Safety and Efficacy of Roflumilast and Pioglitazone in Treating Adults With Nonalcoholic SteatoHepatitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy of Roflumilast and Pioglitazone in Treating Adults With Nonalcoholic SteatoHepatitis (NCT01703260), a randomized phase 2 study. The primary outcome is Amount of Serum Alanine Transaminase (ALT) at Baseline.

2. Clinical Question

The trial was described as randomized, controlled, double-blind, multicenter, and phase 2. Its official title identifies liver enzymes and liver fat content as areas of interest. The record lists baseline ALT as the primary outcome measure and includes secondary outcomes involving AST and liver fat content. These measures indicate the types of questions the study set out to investigate, but the available posted results do not answer whether the groups changed differently over time.

The distinction between a baseline measurement and a treatment outcome is central to interpreting the clinical question. Baseline ALT describes the measured starting value among participants included in the analysis. A comparison of treatment effects would ordinarily require follow-up values or changes from baseline for the treatment groups, along with an analysis comparing those groups. Such results are not reported in the registry record. Accordingly, the baseline values can describe the participants' starting ALT measurements, but cannot establish a treatment effect.

3. Trial Design

This was a randomized, parallel-group phase 2 study described in its official title as double-blind, controlled, and multicenter. The registry record reports quadruple masking. The details of who was masked are not reported in the registry record. It lists three arms: roflumilast plus pioglitazone, roflumilast only, and pioglitazone only. In the single-drug arms, the other study drug was paired with a matching placebo. The treatments were oral and intended for up to 4 months.

The registry enrollment is 20. For the primary baseline ALT measure, the reported arm sizes are 7.0 in the roflumilast-plus-pioglitazone group, 7.0 in the roflumilast-only group, and 6.0 in the pioglitazone-only group. The population definition for this measure is the safety analysis set: participants who were randomized, received at least 1 dose of double-blind study medication, and had a baseline assessment available. The record does not provide a participant flow account that explains enrollment, exclusions, or completion by arm.

The study started in 2013-06 and had primary completion in 2014-09. Its status is terminated. The reason for termination, the timing of that decision, and any consequences for follow-up are not reported in the registry record. The record does not state whether the trial stopped according to a prespecified operational plan or for another reason.

Three active-treatment arms allow the combination to be considered alongside each of its component drugs. However, the posted primary measure is baseline ALT rather than a follow-up or change measure, and the record lists no analyses. As a result, the design description identifies the intended treatment groups, but the posted results do not provide estimates comparing their effects. Further details on allocation procedures, treatment adherence, study visits, and the statistical analysis plan 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.

Before assignment: Participants with a historical diagnosis of nonalcoholic steatohepatitis (NASH) and nonalcoholic fatty liver disease (NAFLD) activity score (NAS) of greater than or equal to (\>=) 3 were enrolled in 1 of 3 treatment groups as follows: roflumilast + pioglitazone; roflumilast only; pioglitazone only.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety analysis set included all participants who were randomized, received at least 1 dose of double-blind study medication and had baseline assessment available.

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: The randomized set included all participants who were randomized.

6. Primary Endpoint

The registry lists the primary outcome as Amount of Serum Alanine Transaminase (ALT) at Baseline, measured over Baseline.

The primary outcome measure is the amount of serum alanine transaminase (ALT) at baseline, expressed in international units per liter (IU/L). ALT is a serum transaminase, as identified in the study description. The registry defines the reported statistic as a mean with standard deviation and specifies baseline as the timeframe. The analysis population is the safety analysis set, restricted to randomized participants who received at least 1 dose of double-blind study medication and had a baseline assessment available.

Baseline measures are useful for describing the participants' starting status and the distribution of a continuous measurement within each treatment group. They can help readers understand the values from which later measurements would be assessed. As a general statistical point, however, a baseline value alone does not measure change during treatment and cannot estimate a difference in treatment effects. For that purpose, follow-up values or changes from baseline and an appropriate between-group analysis would be needed.

