From Baseline (CFB) in Liver Biopsy-derived HSD17B13 Protein Expression Levels.

2. Clinical Question

3. Trial Design

This was a Phase 2a, randomized, open-label, sequential, single-dose dose exploration study of GSK4532990. It enrolled 61 participants and is listed as completed. The registry gives a start date of 2024-01-22 and a primary completion date of 2025-09-23. The expected total study duration for each participant was approximately 24-36 weeks.

Participants were assigned among four GSK4532990 dose groups: 25 mg, 200 mg, 400 mg, and 800 mg. The record labels allocation randomized and the study model sequential, but it does not report the sequence or rules used to move among doses, the allocation ratio, or the number assigned to each group. It also does not report a placebo group. With no masking, participants and study personnel were not blinded to treatment assignment according to the registry designation.

The stated purpose was to characterize how GSK4532990 is processed in the body and how it acts in the liver, while assessing safety and tolerability. The primary endpoint focuses on a predicted change in liver biopsy-derived HSD17B13 protein expression. Secondary outcomes include pharmacokinetic measures, HSD17B13 protein expression, and HSD17B13 mRNA expression. These endpoints address biological activity and drug exposure rather than a clinical outcome such as symptoms or disease progression.

The registry reports six outcome measures and zero analyses. It does not report a detailed statistical analysis plan, the rationale for the total enrollment, a planned allocation by dose, or whether additional interim assessments guided the sequential design. Therefore, the design features can be described, but the operational dose-selection process and the planned statistical comparisons cannot be reconstructed from 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: HSD17B13 Protein Concentration population included all participants in the enrolled population with evaluable baseline biopsy protein assessment. Multiple Monte Carlo simulations of the PKPD model were performed on six participants per dose for the different timepoints. Simulations incorporated individual and residual variability and accounted for parameter uncertainty to generate the predicted mean and 95% CI

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 Predicted Mean Percent Change From Baseline (CFB) in Liver Biopsy-derived HSD17B13 Protein Expression Levels, measured over Baseline (Day 1), At weeks 8, 12 and 28.

The primary endpoint is the predicted mean percent change from baseline in liver biopsy-derived HSD17B13 protein expression. Baseline is defined as the pre-dose value on Day 1. The stated assessment times are weeks 8, 12, and 28. The endpoint expresses the modeled change relative to baseline, rather than a direct clinical outcome.

The registry says the predictions come from a nonlinear mixed-effect pharmacokinetic-pharmacodynamic target engagement model using HSD17B13 protein concentrations from pre-dose and post-dose liver biopsy samples. A target engagement measure can help characterize whether a drug is associated with changes in a biological marker related to its intended activity. In general, modeling can combine information about drug exposure and marker measurements to describe the relationship between them. Those are general statistical purposes; the registry does not provide enough detail to assess the model’s assumptions or performance.

The endpoint is reported conceptually as a predicted mean with a 95% confidence interval. The registry record does not report the mean estimates or interval limits for any dose or assessment time. Consequently, the direction, magnitude, and precision of the primary endpoint cannot be evaluated from the available record. This marker endpoint also does not, on its own, establish a change in a participant’s clinical condition.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary measure as a mean of a continuous outcome and specifies a 95% confidence interval as its dispersion measure. It also describes the endpoint as predicted from a nonlinear mixed-effect pharmacokinetic-pharmacodynamic target engagement model. The distinctions among an observed measurement, a model-based prediction, and uncertainty around an estimate matter when interpreting this type of endpoint.

What does a mean describe?

In general, a mean is a summary of the average value in a defined set of measurements or predictions. For a continuous endpoint such as percent change from baseline, the mean condenses individual values into a single location estimate. That summary is useful for describing the group-level pattern, but it does not show the full distribution of participant responses. A group mean can coexist with substantial variation among individuals. The registry calls this endpoint a predicted mean; it should therefore be distinguished from a simple arithmetic mean of directly observed changes unless the analysis description says otherwise.

What does the confidence interval represent?

