Augmenting Growth Hormone to Ameliorate Nonalcoholic Fatty Liver Disease in Adolescents: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Augmenting Growth Hormone to Ameliorate Nonalcoholic Fatty Liver Disease in Adolescents (NCT02726542), a randomized early/phase 1 study. The primary outcome is Change in Hepatic Fat Fraction.

2. Clinical Question

3. Trial Design

This was a randomized, unmasked, parallel-group, early phase 1 study sponsored by Massachusetts General Hospital. The registry lists enrollment of 24 and a completed status. The listed study start was 2017-05-03, and primary completion was 2020-04-02. These record details describe the trial’s registration and conduct status; they do not establish how many participants contributed data to every endpoint.

The intervention arm was titled Growth Hormone and used somatropin by daily subcutaneous injection. The dose was to begin at 1mg, with titration based on IGF-1 levels. The comparator was titled No Treatment and involved observation only. Because masking was none, the registry indicates that the trial was open-label; it does not describe masking of outcome assessors.

The primary outcome was change in hepatic fat fraction at 24 weeks. Its results use available data, with 9 participants reported in each arm. The registry’s secondary results include AST, ALT and GGT changes, each at 24 weeks, with 10 participants reported in each arm. Although change in visceral adipose tissue is listed among secondary outcomes, its result is not reported in the registry record.

The registry record does not report the sequence-generation method, allocation concealment, a sample-size justification, or detailed eligibility criteria. It also does not report a full statistical analysis plan, interim analyses, or a description of how missing observations were handled. Those details cannot be inferred from the randomized designation or from the posted group summaries.

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

Analysis population for the primary outcome, as defined in the registry: all available data used

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 Change in Hepatic Fat Fraction, measured over 24 weeks.

The primary endpoint was change in hepatic fat fraction from baseline to 24 weeks, measured by hydrogen magnetic resonance spectroscopy. Hepatic fat fraction expresses the amount of fat in the liver as a fraction, and the registry gives its unit as percent. The endpoint is a continuous measure: the reported summaries are means and standard deviations rather than counts of participants meeting a threshold.

Change from baseline focuses the outcome on how the measured value shifted during the study period, rather than only on the value at the end. Comparing changes between randomized groups can describe whether their average trajectories differed. General: a direct between-group estimate is more informative for that question than deciding whether each group’s mean change, separately, differs from zero. The registry posts the two group means and standard deviations, but it does not report a primary between-group estimate, confidence interval, or p-value.

The registry says all available data were used for the primary endpoint. It does not explain whether all participants had measurements at both time points, how missing values were addressed, or whether the endpoint definition had a prespecified hierarchy. Those omissions limit assessment of how the reported group summaries relate to all enrolled participants.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous and summarized by a mean and standard deviation in each group. A mean describes the group’s average change; a standard deviation describes how spread out participant-level changes are around that average. They answer different questions: a similar mean can coexist with different degrees of individual variation. For the primary endpoint, the posted means are -3.0 and 0.3, with standard deviations of 5.2 and 3.5, respectively. The registry gives 9 participants per group for this result.

Why a between-group estimate matters

General: the question in a randomized comparison is usually about the difference between groups, not whether either group changed in isolation. A between-group estimate combines the observed group summaries into a contrast. A confidence interval around that contrast describes the range of values reasonably compatible with the data under the analysis assumptions. It indicates precision as well as direction; a narrow interval is more informative about the size of a contrast than a wide interval.

The registry does not report a primary contrast or its confidence interval. Although the group means can be read descriptively, a new primary estimate should not be represented as an analysis performed by the trial sponsor. The registry’s missing inferential result also means a primary p-value is not available. No inferential conclusion about the primary comparison should be drawn from the separate means alone.

Recalculated secondary comparisons

For AST, ALT and GGT, the registry posts means, standard deviations and group counts, alongside recalculated simple unadjusted comparisons. These recalculated estimates compare the no-treatment group with the growth hormone group. They are not sponsor-reported analyses: the registry lists no reported analyses for these endpoints. The recalculated mean differences and intervals are descriptive inferential summaries based on posted statistics, without adjustment for other outcomes or additional covariates.

General: a two-sided confidence interval that includes zero is compatible with both a zero difference and differences in either direction at the interval’s stated confidence level. It does not show that the groups are equivalent, nor that there is no effect. Equivalence requires a prespecified margin and an analysis designed to assess whether the interval lies within that margin; neither is reported in the registry record. A p-value measures how unusual data at least as incompatible with a null hypothesis would be under assumptions of that test. It is not the probability that the null hypothesis is true, and it does not measure clinical importance.

The secondary estimates have unadjusted p-values of 0.521 for AST, 0.431 for ALT and 0.158 for GGT. These values are reported as recalculated unadjusted results, not as the trial’s planned or sponsor-reported tests. Because several outcomes are listed, interpretation of individual unadjusted p-values should be cautious. The record does not report a multiplicity strategy, so it is not possible to determine whether the analyses were intended as confirmatory or exploratory.