The study purpose refers to evaluating the effect of therapy on ALT, but the posted primary outcome is explicitly a baseline amount. The record does not report a follow-up ALT result, a change from baseline, a treatment contrast, a confidence interval, or a p-value for ALT. The reported means and standard deviations therefore describe baseline ALT only. They should not be interpreted as evidence of improvement, worsening, or a difference caused by treatment.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The outcome is continuous: ALT is measured on a numeric scale in international units per liter (IU/L). For this measure, the registry reports a mean and standard deviation in each treatment group. A mean describes the center of the observed group values, while the standard deviation describes their spread. These are descriptive statistics. They do not by themselves test a hypothesis or quantify a treatment contrast.

What does a mean describe?

For a continuous measurement, the mean is the sum of the observed values divided by the number of observations. The registry lists means of 101.7 IU/L, 83.4 IU/L, and 118.8 IU/L for the three arms. The mean is a compact description of the group's average, but it can be influenced by unusually high or low observations. With small groups, individual observations can have a noticeable influence on the summary. The registry record does not provide individual participant values, so the shape of each group's measurement distribution cannot be assessed.

What does the standard deviation describe?

The standard deviation summarizes the typical spread of observations around the mean. The posted values are 56.98, 31.63, and 36.76 IU/L for the combination, roflumilast-only, and pioglitazone-only arms, respectively. As a general statistical concept, the standard deviation is a measure of variation among observations, not a measure of how precisely a treatment effect has been estimated. Precision for a group mean or for a between-group contrast is usually described with a confidence interval or another uncertainty measure. Such intervals are not reported for this primary baseline outcome.

Why is baseline not a treatment comparison?

The primary measure is ALT at baseline. A baseline summary describes values before the period of treatment being evaluated; it does not quantify the subsequent response. A treatment-effect analysis would need a defined post-baseline endpoint, such as a follow-up value or a change from baseline, and an analysis that compares treatment groups. The record does list secondary outcomes including percent change in AST at Month 4 and change in liver fat content at Month 4, but their results and analyses are not reported in the registry record. No follow-up ALT result or treatment contrast is posted.

Randomization is intended, as a general design principle, to support fair comparisons by assigning participants to groups through a chance-based process. It does not remove uncertainty, particularly when groups are small, and it does not substitute for reporting the outcome comparison. The registry describes the study as randomized and parallel-group, but does not report the allocation method or a statistical analysis of treatment outcomes. Thus, randomization is part of the design description, not a result demonstrating comparability or treatment effect.

What do confidence intervals and p-values add?

As a general statistical explanation, a confidence interval gives a range of values compatible with an estimated quantity under the analysis assumptions. For a treatment contrast, it conveys the direction and precision of the estimated difference. A p-value assesses how compatible the observed data are with a specified null hypothesis under an assumed statistical model; it is not the probability that a treatment works, nor a measure of clinical importance. Neither confidence intervals nor p-values for a treatment comparison are reported in the registry record. Because the primary outcome is a baseline measure, an inferential comparison of baseline values would not answer the stated treatment-effect question in any case.

Why can over-dispersion matter?

Over-dispersion is a general issue for some statistical models, particularly when observed variability exceeds what the model assumes. It can make standard errors too small and confidence intervals too narrow if ignored. The posted primary outcome uses a continuous mean-and-standard-deviation summary, and no fitted model or count outcome analysis is reported. The registry record does not report an over-dispersion assessment or a model specification. It would therefore be inappropriate to claim that over-dispersion was present, absent, or addressed in this study.

The registry record lists 0 analyses. This identifies the absence of posted analysis entries, rather than a detailed account of what investigators may have calculated. The statistical analysis plan, the model used for any treatment comparison, handling of repeated measurements, and adjustment for baseline values are not reported in the registry record. No new effect estimates can be derived from the arm-level means and standard deviations without making unsupported assumptions about the analysis and the endpoint.

Sample-size calculations are design tools, not findings from participants. General planning illustrations based on an observed effect would depend on an effect estimate and assumptions about significance level, allocation, and the statistical model. The registry record does not report its own sample-size justification. Any planning illustration should not be described as observed power, and the baseline means are not a treatment effect on which to base an efficacy interpretation.

