Study to Evaluate the Effect of GWP42003 on Liver Fat Levels in Participants With Fatty Liver Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study to Evaluate the Effect of GWP42003 on Liver Fat Levels in Participants With Fatty Liver Disease (NCT01284634), a randomized phase 2 study. The primary outcome is Percent Change From Baseline To The End Of Treatment (EOT) In Mean Liver Triglyceride Levels.

2. Clinical Question

3. Trial Design

This was a Phase 2, randomized, parallel-group, placebo-controlled pilot and dose-ranging study sponsored by GW Research Ltd. The registry record lists enrollment of 25 and describes the condition as fatty liver. The study status is completed. Its recorded start date is 2011-05-03 and its primary completion date is 2012-07-13.

Participants were assigned to one of three daily GWP42003 dose groups or placebo. The dose groups were 200 mg/day, 400 mg/day, and 800 mg/day. Each treatment description specifies self-administration twice daily for 8 weeks. The GWP42003 capsule count varied by assigned dose; placebo capsules were described as matching the active capsules. Doses were scheduled before breakfast and before the evening meal, typically about 12 hours apart, according to the registry descriptions.

The design is described as randomized and parallel. The official title calls the study partially blind, while the registry masking information lists quadruple masking. These descriptions are both present in the record; the exact people who were masked, and how masking was implemented or maintained, are not reported in the registry record.

The primary outcome was a continuous measure: percent change from baseline to the end of treatment in mean liver triglyceride levels. The reported endpoint timeframe was baseline to EOT (Day 57) or early termination. The primary analysis set included randomized participants who received at least one dose and had on-treatment efficacy data; participants with no relevant post-randomization efficacy data were excluded. Although overall enrollment is reported, the record does not provide a single primary-analysis total separate from the arm-level outcome counts.

The record lists five secondary outcomes concerning serum cholesterol measures and triglycerides, and six outcome measures overall. It reports three pairwise analyses of each active dose against placebo for the primary outcome. A detailed protocol-level statistical analysis plan, including any planned adjustment for multiple comparisons, is 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: ITT analysis set: All participants who were randomized, received at least one dose of study medication, and had on-treatment efficacy data. Participants who did not have any relevant post-randomization efficacy data were excluded from the analysis.

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.

Population: Safety Analysis Set: Participants who received at least 1 dose of investigational medicinal product (IMP); analyzed as per actual treatment received.

6. Primary Endpoint

The registry lists the primary outcome as Percent Change From Baseline To The End Of Treatment (EOT) In Mean Liver Triglyceride Levels, measured over Baseline to EOT (Day 57) or Early Termination (ET).

The primary endpoint was percent change from baseline to the end of treatment in mean liver triglyceride levels. Liver triglyceride is the quantity being measured; the endpoint expresses how the group mean changed relative to baseline. The measure was obtained using magnetic resonance imaging or magnetic resonance scanning, according to the registry description. The timeframe was baseline to EOT (Day 57) or early termination.

The direction of the endpoint matters for interpretation. The registry record defines a reduction from baseline, represented by a negative value, as an improvement in condition. A positive value represents an increase from baseline. The posted arm summaries report a mean and a standard deviation for each group. The standard deviation describes spread among the observations and is not itself a measure of the precision of the estimated group mean.

Using an imaging-based liver-fat measure gives the trial a continuous endpoint that can capture changes across a range rather than only recording whether a threshold was crossed. In general, continuous outcomes can provide information about both the direction and magnitude of change. That general advantage does not remove uncertainty: estimates depend on the number of participants with evaluable data, the variability of their measurements, and the analysis used to compare groups.

The primary endpoint is distinct from the secondary serum lipid outcomes. Changes in cholesterol or serum triglycerides are not interchangeable with changes in liver triglyceride levels. The registry record identifies the outcome and its measurement approach, but does not report additional details about image reading, measurement repeatability, or how missing endpoint measurements were handled beyond the stated analysis-population definition.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous, and the registry reports group means and standard deviations. In general, a mean describes the central value of a set of measurements, while a standard deviation describes how dispersed individual measurements are around that mean. A standard deviation is not a confidence interval and does not directly state how precisely the group mean or the between-group difference has been estimated. Precision also depends on the amount of data available.

What does the reported mean difference represent?