Variation, uncertainty and small samples

General: standard errors and confidence intervals reflect both variability and the amount of information in the data. With few observations and substantial dispersion, estimates can remain uncertain. The secondary results report 10 participants per group and show different levels of spread across measures. In particular, the posted GGT standard deviation is 30.0 in the growth hormone group and 9.0 in the no-treatment group. A difference in dispersion can affect the uncertainty of a comparison; a pooled-variance approach relies on assumptions about how the groups’ variability relates. The recalculated summaries use the method represented in the posted results, but the registry record does not specify a full statistical analysis plan.

The primary sample sizes are smaller than the secondary endpoint sample sizes listed in the results. That difference should not be explained by assuming a particular missing-data process: the registry does not say why the counts differ. It states that all available data were used for the primary outcome but does not describe imputation, exclusions, or a complete-case rule. General: missing-data methods matter because an analysis of available observations may not represent all randomized participants if the presence of a measurement is related to the outcome or group.

Planning illustrations are not observed power

The registry includes planning illustrations based on observed effects, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. For each secondary outcome, the illustrations consider planning for 100%, 75% and 50% of the observed effect and give participants per arm for 80% and 90% power. These values are hypothetical planning quantities, not actual enrollment targets, and not a measure of the trial’s observed power. A planning calculation based on an observed effect is especially uncertain when that estimate comes from a small study. The registry record does not report the trial’s own sample-size justification, so the illustrations should not be mistaken for the design rationale.

General: over-dispersion means observed variation is greater than a model expects. When a method assumes less variation than is present, it can understate uncertainty. The registry’s posted continuous summaries report standard deviations, but it does not report a model-based assessment of over-dispersion or a robust-variance method. More broadly, the available record does not specify whether the analyses accounted for baseline values, repeated measurement, or other covariates. No such adjustment should be presumed.

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("Growth Hormone", "No Treatment")
n   <- c(9, 9)
m   <- c(-3, 0.3)
sp  <- c(5.2, 3.5)
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  Growth Hormone  9  -3.000  5.200  -6.997  0.997
CARM  No Treatment  9  0.300  3.500  -2.390  2.990

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome and several secondary outcomes, but a multiplicity strategy is not reported in the registry record. It is therefore unclear whether the secondary outcomes were intended to be tested as a family, whether any order of testing was prespecified, or whether p-values were to be adjusted. The recalculated AST, ALT and GGT p-values are explicitly unadjusted. They should not be interpreted as if they had been protected against multiple comparisons.

The registry’s primary endpoint description says all available data were used. That states how the posted result was characterized, but it does not explain the analysis population in enough detail to determine whether all randomized participants were included, whether only those with observed paired measurements contributed, or whether any missing values were imputed. Missing-data methods and reasons for missingness are not reported in the registry record. The primary group counts are 9 in each arm; the secondary group counts are 10 in each arm. No explanation for this difference is given.

Randomized allocation helps define the intended comparison, but the registry does not identify an intention-to-treat analysis or another named analysis population. It does not report participant flow by analysis, protocol deviations, or exclusions from specific endpoint analyses. Those details are important for judging whether the analysis preserved the randomized groups and whether available-case summaries could be affected by selective outcome availability.

Interim analyses, stopping rules, and any adjustment associated with interim monitoring are not reported in the registry record. A full statistical analysis plan is also not reported. Consequently, the record cannot establish whether the reported endpoints, contrasts, missing-data procedures or multiplicity decisions were prespecified. These are limits on interpretation of the posted results, not evidence that a particular procedure was or was not used.

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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hospitalization (7.7%).

The safety record lists 13 participants at risk in the growth hormone group and 11 in the no-treatment group for serious adverse events. It reports 1 participant affected in the growth hormone group and 0 in the no-treatment group. Deaths are reported as 0 in both groups. The listed top serious-event term is Hospitalization, at 7.7%.

For other adverse events, the record lists 13 participants at risk and 9 affected in the growth hormone group, and 11 at risk and 1 affected in the no-treatment group. These counts are distinct from the serious-event counts and should not be combined with them. The registry record does not provide further event-level detail in the posted safety summary, such as the nature, timing, duration or relationship of each event to study treatment.

These are descriptive counts from a small study, and the groups have different numbers at risk. They do not by themselves estimate comparative safety precisely, establish causation, or rule out uncommon events. The 7.7% figure is the listed proportion for the top serious-event term; the record does not supply enough detail to infer broader safety conclusions from that term alone.

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

The posted primary results show a mean decrease in hepatic fat fraction in the growth hormone group and a mean increase in the observation group over 24 weeks. The group means were -3.0 and 0.3, respectively. These values describe observed changes among the participants contributing data; the registry does not report a primary between-group estimate or an interval that quantifies its uncertainty.