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("Roflumilast + Pioglitazone", "Roflumilast Only", "Pioglitazone Only")
n   <- c(7, 7, 6)
m   <- c(101.7, 83.4, 118.8)
sp  <- c(56.98, 31.63, 36.76)
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  Roflumilast + Pioglitazone  7  101.700  56.980  49.002  154.398
CARM  Roflumilast Only  7  83.400  31.630  54.147  112.653
CARM  Pioglitazone Only  6  118.800  36.760  80.223  157.377

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 6 outcome measures, including the primary baseline ALT measure and secondary measures concerning AST and liver fat content. Four secondary outcomes are identified: baseline AST, percent change from baseline in AST at Month 4, baseline liver fat content, and change from baseline in liver fat content at Month 4. Results for these secondary outcomes are not reported in the registry record. It lists 0 analyses, so there are no posted analysis results from which to assess the pattern of findings across outcomes.

Multiplicity is the general statistical problem that arises when several outcomes, time points, or comparisons are tested. The more tests considered, the greater the chance of obtaining at least one apparently notable result by chance if no adjustment or prespecified interpretation is used. A statistical plan may describe how multiple tests are handled, but such a plan is not reported in the registry record. It is therefore not possible to determine whether any multiplicity adjustment was planned or applied.

Missing-data handling is also not reported in the registry record. The primary outcome population is defined as randomized participants who received at least 1 dose of double-blind study medication and had baseline assessment available. This defines who contributed to the posted baseline measure, but does not explain missing follow-up measurements, withdrawals, or how incomplete outcome data were handled. A participant flow account and missing-data methods are not reported.

The registry record does not report whether analyses followed an intention-to-treat approach, how protocol deviations were handled, or whether the analysis populations differed across endpoints. These omissions limit assessment of potential bias and of how the available participants represent the randomized groups. The record's terminated status does not explain the missing analyses or outcomes; the reason for termination is not reported in the registry record.

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 0% reporting threshold.

Most frequent non-serious adverse event terms

For serious adverse events, the posted affected counts are 0 of 7 participants in the roflumilast-plus-pioglitazone group, 0 of 7 in the roflumilast-only group, and 0 of 6 in the pioglitazone-only group. These figures mean that no serious adverse events are recorded among the listed participants at risk in these groups. They should be read as counts within a small study, not as proof that serious events cannot occur or as a precise estimate of comparative safety.

The record also lists other adverse events affecting 5 of 7 participants in the combination group, 4 of 7 in the roflumilast-only group, and 4 of 6 in the pioglitazone-only group. These are separate from the serious adverse event counts. The registry record does not provide the terms for serious adverse events, and the serious-event term list is empty. Death counts are not reported in the registry record.

As a general statistical point, event counts in small groups are imprecise: a count of zero in a group does not establish zero underlying risk. The registry record does not report exposure-adjusted event rates, uncertainty intervals, or a comparative safety analysis. These figures describe the posted counts only and do not support a conclusion that one regimen is safer than another.

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

The posted results establish that the registry record contains baseline ALT summaries for three treatment groups. The means are 101.7 IU/L for roflumilast plus pioglitazone, 83.4 IU/L for roflumilast only, and 118.8 IU/L for pioglitazone only. The corresponding standard deviations are 56.98, 31.63, and 36.76. The analysis group sizes are 7.0, 7.0, and 6.0. These quantities characterize starting measurements and observed variation among participants in the safety analysis set.

They do not show how ALT changed after participants began treatment. They do not estimate the effect of combining roflumilast with pioglitazone, compare either single-drug group with the combination, or establish a difference in liver fat content. The record lists secondary outcomes that could address AST or liver fat change, but their results are not reported in the registry record. No treatment-effect estimates, confidence intervals, or p-values are posted.

As a general statistical principle, a study's design and its results answer different questions. Randomization and masking describe how a trial was organized; they do not constitute evidence for an outcome difference. Likewise, a baseline mean is descriptive and should not be treated as a follow-up result. The registry record identifies the study as terminated, but does not report why it terminated or how that status affected the planned analyses. Those unknowns cannot be resolved from the posted baseline summaries.