The registry lists three analyses using linear regression and labels the parameter as “Mean Difference (Final Values).” Each compares one GWP42003 dose group with placebo. A mean difference is a contrast between group values: negative estimates in these comparisons indicate a lower final value in the dose group relative to placebo under the reported contrast. The record does not provide enough detail to establish whether the regression included covariates, how the model was parameterized, or whether any baseline adjustment was used. Those details are not reported in the registry record.

The arm summaries are described as percent change from baseline in group mean levels, whereas the pairwise analysis parameter is described as a difference in final values. The record does not explain the full relationship between these summaries and the regression estimand. These results therefore should not be treated as interchangeable calculations. The reported regression estimates and their confidence intervals are the registry's stated comparisons; the arm means provide descriptive context.

How should the confidence intervals be read?

A confidence interval expresses uncertainty around an estimated contrast under the statistical model and its assumptions. A narrower interval indicates greater precision than a wider interval, all else being equal. Here, each reported interval spans zero, so the estimates do not rule out zero difference at the stated 90% confidence level. They also include values on both sides of zero, which means the data are compatible with differing directions of effect under the reported analysis. The intervals should not be interpreted as a range containing a fixed proportion of individual participant outcomes.

The 90% confidence level is the level reported for all three comparisons. The associated p-values are 0.222, 0.133, and 0.302. In general, a p-value measures how unusual data at least as extreme as those observed would be under a specified null model and analysis assumptions. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical relevance of a difference. The registry record does not report a prespecified decision threshold for clinical importance.

Why does small sample size matter?

The primary outcome groups contain 7.0, 6.0, 6.0, and 5.0 observations, respectively. With limited observations, the estimate of a group mean and the contrast between groups can be strongly influenced by individual measurements. This can produce wide uncertainty even when the observed means appear different. It also limits the ability to characterize the distribution of the outcome or to examine whether a small number of unusual measurements affected the estimates.

In general, variability reduces precision because a noisier outcome makes it harder to distinguish a systematic difference from random variation. The reported standard deviations vary across arms, and the placebo standard deviation is 17.97. No distributional diagnostics or participant-level measurements are reported in the registry record. Consequently, the assumptions underlying the linear regression results cannot be assessed from the posted summary information alone.

What does the planning illustration mean?

Sample-size illustrations based on an observed effect are planning exercises, not observed power. The stated assumptions for such illustrations are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is defined as a fraction of the observed effect. In general, these calculations ask how much information a future study might need under selected assumptions. They do not show the probability that this completed trial detected an effect, and they do not establish that the observed effect is stable or clinically important.

The registry record does not report the trial's own sample-size justification or planned target assumptions. It also does not report an interim analysis plan, stopping rules, or a full statistical analysis plan. Those omissions prevent a reader from judging whether the reported analyses follow a prespecified strategy or how the study's design anticipated its precision.

How should multiple comparisons be considered?

Three dose-versus-placebo analyses are reported for the primary outcome. When several hypotheses are tested, the chance of at least one apparently unusual result can increase if no adjustment is made. In general, multiplicity procedures can address this issue, but they may change the interpretation of p-values or confidence intervals. Whether an adjustment was planned or applied is not reported in the registry record. The three comparisons should therefore be read as the three reported estimates, without assuming a multiplicity correction or its absence.

Likewise, five secondary outcomes are posted in addition to the primary outcome. The record does not identify which outcomes or comparisons were designated as confirmatory, nor does it report a hierarchy or familywise error strategy. Selective interpretation across outcomes is a general risk when many analyses are considered, so the primary endpoint and the secondary measures should remain clearly distinguished.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("GWP42003 200 mg/Day Dose", "GWP42003 400 mg/Day Dose", "GWP42003 800 mg/Day Dose", "Placebo")
n   <- c(7, 6, 6, 5)
m   <- c(-0.68, -0.28, 0.65, 6.36)
sp  <- c(4.97, 8.6, 5.28, 17.97)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  GWP42003 200 mg/Day Dose  7  -0.680  4.970  -5.276  3.916
CARM  GWP42003 400 mg/Day Dose  6  -0.280  8.600  -9.305  8.745
CARM  GWP42003 800 mg/Day Dose  6  0.650  5.280  -4.891  6.191
CARM  Placebo  5  6.360  17.970  -15.953  28.673

11. Analyses Reported by the Sponsor

The registry reports three linear-regression comparisons of each GWP42003 dose with placebo. Those are the analyses to use when describing the posted inferential results: mean differences in final values, 90% confidence intervals, and p-values. The group means and standard deviations are descriptive summaries and do not replace those regression comparisons.