In general, a confidence interval describes uncertainty in an estimated quantity under the assumptions of the method used to calculate it. Its width reflects the estimate’s statistical precision: a wider interval indicates more uncertainty than a narrower interval, all else being equal. A confidence interval is not a range containing every participant’s response, nor does it directly describe the probability that the fixed true value lies within a particular calculated interval. Here, the registry specifies 95% confidence intervals but does not report their limits. Their widths and the resulting precision cannot be assessed.

Why use a pharmacokinetic-pharmacodynamic model?

In general, pharmacokinetic methods describe drug exposure over time, while pharmacodynamic methods relate exposure to a biological response or marker. A nonlinear mixed-effect model can represent a relationship that does not follow a straight line and can account for variation among individuals through model components. The registry says the primary predictions used HSD17B13 protein concentrations from pre-dose and post-dose liver biopsy samples and came from a target engagement model. This description does not identify the model’s equation, covariates, parameter estimates, diagnostic checks, or handling of observations below measurement limits; those details are not reported in the registry record.

The stated endpoint is a predicted percent change relative to baseline. In general, expressing change relative to a baseline measurement can help describe the degree of change against each participant’s starting level. But the estimate may depend on how baseline values are measured and on the statistical model used to combine repeated observations. The registry defines baseline as the pre-dose value on Day 1, but further details about baseline adjustment and the analysis dataset are not reported.

How should simulated estimates be understood?

The registry states that multiple Monte Carlo simulations of the pharmacokinetic-pharmacodynamic model were performed on six participants per dose for different time points. In general, Monte Carlo simulation uses repeated draws from specified distributions or model components to examine the range of predictions implied by a model. Simulation can propagate uncertainty or represent variability, depending on how it is specified. It does not mean that the simulated values are additional participants or newly observed biopsy results. The record’s description is truncated after noting that simulations incorporated individual-level information, so the remaining simulation details are not reported in the registry record.

The stated six participants per dose is part of the simulation description; it should not be mistaken for the total number randomized to a dose or for the full study enrollment. The record does not explain how those participants were selected for the simulations, whether the same participants contributed at each time point, or how uncertainty in model parameters was represented. These unknowns limit independent assessment of the predicted means.

What can the dose groups establish?

Randomized allocation can support comparisons between assigned groups when the analysis and allocation details are clear. Here, the groups received different doses of the same drug, and the trial was open-label. The registry does not report the allocation ratio, group sizes, or a named statistical comparison between dose levels. It also does not give a single estimate contrasting one dose with another. The design description alone therefore does not establish the statistical strength of a dose-response comparison.

The word sequential identifies the recorded trial model, but it is not enough to determine the operational sequence. In general, sequential dose designs may use information from earlier stages when proceeding to later stages; however, the registry does not state whether or how that occurred in this study. No particular dose-escalation rule, stopping boundary, or interim decision process can be inferred from that label.

What is missing from the statistical account?

The registry reports six outcome measures and zero analyses. It does not report numerical results for the primary endpoint, confidence interval limits, p-values, or dose-specific estimates. The statistical analysis plan, model diagnostics, multiplicity procedures, missing-data method, and planned sample-size justification are also not reported in the registry record. General statistical principles can explain why these details matter, but their absence prevents assessment of whether the planned analyses were appropriate or how robust the reported results would be.

Sample-size illustrations, when based on an observed effect, are planning calculations rather than observed trial outcomes or observed power. No numerical planning illustration can be interpreted without the estimate and assumptions used in that calculation. More generally, precision depends on the data available and the analysis; a study’s enrollment count by itself does not tell how precise a modeled endpoint estimate is.

10. Multiplicity, Missing Data and Analysis Populations

The primary endpoint includes predictions at weeks 8, 12, and 28 and spans four dose groups. Multiple doses and assessment times create several possible comparisons or interpretations. In general, when multiple statistical claims are made, the chance of at least one false-positive finding can increase unless the analysis plan controls or otherwise addresses multiplicity. The registry does not report a multiplicity adjustment, testing hierarchy, prespecified primary time point, or strategy for summarizing results across doses and times.

