A Dose-ranging Study of DRM01 in Subjects With Acne Vulgaris: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Dose-ranging Study of DRM01 in Subjects With Acne Vulgaris (NCT02431052), a randomized phase 2 study. The primary outcome is Mean Absolute Change in Acne Lesion Counts (Inflammatory) From Baseline to Week 12.

2. Clinical Question

The trial was designed as a dose-ranging comparison rather than a single active-versus-control contrast. It included five groups: two vehicle regimens and three active-dose regimens. The presence of both once-daily and twice-daily regimens permits comparisons across application schedules, while the active-dose groups include two concentrations.

The posted arm means describe changes from baseline. Negative values indicate that the Week 12 counts were lower than baseline on average, according to the direction of the change measure. The key statistical question is the size and precision of differences between each active regimen and the once-daily vehicle group identified as the control in the posted comparisons. These pairwise calculations are unadjusted re-calculations; the registry also reports a separate overall analysis across five groups.

The registry record does not report a separate primary question about safety, nor does it report posted results for secondary outcome measures. The record’s brief summary states that the study objectives included assessment of safety and efficacy compared with vehicle, but the posted numerical efficacy results concern inflammatory lesion-count change.

3. Trial Design

This was a Phase 2, randomized, triple-masked, parallel-group, vehicle-controlled dose-ranging study of topical gel in subjects with acne vulgaris. The registry record lists enrollment of 420 and status as completed. The study start is recorded as 2015-04, with primary completion in 2016-04.

Five regimens are represented in the posted arm results. The once-daily vehicle group included 52 participants, and the twice-daily vehicle group included 50. The active groups were 4.0% once daily, with 106 participants; 7.5% once daily, with 110; and 7.5% twice daily, with 101. Each regimen was applied to the face for 12 weeks. The registry describes the overall study as randomized and parallel, but further details of sequence generation, allocation concealment, or masking procedures are not reported in the registry record.

The primary outcome was mean absolute change in inflammatory acne lesion counts from baseline to Week 12, measured in lesions. Its population is listed as Intent-to-Treat, and the reported arm statistic is a least-squares mean. The arm-level dispersion is identified as standard deviation. The registry’s reported analysis lists a Cochran-Mantel-Haenszel method and an overall p-value across the five groups, but does not give a parameter estimate or confidence interval for that analysis.

The registry record does not report the study’s original sample-size justification, allocation ratios as planned, or detailed statistical analysis plan. It also does not report whether interim analyses were planned or conducted. The posted arm counts and results describe analyzed outcome data and should not be treated as a substitute for those design details.

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: Intent-to-Treat

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: Intent-to-Treat

6. Primary Endpoint

The registry lists the primary outcome as Mean Absolute Change in Acne Lesion Counts (Inflammatory) From Baseline to Week 12, measured over Baseline and Week 12.

The primary endpoint is mean absolute change in inflammatory acne lesion counts from baseline to Week 12. A lesion count is a quantitative measure, so the endpoint is treated as continuous and summarized as a mean. The registry identifies the reported mean as a least-squares mean and gives lesions as the unit.

Change scores express how a participant’s measurement at follow-up differs from their baseline measurement. In this record, negative arm means indicate an average decrease in counted inflammatory lesions by Week 12. The endpoint therefore describes the direction and average magnitude of change within each study group. Comparisons between group means address whether the observed average change differed between regimens.

Inflammatory lesion counts provide a measurable outcome for a dose-ranging study because they can be recorded at baseline and again at a specified follow-up time. General statistical context: a group mean condenses participant-level measurements into one summary, but does not show the full distribution of individual responses. The standard deviation describes variation among observations; it is not the same as the uncertainty around the mean or around a difference between means.

The endpoint is limited to inflammatory lesion counts at the specified time point. It does not, by itself, describe other acne features, longer-term outcomes, participant experience, or safety. Those outcomes cannot be inferred from this measure. The registry record does not report posted secondary outcome results in the results.