That distinction matters. Descriptive means can suggest the direction of the observed group summaries, but without a reported contrast and its precision, the size and uncertainty of the difference cannot be evaluated directly from the record. The standard deviations also indicate variation among participants, so the group averages do not describe every participant’s change. Neither the primary summary nor the recalculated secondary comparisons establish that one group’s outcome would be expected for every person.

The secondary results are unadjusted recalculations and are not sponsor-reported analyses. Their confidence intervals include zero, and they do not resolve the uncertainty around the primary result. The registry record also lacks a reported multiplicity approach and a detailed account of missing data and analysis populations. Those gaps constrain how strongly the observed summaries can be interpreted.

These results do not establish equivalence between the groups, because an equivalence margin and an equivalence analysis are not reported. They also do not prove absence of a difference: a non-significant unadjusted secondary result is not evidence that the true difference is exactly zero. Conversely, the direction of the primary group means alone is not evidence of a definitive treatment effect.

The study is recorded as completed, with enrollment of 24, but the primary analysis reports 9 participants per group. The registry does not explain the difference between overall enrollment and endpoint counts. Readers should therefore interpret the posted outcome summaries as limited evidence from the recorded analyses, rather than as a complete account of outcomes for all enrolled participants.

14. Why This Trial Matters Statistically

Statistically, the record illustrates why a randomized design and a reported group contrast are not the same thing. Randomization describes how assignment was intended to occur. To assess the primary question, readers also need a between-group estimate and an uncertainty interval. Here, the registry reports group means and standard deviations for hepatic fat fraction but does not report that primary inferential comparison. The apparent separation between the group means is therefore not accompanied by a registry-reported measure of its precision.

The record also shows how analysis counts can differ across outcomes. The primary endpoint lists 9 participants per group, while the secondary outcomes list 10 per group. Without participant flow and missing-data detail, those differences cannot be attributed to any particular reason. This matters because analysis populations and outcome availability determine which observations inform each summary.

Another statistical lesson concerns secondary outcomes. Several unadjusted comparisons may offer useful exploratory context, but they do not carry the same interpretive weight as a well-reported prespecified primary contrast. Their confidence intervals are wide and include zero, and the record does not describe a multiplicity strategy. The secondary outcomes thus cannot compensate for missing precision on the primary question.

Finally, the planning illustrations use observed effects as inputs and are explicitly hypothetical. They help show how sample-size requirements can change when the effect targeted for detection is smaller than the observed estimate. They are not the trial’s actual design justification and do not describe observed power. The registry does not report a sample-size rationale for the trial itself. Separating these planning illustrations from the study’s actual design is essential when interpreting a small study’s results.

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 limitations are specific to what the registry reports for this trial. The primary endpoint has group-level summaries but no reported between-group estimate, confidence interval or p-value. That prevents direct assessment of the precision of the primary contrast. The primary outcome counts are 9 per group despite total enrollment of 24, and the record does not explain the participant flow or why fewer participants appear in these summaries.

The secondary outcomes have 10 participants per group, but their analyses are recalculated, unadjusted comparisons rather than reported sponsor analyses. The registry does not report multiplicity control, a hierarchy for interpreting outcomes, or a full statistical analysis plan. Change in visceral adipose tissue is listed as a secondary outcome, but its result is not reported in the registry record.

The registry describes an open-label comparison of daily somatropin with observation only. It does not report whether outcome assessment was masked, allocation concealment, detailed inclusion and exclusion criteria, or the treatment exposure actually received. The dose description says treatment would begin at 1mg and be titrated based on IGF-1 levels; the results record does not give achieved dosing or adherence information.

Further, missing-data procedures, analysis populations, participant flow, interim analyses and the trial’s own sample-size justification are not reported in the registry record. These omissions do not establish that the corresponding procedures were absent, but they limit evaluation of potential bias, precision and prespecification. The listed safety counts are also too limited to support detailed comparisons of event type, timing or treatment relationship from this record alone.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Augmenting Growth Hormone to Ameliorate Nonalcoholic Fatty Liver Disease in Adolescents — NCT02726542.
• Registry record last updated 2022-09-07; status last verified 2022-08.
• Results first posted 2021-04-26.
• 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

This completed, randomized, unmasked, parallel early phase 1 study compared somatropin with observation in a study concerning NAFLD and increased liver fat in young people. Enrollment was 24. The primary outcome was change in hepatic fat fraction at 24 weeks, measured by hydrogen magnetic resonance spectroscopy. The registry reports mean changes of -3.0 with a standard deviation of 5.2 for growth hormone and 0.3 with a standard deviation of 3.5 for no treatment, with 9 participants per group.

No primary between-group inferential result is reported. Recalculated unadjusted comparisons are available for AST, ALT and GGT; no result for change in visceral adipose tissue is reported in the registry record. The registry does not report the full analysis plan, missing-data methods, multiplicity strategy or the trial’s own sample-size justification. The posted results therefore support descriptive reading, with important limits on inference and precision.