The record lists six outcome measures, including AUC0-t, Cmax, changes in HSD17B13 protein expression, and changes in HSD17B13 mRNA expression. It reports zero analyses and does not specify which outcomes were tested, whether any analyses were confirmatory or exploratory, or how results across outcomes were to be interpreted. No multiplicity method should be assumed.

The primary endpoint population is described as enrolled participants with an evaluable baseline biopsy protein assessment. That statement identifies a criterion for inclusion in the stated population, but it does not give the number meeting the criterion or explain what happened to participants without evaluable samples. The registry does not report a missing-data method, imputation approach, sensitivity analysis, or reasons for missing endpoint measurements. It is therefore not possible to assess whether missing biopsy data could have affected the estimates.

The record also does not fully define the analysis populations for the pharmacokinetic, pharmacodynamic, or safety outcomes. The enrollment total is 61, while the model description says simulations were performed on six participants per dose at different time points. The relation between that simulation set and the analysis population is not explained. Allocation counts, exclusions, and participant flow by dose are not reported in the registry record. Without those details, readers cannot determine precisely who contributed to each endpoint or how broadly the estimates represent the enrolled population.

11. 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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hepatic enzyme increased (6.7%), chest pain (5.9%), scrotal cellulitis (5.9%).

The registry reports serious adverse events separately from other adverse events and deaths for each dose group. In the 25 mg group, 1 of 17 participants at risk had a serious adverse event; in the 200 mg group, 0 of 15; in the 400 mg group, 1 of 15; and in the 800 mg group, 0 of 14. The reported death count was 0 in each group. These are group-level counts as recorded, not estimates of comparative risk.

The record lists “Hepatic enzyme increased” at 6.7%, “Chest pain” at 5.9%, and “Scrotal cellulitis” at 5.9% among the top serious adverse event terms. It does not identify the dose group for each listed term in the available information. The percentages should therefore not be assigned to a particular group based on the group counts.

In general, adverse-event counts depend on the definition of an event, the period of observation, and how events are collected and coded. The registry record does not provide additional detail about event duration, severity grading, relatedness assessments, or exposure-adjusted rates. These data describe reported events during the study; by themselves, they cannot establish causation or support a reliable comparison of safety between doses. The absence of a reported death in these groups is a description of the posted count, not a guarantee about risk beyond the study.

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

The study was designed to explore single-dose GSK4532990 across four dose groups and assess pharmacokinetics, pharmacodynamics, safety, and tolerability in adults with NASH or suspected NASH. The primary endpoint was a modeled prediction of mean percent change from baseline in liver biopsy-derived HSD17B13 protein expression, with assessments at weeks 8, 12, and 28. These details describe the question the study set out to examine.

The registry record does not report the numerical primary endpoint estimates or confidence intervals. It therefore does not show the size, direction, or precision of modeled changes in HSD17B13 protein expression. Nor does it provide numerical pharmacokinetic or secondary biomarker results in the record information available. No conclusion about the magnitude of biological activity or a pattern across doses can be drawn from the endpoint description alone.

The endpoint is a biological marker assessed in liver biopsy samples and analyzed through a pharmacokinetic-pharmacodynamic model. In general, a marker can help characterize biological response, but a marker result is not automatically equivalent to a change in symptoms, disease course, or another clinical outcome. The registry does not report a clinical efficacy endpoint in the listed primary or secondary outcomes. Accordingly, the study’s stated biomarker focus should not be interpreted as evidence of a clinical benefit or lack of benefit.

The posted safety counts document serious and other adverse events in the dose groups, along with zero reported deaths in each group. Those observations are useful descriptions of what was recorded in this study, but the available record does not supply the detail needed to attribute events to treatment or make a definitive comparison across doses. No placebo group is listed, and the study was open-label.

In general, a completed study can still have limitations in the registry results available to readers if outcome estimates or analysis details are not reported. Here, the design and endpoint definitions are more fully described than the numerical efficacy results. The absence of numerical results in the registry record should be read as an information gap, not as a finding of no effect. The conclusions that can be drawn are consequently limited to the study’s design, planned measurements, and posted safety counts.