7. Results Posted in the Registry

• Olumacostat Glasaretil Gel, Vehicle BID: mean -11.00 versus -10.50 on control; difference -0.50 (95% CI -4.15 to 3.15), standardized effect d = -0.05; z-test p = 0.789.
• Olumacostat Glasaretil Gel, 4.0% QD: mean -14.60 versus -10.50 on control; difference -4.10 (95% CI -7.29 to -0.91), standardized effect d = -0.44; z-test p = 0.012.
• Olumacostat Glasaretil Gel, 7.5% QD: mean -14.50 versus -10.50 on control; difference -4.00 (95% CI -7.20 to -0.80), standardized effect d = -0.42; z-test p = 0.014.
• Olumacostat Glasaretil Gel, 7.5% BID: mean -15.00 versus -10.50 on control; difference -4.50 (95% CI -7.73 to -1.27), standardized effect d = -0.47; z-test p = 0.006.

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is continuous, so the posted summaries are means and standard deviations. A mean describes the center of a group’s observed changes. A standard deviation describes how dispersed the individual changes are around that group mean. General statistical context: the standard deviation is about participant-to-participant variability, whereas a confidence interval for a mean or mean difference describes uncertainty in an estimated quantity.

What does a mean difference describe?

Each recalculated comparison subtracts the once-daily vehicle mean from the mean in another arm. A negative difference means that the comparison arm’s reported mean change was more negative than the reference arm’s mean change. Since the endpoint is change in lesion counts, this describes a greater average decrease relative to the reference, not a claim about every participant. The posted differences are -0.5 lesions for twice-daily vehicle, -4.1 for 4.0% once daily, -4.0 for 7.5% once daily, and -4.5 for 7.5% twice daily.

How should the re-calculated tests be read?

The listed pairwise p-values are unadjusted and were recalculated from posted summary statistics. A p-value measures how compatible the observed data are with a specified null model, under the assumptions of the test; it is not the probability that the null hypothesis is true. The recalculated values are 0.789, 0.012, 0.014, and 0.006 for the comparisons in the order. These are not the same as the registry’s overall p-value, because the registry reports a Cochran-Mantel-Haenszel analysis spanning all five groups rather than these particular pairwise estimates.

The registry record does not report the model specification, adjustment covariates, or exact contrast procedure used for the least-squares means. It does not explain how the Cochran-Mantel-Haenszel test was applied to the continuous primary endpoint. A detailed statistical plan is not reported in the registry record. Therefore, the re-calculated comparisons should be regarded as a transparent but simplified analysis rather than a reconstruction of the sponsor’s complete model.

Why is the pooled standard deviation approximate?

The pairwise re-calculations use an approximate pooled standard deviation formed from posted group-level dispersion summaries. Pooling assumes that a common variability estimate is a reasonable summary for the groups being compared. General statistical context: if group variability differs materially or the analysis model handles covariates or repeated measurements differently, a simple pooled calculation may not reproduce the model-based result. The registry gives standard deviations for the arms, but does not provide enough model detail to determine how these relate to the sponsor’s analysis.

What does the overall analysis contribute?

The registry lists a Cochran-Mantel-Haenszel method, identifies all five groups, and reports p < 0.001. It does not post the method’s parameter, value, confidence limits, or confidence level. An overall test addresses a different question from any one pairwise contrast: broadly, whether there is evidence of a difference across the collection of groups, rather than the size of a particular arm-to-control difference. The precise interpretation depends on the underlying test specification, which is not reported.

Because the registry’s overall result and the simple pairwise re-calculations differ, the sponsor’s own analysis takes precedence. The pairwise statistics remain useful for illustrating estimated differences and uncertainty based on the posted arm summaries, but they should not be presented as the trial’s official analysis.

What do confidence intervals add?

A confidence interval gives a range of values compatible with the data and the method used to calculate it. General statistical context: a narrower interval indicates greater precision, while a wider one indicates more uncertainty. If an interval for a mean difference includes zero, values corresponding to no average difference remain compatible with that interval. If it excludes zero, the interval is inconsistent with a zero difference under the same method and confidence level. Neither feature alone establishes clinical importance.