The record does not report separate recalculated comparison estimates. It is therefore not possible to set out a numerical recalculation alongside the reported analyses from the registry information. Nor does the record provide enough model detail to reproduce the regression estimates independently from summary statistics alone. The relationship between the descriptive percent-change means and the reported final-value contrasts is not fully explained in the record.

The reported comparisons are imprecise, with confidence intervals spanning zero. No discrepancy between a sponsor analysis and a recalculated analysis is indicated in the registry record. The registry's stated analysis should be distinguished from any independent calculation based on summary data; if a disagreement were present, the sponsor's own analysis would take precedence under the reporting convention used for these results. Since no separate recalculated estimates are reported, the interpretation rests on the registry's stated estimates and their uncertainty.

12. Multiplicity, Missing Data and Analysis Populations

The primary analysis population was defined as randomized participants who received at least one dose of study medication and had on-treatment efficacy data. Participants who had no relevant post-randomization efficacy data were excluded. This is the registry's stated population definition; the record does not provide a separate total for the primary analysis population or describe additional rules for handling missing primary endpoint values.

The outcome counts for the primary endpoint were 7.0 for 200 mg/day, 6.0 for 400 mg/day, 6.0 for 800 mg/day, and 5.0 for placebo. These are the reported group counts for that outcome. The overall enrollment was 25, but enrollment and the endpoint analysis counts are not the same quantity. The reasons that participants may not have contributed to the primary analysis, and the distribution of missingness by reason, are not reported in the registry record.

Three pairwise analyses compare each dose with placebo. Five secondary outcomes are posted, and the record lists six outcome measures overall. In general, testing multiple doses and outcomes raises multiplicity questions because interpreting each result independently can overstate how much evidence a set of tests provides. The registry record does not report whether the analyses were adjusted for multiplicity, whether a hierarchy was specified, or which tests were intended as confirmatory. No adjustment should be assumed from the information recorded.

The record also does not report an interim analysis plan, a prespecified missing-data sensitivity analysis, or details of imputation. The stated exclusion of participants without relevant post-randomization efficacy data identifies one part of the analysis-population rule, but does not explain how other incomplete observations were treated. The implications of missing data depend on why data are missing and on the analysis assumptions; those reasons and assumptions are not reported in the registry record.

The secondary outcomes concern serum total cholesterol, HDL cholesterol, LDL cholesterol, the HDL-to-LDL ratio, and serum triglycerides. They are related measures but are not substitutes for the primary liver triglyceride endpoint. Keeping the endpoint families separate helps avoid treating findings on different measures as though they answered the same question.

13. 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's serious adverse event summary reports zero affected participants in each listed group: 200 mg/day, 400 mg/day, 800 mg GWP42003, and placebo. The corresponding serious-event at-risk counts are 7, 6, 7, and 5. These are small groups, so the absence of a reported serious event in this record should not be read as evidence that serious events are impossible or that safety is established.

The safety summary also reports other adverse events affecting 6 of 7 participants in the 200 mg/day group, 5 of 6 in the 400 mg/day group, 7 of 7 in the 800 mg group, and 5 of 5 in placebo. These figures describe participants affected, not the total number of event episodes or their severity. The registry record does not provide the individual event terms in the registry record for this summary.

For the 800 mg group, the serious-event at-risk count is 7, while the primary endpoint outcome count is 6.0. These counts refer to different summaries; the reason for their difference is not reported in the registry record. Death counts are not reported in the registry record. Safety results from a small pilot study are limited in their ability to characterize uncommon events or compare rates precisely.

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

The posted results describe a small randomized pilot study of three GWP42003 doses compared with placebo in participants with fatty liver disease. For the primary endpoint, the reported group means were negative in the 200 mg/day and 400 mg/day groups and positive in the 800 mg/day and placebo groups. The registry defines negative change as a reduction from baseline. The regression comparisons, however, have broad 90% confidence intervals that include zero, so the estimates remain uncertain.

These results do not establish that any dose differs from placebo, and they do not establish that the groups are equivalent. Failure to show a statistically significant difference is not proof of no difference. Conversely, an observed mean difference is not by itself proof of a treatment effect. The confidence intervals indicate that a range of possible contrasts remains compatible with the data and the reported analysis.

The trial's pilot and dose-ranging description is important context. The record does not report a clinical threshold for an important change in liver triglyceride levels, so the clinical importance of the numerical estimates cannot be determined from the posted results alone. Nor does the record report a formal dose-response test. The pattern of arm means should therefore not be treated as a demonstrated dose-response relationship.