The record therefore supports a narrow conclusion: it reports starting ALT values and safety counts for the listed groups. It does not support a conclusion about whether either treatment changed ALT, AST, or liver fat, or whether the combination differed from either single-drug regimen. The absence of posted efficacy analyses is not evidence of no effect, just as the study's stated purpose is not evidence of an effect. Interpretation should remain limited to the measurements and counts that are actually reported.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of the difference between a study question and the results available for evaluating it. The official title and brief description concern the effect of treatment on liver enzymes and liver fat, but the posted primary outcome is an ALT measurement at baseline. Baseline summaries can describe the starting measurements; they cannot answer whether treatment groups experienced different changes after treatment.

The arm-level summaries also show why group size and uncertainty matter. The reported groups are small, with 7.0, 7.0, and 6.0 participants contributing to the primary measure. Means and standard deviations convey central tendency and within-group spread, but without inferential comparisons they do not quantify the uncertainty around a treatment effect. A randomized design is not a substitute for outcome estimates, confidence intervals, or a clearly reported analysis.

The record lists secondary outcome measures relating to AST and liver fat, yet does not report their results. It also lists 0 analyses. This makes it impossible to assess the intended outcome pattern, the precision of estimates, or whether findings were consistent across measures. The terminated status adds context, but the reason for termination is not reported in the registry record and should not be inferred.

These features matter to biostatisticians and clinical researchers because transparent endpoint definitions and analysis reporting are needed to connect trial design with evidence. A baseline measure can be a useful descriptive component, but it should not be mistaken for an efficacy result. In this record, the most defensible statistical reading is limited: baseline ALT and safety counts are reported, while treatment-effect analyses are not.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The principal limitation for interpretation is that the primary outcome posted is baseline ALT rather than a post-treatment value or change from baseline. No treatment contrast for ALT is reported. The record lists secondary measures that include changes in AST and liver fat, but does not report their results or analyses. Consequently, the clinical treatment question cannot be answered using the posted outcome data.

The arm sizes contributing to the primary measure are 7.0, 7.0, and 6.0. With groups of this size, descriptive summaries alone provide limited information about the precision of any comparison. The registry record does not report confidence intervals, p-values, a sample-size justification, or a statistical model for treatment comparisons. It is not possible to assess whether the study was designed to estimate a particular effect with a stated level of precision.

The primary outcome population is restricted to randomized participants who received at least 1 dose of study medication and had a baseline assessment available. A participant flow account, details of missing follow-up measurements, and methods for handling missing data are not reported in the registry record. The record also does not describe allocation procedures, adherence, protocol deviations, or the roles covered by quadruple masking.

The study is listed as terminated, but the reason and timing of termination are not reported in the registry record. This status cannot be used to infer why treatment outcomes are absent. The lack of posted results leaves open whether analyses were completed outside the record, but does not justify assumptions about their direction or findings. These limitations apply to what can be concluded from the registry record, not necessarily to information that may exist elsewhere.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Safety and Efficacy of Roflumilast and Pioglitazone in Treating Adults With Nonalcoholic SteatoHepatitis — NCT01703260.
• Registry record last updated 2017-02-01; status last verified 2016-09.
• Results first posted 2016-02-23.
• 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

NCT01703260 is a randomized, parallel-group phase 2 study of roflumilast and pioglitazone in adults with nonalcoholic steatohepatitis. The official title describes a double-blind, controlled, multicenter study; the registry lists quadruple masking. The three arms are the combination, roflumilast only, and pioglitazone only, with oral treatment for up to 4 months. Enrollment is 20, the study started in 2013-06, primary completion was in 2014-09, and status is terminated.

The posted primary measure is mean serum ALT at baseline, with standard deviations, in a safety analysis set. The record also lists secondary AST and liver-fat outcomes, but no analyses are posted. Baseline summaries and adverse event counts describe the reported data; they do not estimate treatment effects. The reason for termination, statistical analysis plan, sample-size justification, and missing-data methods are not reported in the registry record.