The pairwise intervals are based on approximate standard deviations and unadjusted calculations. They do not incorporate a reported multiplicity correction, and the registry does not provide confidence intervals for its overall result. These limits matter when comparing the recalculation with the sponsor analysis or using the estimates for planning.

10. Confidence Intervals and Precision

The recalculated confidence intervals describe uncertainty around each pairwise mean difference, conditional on the simple method used. For twice-daily vehicle versus once-daily vehicle, the estimated difference was -0.5 lesions and the 95% confidence interval was -4.155 to 3.155. The interval includes zero and covers effects in either direction, so this contrast is imprecisely estimated relative to the range of values it allows.

For the active groups, the recalculated intervals exclude zero: -7.288 to -0.912 for 4.0% once daily, -7.198 to -0.802 for 7.5% once daily, and -7.726 to -1.274 for 7.5% twice daily. These intervals indicate uncertainty around the estimated mean differences, rather than certainty that the true difference equals the point estimate. Their widths also show that several values remain compatible with the posted summaries.

General statistical context: a confidence interval is not a range containing a fixed proportion of individual participant outcomes. It concerns the uncertainty of an estimated population quantity under the analysis assumptions. The standard deviations reported for arms instead describe variability in participant changes. These two kinds of spread answer different questions.

The intervals are recalculated from summary statistics and rely on an approximate pooled standard deviation. They are unadjusted for multiple comparisons. The sponsor’s reported overall analysis has no posted confidence interval, so its precision cannot be assessed from the registry result. In addition, the registry record does not report details needed to assess whether the sponsor’s model accounted for covariates or other design features. Confidence intervals from the simple recalculations should therefore be read as estimates under that simplified approach, not as replacements for sponsor-model intervals.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size figures are planning illustrations based on the observed comparison effects, not the original trial design. For these continuous-outcome comparisons, the listed figures are participants per arm, not total participants across both arms. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The registry record does not report the original sample-size justification or the assumptions used to design the trial.

The 80% power column means the number of participants per arm estimated to give an 80% chance of detecting the selected effect under those planning assumptions, if that effect were the true one. The 90% power column uses a higher target probability of detection and therefore generally calls for more participants per arm for the same assumed effect. These are prospective planning quantities, not descriptions of what happened in the completed trial.

Rows that plan for a smaller fraction of the observed effect require more participants per arm: smaller differences are harder to distinguish from random variation when other assumptions stay fixed. This is not observed power. Observed results are described by the estimates, uncertainty intervals, and analyses; a planning calculation based on the observed effect does not measure the probability that the completed study succeeded.

The figures are specific to each comparison and its observed estimate. They should not be interpreted as a single recommended sample size for a future study without considering the original design, analysis plan, multiplicity, outcome variability, and other assumptions, many of which are not reported in the registry record. The assumptions stated for these illustrations do not establish what the sponsor used when planning the trial.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.500; pooled standard deviation σ = 9.430.
2. n = 2 × (1.96 + 0.842)² × 9.430² / -0.500² = 5,583.6, rounded up to 5,584 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -0.5                      # observed difference in means
sigma <- 9.42995625652632                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(2431052)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 5,584 (80% power), 7,475 (90% power)
 75% of observed effect: n per arm = 9,927 (80% power), 13,289 (90% power)
 50% of observed effect: n per arm = 22,335 (80% power), 29,900 (90% power)
power.t.test: 5584.6 
Simulated power at n = 5,584 per arm: 0.799 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Analyses Reported by the Sponsor

The registry reports a Cochran-Mantel-Haenszel analysis across the five groups with p < 0.001. It does not give an effect estimate, confidence interval, or more detailed specification for that analysis. Separately, the pairwise comparisons against once-daily vehicle are simple unadjusted re-calculations from posted group means and standard deviations.

The recalculated p-values are 0.789 for twice-daily vehicle, 0.012 for 4.0% once daily, 0.014 for 7.5% once daily, and 0.006 for 7.5% twice daily. They do not reproduce the registry’s overall p-value because they address individual contrasts using a simplified method rather than the reported overall analysis. The registry record does not explain the relation between the overall test and those pairwise comparisons.