The outcome is an imaging-based liver triglyceride measure over the treatment period specified in the record. It does not, by itself, establish effects on other clinical outcomes or longer-term changes. Secondary serum lipid measures are distinct endpoints, and their results are not included in the primary comparison estimates.

Overall, the registry record supports a cautious statistical reading: it documents the design, the observed summaries, and three uncertain dose-versus-placebo comparisons. It does not provide enough information to resolve the precision limitations, assess all model assumptions, or determine whether the observed differences have clinical importance. The results are descriptive and exploratory in the context of this pilot study, rather than a conclusive answer about treatment effects.

15. Why This Trial Matters Statistically

This trial illustrates why a group mean difference should be read together with its uncertainty. The arm summaries show different observed values, but the reported confidence intervals for every dose-versus-placebo comparison span zero. A point estimate alone can make a result appear more definite than the data warrant; the interval shows how much uncertainty remains under the reported analysis.

It also demonstrates the distinction between a descriptive endpoint summary and an inferential comparison. The group means describe observed changes within each arm, while the regression analyses estimate contrasts between a dose group and placebo. Their stated parameter is a difference in final values, and the registry does not fully document how that analysis relates to the percent-change summaries. Clear separation of these quantities prevents an arm-level pattern from being mistaken for a direct treatment comparison.

The study's small outcome groups and variable measurements make precision a central issue. In general, a pilot study can help characterize an endpoint and produce estimates for planning, but an observed effect used in planning is not proof of a stable effect. Any planning illustration based on these estimates depends on assumptions about the effect to detect, allocation, significance level, and approximation; it is not observed power.

Finally, the record highlights the importance of transparent reporting about analysis populations, missing data, masking, and multiplicity. Some key details are not reported in the registry record. That absence limits what readers can conclude about robustness and prespecification, even when the main estimates and their uncertainty are available. The statistical value of a pilot result lies partly in identifying what is estimated and where more precise, fully documented evidence would be needed.

16. 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 limitations are specific to the information recorded for this trial. The primary outcome counts are small, ranging from 5.0 to 7.0 across groups. With few observations, estimates may be sensitive to individual values, and the standard deviations and confidence intervals indicate that the comparisons are uncertain. Participant-level data are not reported, so the influence of individual observations cannot be assessed.

The registry describes the endpoint as percent change in group mean levels, but describes the regression parameter as a mean difference in final values. The full analysis model and the connection between those estimands are not reported. The record does not state whether baseline values or other covariates were included in regression, and no diagnostic information is available for checking model assumptions.

The official title calls the study partially blind, while the masking information lists quadruple masking. The record does not explain this difference or identify which people were masked. Details of the randomization procedure and allocation concealment are also not reported in the registry record.

The primary analysis excludes participants without relevant post-randomization efficacy data, but further missing-data methods and reasons for missing measurements are not reported. The record does not provide a sample-size justification, an interim analysis plan, or enough information to assess prespecification. It also does not report whether multiplicity adjustments were used across the three primary comparisons and the secondary outcomes.

For safety, the listed at-risk count in the 800 mg group differs between the serious-adverse-event summary and the primary outcome summary; the reason is not reported. Death counts and individual adverse-event terms are not reported in the registry record. Together, these omissions constrain interpretation beyond the uncertainties already reflected in the estimates.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Study to Evaluate the Effect of GWP42003 on Liver Fat Levels in Participants With Fatty Liver Disease — NCT01284634.
• Registry record last updated 2018-08-08; status last verified 2018-07.
• Results first posted 2014-01-20.
• 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.

20. Record Summary

NCT01284634 is a completed Phase 2 randomized, parallel, placebo-controlled pilot dose-ranging study sponsored by GW Research Ltd. It enrolled 25 participants with fatty liver disease and evaluated GWP42003 at 200 mg/day, 400 mg/day, and 800 mg/day against placebo over 8 weeks.

The primary outcome was percent change from baseline to EOT (Day 57) or early termination in mean liver triglyceride levels, measured using magnetic resonance imaging or magnetic resonance scanning. The registry reports three linear-regression comparisons against placebo, each with a 90% confidence interval spanning zero. The outcome groups were small, and the estimates are imprecise.

The record reports five secondary outcomes and serious adverse event counts of zero affected participants in each listed group. Important details about model specification, missing-data methods, multiplicity, masking procedures, and the trial's own sample-size justification are not reported in the registry record.