13. Why This Trial Matters Statistically

This study raises several statistical questions that are important in early dose exploration. It is randomized across four doses, but it is also open-label and described as sequential. Those features make it important to know how allocation was implemented, whether decisions about later dose groups depended on earlier observations, and how comparisons were planned. The registry record does not provide those operational details, so the label sequential cannot be translated into a particular design or analysis procedure.

The primary endpoint illustrates the distinction between direct measurement and model-based inference. The measure is a predicted mean percent change in biopsy-derived protein expression, obtained from a nonlinear mixed-effect pharmacokinetic-pharmacodynamic model. This can be a useful way to connect drug exposure to a biological marker, but interpretation depends on model specification, data quality, and uncertainty. Numerical predictions, confidence intervals, and model checks are not reported in the registry record, leaving the principal statistical result unavailable for assessment.

The endpoint is also considered at several times and across multiple doses, alongside additional pharmacokinetic and biomarker outcomes. In general, this creates choices about which contrasts deserve emphasis and how to address multiple comparisons. Without a reported statistical plan, readers cannot determine whether the analysis controlled multiplicity or which results were intended to be primary beyond the stated endpoint.

Finally, the record shows why reporting the analysis population and missing-data approach matters. The primary endpoint population requires an evaluable baseline biopsy assessment, while the model description refers to simulations on six participants per dose. The relationship between these descriptions, and the number of participants contributing to each estimate, is not reported. These details affect how readers judge the scope and precision of the findings. The trial is statistically informative as a design example, but the missing numerical outcome results prevent evaluation of its principal estimate.

14. 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 main limitation for interpreting this trial from the registry record is the lack of numerical primary endpoint results. Although the predicted mean percent change and 95% confidence interval are specified, neither estimates nor interval limits are reported. This prevents assessment of effect direction, magnitude, precision, and differences across dose groups or time points.

The record does not report a complete statistical analysis plan, model equation, model diagnostics, parameter estimates, or sensitivity analyses. The simulation description states that six participants per dose were used for different time points, but the rest of that description is incomplete. How participants were selected, whether the same participants contributed across times, and how uncertainty was propagated are not reported in the registry record.

Participant flow and dose-specific enrollment are also unavailable. The total enrollment was 61, but the number randomized to each group, the number with evaluable biopsy samples, and the number contributing to each analysis are not reported. The registry does not state how missing biopsy results were handled or whether analyses used a particular missing-data assumption. This limits evaluation of selection and missing-data effects.

The study was open-label and lists no placebo or untreated comparator. The registry record also does not report the allocation ratio, sequence rules for the sequential design, a sample-size justification, or a multiplicity strategy. These omissions make it difficult to reconstruct the intended inferential framework or assess how dose comparisons were to be interpreted.

Safety reporting includes group counts and selected serious-event terms, but the available record does not provide detailed definitions, event timing, relatedness assessments, or exposure-adjusted comparisons. These limitations do not negate the observations that are posted; they restrict the conclusions that can be drawn from them.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Dose Exploration Study of GSK4532990 in. Participants With NASH or Suspected NASH — NCT06104319.
• Registry record last updated 2026-10-27; status last verified 2026-10.
• Results first posted 2026-10-27.
• 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.

18. Record Summary

NCT06104319 is a completed Phase 2a randomized, open-label, sequential, single-dose study of GSK4532990 in adults with NASH or suspected NASH. The record lists 61 participants and four dose groups: 25 mg, 200 mg, 400 mg, and 800 mg. Its primary endpoint is the model-predicted mean percent change from baseline in liver biopsy-derived HSD17B13 protein expression at weeks 8, 12, and 28.

The registry describes a nonlinear mixed-effect pharmacokinetic-pharmacodynamic target engagement model but does not report numerical primary endpoint estimates or confidence intervals. It lists six outcome measures and zero analyses. Serious adverse event counts were 1, 0, 1, and 0 across the dose groups, respectively; reported deaths were 0 in each group. Missing-data methods, multiplicity procedures, detailed analysis populations, and the statistical analysis plan are not reported in the registry record.