Where the sponsor’s analysis and recalculation differ, the sponsor’s own analysis takes precedence. The recalculated values can help readers understand the scale and precision of pairwise differences from the posted summaries, but should not be described as official trial findings. The absence of the sponsor’s estimate and interval prevents a direct comparison of effect size and precision across the two approaches.

14. Multiplicity, Missing Data and Analysis Populations

Four pairwise comparisons were recalculated against the once-daily vehicle group. Their p-values are unadjusted. When multiple hypotheses are tested, the chance of at least one apparently unusual result can increase if each test is treated separately. General statistical context: multiplicity procedures address this issue by controlling an error rate or by defining a testing hierarchy. The registry record does not report whether the sponsor used a multiplicity adjustment, a hierarchy, or another strategy for the primary endpoint comparisons.

The reported Cochran-Mantel-Haenszel p-value is an overall result across five groups, but the registry does not report its test statistic, effect estimate, confidence interval, or detailed interpretation. It is not possible from the posted information to determine whether it was intended to replace, precede, or complement particular pairwise tests. The re-calculations should therefore remain clearly distinguished from the sponsor’s reported result.

The primary outcome population is listed as Intent-to-Treat. The registry record does not report the precise population definition, the number included in each outcome analysis beyond the posted arm counts, or how participants with missing Week 12 measurements were handled. It does not report whether missing observations were imputed, whether a repeated-measures model was used, or whether sensitivity analyses were conducted. These details can affect both estimates and uncertainty.

The registry also does not report the analysis model behind the least-squares means, any covariate adjustment, or whether the reported standard deviations correspond to the same estimand as those means. The original statistical analysis plan and interim-analysis details are not reported in the registry record. These omissions limit the ability to reproduce the sponsor analysis or assess its assumptions. No secondary outcome results are posted in the outcome information.

15. 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: uterine leiomyoma (0.9%).

The registry lists serious adverse events for each group with the number at risk and number affected. There were no serious adverse events reported among 52 participants in the once-daily vehicle group, 50 in the twice-daily vehicle group, 106 in the 4.0% once-daily group, or 101 in the 7.5% twice-daily group. One serious adverse event was reported among 109 participants at risk in the 7.5% once-daily group. The listed serious term was uterine leiomyoma, reported as 0.9%.

No deaths were reported in any group. The registry also lists other adverse events affecting 0 participants in each vehicle group, 4 in the 4.0% once-daily group, 6 in the 7.5% once-daily group, and 7 in the 7.5% twice-daily group. These are counts of participants affected in the registry’s categories, not counts of all event episodes.

The safety denominators are not identical to every efficacy-arm count; for example, the 7.5% once-daily safety group lists 109 at risk while the efficacy result lists 110. The registry record does not explain this difference. These figures are descriptive, and the small number of reported serious events does not by itself establish comparative safety. The record does not provide the detail needed to assess event timing, duration, relatedness, or complete safety follow-up.

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

The posted outcome data show mean reductions in inflammatory lesion counts from baseline to Week 12 in all five groups. The average changes were more negative in the active-dose groups than in the once-daily vehicle group, while the two vehicle-group means were close. The recalculated pairwise estimates quantify these differences, but they are not the sponsor’s reported overall analysis.

The sponsor-reported result is an overall Cochran-Mantel-Haenszel p-value of 0.001 across the five groups, with no posted estimate or confidence interval. The pairwise re-calculations are unadjusted and use an approximate pooled standard deviation. Given the documented discrepancy, the sponsor’s own analysis takes precedence. The record does not provide enough methodological detail to reconcile the overall test with the pairwise calculations.

These results do not establish that every participant had the group-average change, or that one regimen is preferable for an individual. Group means conceal variation in individual outcomes. Statistical evidence about a difference is also distinct from whether its size is important to participants; the registry record does not specify a threshold for clinical importance.

The results are limited to the posted primary endpoint, the stated population, and the baseline-to-Week-12 interval. They do not establish longer-term effects, effects on other outcomes, or conclusions beyond the studied population. Secondary outcome results are not posted in the registry information. Missing-data methods and important model details are also not reported, which limits independent assessment of the estimates.

Safety information should be read separately from efficacy summaries. The record lists one serious adverse event in one active-dose group and none in the other groups, as well as other adverse-event counts. Those descriptive counts are not sufficient to determine comparative risks, particularly without further event detail or a larger body of safety evidence. No medical conclusion should be drawn from the statistical summaries alone.

17. Why This Trial Matters Statistically

This trial is statistically interesting because it compares multiple dose and schedule regimens against vehicle while reporting a continuous change endpoint. That design yields several possible contrasts, not just one active-control comparison. The choice of primary contrast, handling of multiplicity, and interpretation of the overall test are therefore central to understanding the evidence.

The registry provides arm-level least-squares means and standard deviations, plus a sponsor-reported overall test. Those summaries permit simple pairwise re-calculations, but they do not reveal the full model behind the sponsor’s result. The distinction between a model-based overall analysis and unadjusted summary-statistic comparisons illustrates why estimates that look similar in subject matter may answer different statistical questions.

Precision also matters. The active-dose pairwise re-calculations have confidence intervals that exclude zero, while the twice-daily vehicle comparison has an interval that includes zero and is broad enough to allow differences in either direction. Such intervals convey more than a significance label: they show the range of estimates compatible with the simplified method. They do not, however, address multiplicity or replace the sponsor’s analysis.

Finally, the record makes clear how incomplete reporting can constrain interpretation. The original sample-size rationale, detailed analysis model, missing-data procedures, and multiplicity strategy are not reported. Planning illustrations based on observed effects are not evidence of observed power or of the assumptions used in the trial’s design. A careful reading keeps these statistical limits separate from the numerical pattern in the arm summaries.

18. 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 registry record leaves important features of the analysis unspecified. Although the primary endpoint is described as a least-squares mean in the Intent-to-Treat population, the model, covariates, contrast definitions, and handling of repeated or missing measurements are not reported. The reported standard deviations may not capture all features of the sponsor’s model-based analysis.

The sponsor’s reported overall p-value has no accompanying effect estimate or confidence interval. The record identifies a Cochran-Mantel-Haenszel method but does not explain how that method was applied to this continuous endpoint. The simple pairwise re-calculations do not resolve that reporting gap and differ from the sponsor’s reported result; the sponsor analysis takes precedence.

The pairwise re-calculations are unadjusted, and no multiplicity strategy is reported. Their p-values should not be treated as confirmatory tests without knowing the planned testing procedure. The registry record also does not report the original sample-size justification, interim analyses, or the detailed statistical analysis plan. Planning figures based on observed effects cannot fill those gaps.

The efficacy groups and safety groups have slightly different listed denominators in at least one arm, and the reason is not reported. The primary results are limited to inflammatory lesion-count change at Week 12; secondary outcomes are not posted in the registry information. Further details about adverse-event timing, relatedness, and follow-up are not reported, limiting interpretation of the sparse serious-event counts.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Dose-ranging Study of DRM01 in Subjects With Acne Vulgaris — NCT02431052.
• Registry record last updated 2021-07-20; status last verified 2021-07.
• Results first posted 2019-02-26.
• 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.

22. Record Summary

The registry describes a completed Phase 2 randomized, triple-masked, parallel, vehicle-controlled dose-ranging study of topical olumacostat glasaretil in subjects with acne vulgaris. Enrollment was 420. The study compared two vehicle regimens with 4.0% once daily, 7.5% once daily, and 7.5% twice daily, applied to the face for 12 weeks.

The primary endpoint was mean absolute change in inflammatory lesion counts from baseline to Week 12 in the Intent-to-Treat population. All five arm means were negative. The registry reports an overall Cochran-Mantel-Haenszel p-value of 0.001, without an effect estimate or confidence interval. Simple pairwise re-calculations are unadjusted and should not override the sponsor’s reported analysis.

The record lists no deaths and one serious adverse event in the 7.5% once-daily group. Details of the original sample-size rationale, analysis model, multiplicity strategy, and missing-data handling are not reported in the registry record.