Effects of Isotretinoin on CYP2D6 Activity: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effects of Isotretinoin on CYP2D6 Activity (NCT03076021), a na phase 4 study. The primary outcome is CYP2D6 Activity as Measured by the Molar Concentration Ratio of Dextromethorphan (DM)/Dextrophan (DX).

2. Clinical Question

The statistical question is about a within-person change in a metabolic ratio, not a comparison of independently assigned treatment groups. That distinction matters: the two measurements from a participant are related, and the natural analysis would need to account for that pairing. The actual analysis method is not reported in the registry record. The record also does not state how many participants contributed both endpoint measurements or how the analysis handled any missing measurements.

3. Trial Design

This is a completed Phase 4, single-group study sponsored by the University of Washington. The registry lists an enrollment of 33 and describes the study population as adolescents. The study started on 2016-07-26, and the primary completion date is 2023-11-27. The listed condition is CYP2D6 polymorphism. Allocation is recorded as not applicable, masking as none, and the model as single group.

The group is described as adolescents receiving dextromethorphan before and after isotretinoin. In the primary endpoint description, each person serves as their own control: CYP2D6 activity is measured before isotretinoin and with isotretinoin. This structure can reduce the influence of stable differences between participants because each participant contributes to both conditions. It does not, by itself, remove possible effects of time, changes in other exposures, or differences in measurement circumstances.

Because this is not a randomized comparison, the pre-isotretinoin and with-isotretinoin observations should not be read as independent treatment arms. The registry itself cautions that the arms are not a like-for-like control comparison. The timing is described broadly as 1 week to 6 months for the primary outcome; more detail about the exact measurement schedule is not reported in the registry record. The record does not report the sequence or duration of individual treatment periods, dose, adherence, or other protocol details.

The enrollment count does not establish the number included in the endpoint analysis. The registry says that only cytochrome P450 extensive metabolizers were included in that analysis and that the analysis population was smaller than the number who completed the study. The exact number analyzed, the number completing the study, and reasons for exclusions are not reported in the registry record. No separate randomized control group is listed.

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.

Pre-isotretinoin

Isotretinoin

Analysis population for the primary outcome, as defined in the registry: The number of subjects included in the analysis is less than the number that completed the study because only cytochrome P450 extensive metabolizers were included in the analysis. Each person served as their own control (pre-isotretinoin and with isotretinoin). Pre-isotretinoin study was completed prior to starting isotretinoin and then the same study was completed after the subject had been on isotretinoin for 1 week to 6 months.

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: All subjects that completed the study.

6. Primary Endpoint

The registry lists the primary outcome as CYP2D6 Activity as Measured by the Molar Concentration Ratio of Dextromethorphan (DM)/Dextrophan (DX), measured over 1 week to 6 months.

The primary endpoint is CYP2D6 activity measured by the molar concentration ratio of dextromethorphan (DM) to dextrophan (DX). Dextromethorphan is used as a probe, and the ratio represents the relative concentrations of the parent compound and its metabolite. The registry describes the ratio as unit-less because the molar concentration units cancel. The endpoint is specified as a geometric mean with a 90% confidence interval, over the timeframe of 1 week to 6 months.

As a general statistical interpretation, a metabolic ratio is used as an indirect measure of enzyme activity: a change in the balance between the parent compound and metabolite can reflect a change in metabolism. The direction and magnitude of that interpretation depend on the endpoint definition and the study context. The registry’s stated aim was to investigate whether isotretinoin decreases CYP2D6 activity; that aim is not itself a result.

A geometric mean is suited to positive measurements that are more naturally interpreted on a multiplicative scale. A confidence interval expresses uncertainty around an estimate under the statistical method used; it does not show the range of individual participant values. The actual geometric means, confidence interval limits, and comparison estimate are not reported in the registry record. The registry also does not report the laboratory procedures or the specific calculation used to produce the endpoint estimates.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is a continuous metabolic ratio, and the registry specifies that it be summarized as a geometric mean with a 90% confidence interval. The comparison is between measurements made in the same participant before and with isotretinoin. The registry does not report the detailed analysis procedure, so the following concepts explain how such summaries and paired comparisons are generally interpreted rather than describing a confirmed analysis plan for this study.

What a geometric mean conveys

General: The arithmetic mean summarizes values by adding them and dividing by their count. The geometric mean instead works on a multiplicative scale: values are transformed, averaged, and returned to the original scale. It is often used for positive measurements whose variation is asymmetric or whose effects are naturally expressed as relative changes. A geometric mean describes a central tendency; it does not convey the full spread of individual responses.

For a paired design, there are different ways to summarize the measurements. An analysis might summarize each condition separately, or it might analyze within-person changes on a transformed scale and report a contrast. Those approaches do not answer exactly the same descriptive question. The registry record gives the endpoint summary type but does not say which paired estimator or transformation was used, nor does it provide the resulting estimates.

How to read a confidence interval

General: A confidence interval is an uncertainty interval calculated from the observed data under a specified statistical method and its assumptions. A 90% confidence interval describes the performance of the interval procedure over repeated studies under those assumptions; it is not a range that contains a fixed proportion of individual participants. The interval’s width reflects statistical precision: a narrower interval indicates less uncertainty about the estimated quantity under the method, while a wider interval indicates more.

For this endpoint, the registry specifies a 90% confidence interval but does not report its limits. Consequently, the estimate’s precision cannot be assessed from the registry record. The interval also cannot be used to determine whether a comparison is compatible with no change unless the estimate and the relevant contrast are defined and reported.

Why pairing matters

General: Two observations from one person are usually related because they share characteristics of that person. A paired analysis accounts for that relationship by focusing on within-person changes or by using a model that represents the correlation. Treating paired observations as if they came from unrelated people can misrepresent uncertainty. The study description says each participant served as their own control, but the registry record does not identify the statistical method used to account for pairing.

A within-person design can control for participant characteristics that remain stable across measurement occasions. It cannot automatically distinguish an isotretinoin effect from other changes occurring between the measurements. That is why timing, measurement procedures, concurrent exposures, and protocol details matter when interpreting a pre–post comparison. Those details are not reported in the registry record.

What the planning assumptions mean

General: A sample-size calculation estimates the amount of information needed to detect a specified effect under chosen assumptions. The registry describes planning illustrations using a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect expressed as a stated fraction of the observed effect. It also distinguishes participant counts per arm from total event counts for time-to-event calculations. These are descriptions of planning assumptions, not evidence that the trial was designed or analyzed using those assumptions.

The record does not report the observed effect used for the illustrations or the resulting participant requirements. The actual trial is a single-group, within-person study rather than a randomized two-arm comparison, so the equal-allocation assumption belongs to the planning illustration and should not be mistaken for the study’s allocation. The trial’s own sample-size justification is not reported in the registry record. Any planning count is an illustration, not an observed power estimate, and it cannot establish whether the endpoint analysis was sufficiently precise.

What cannot be reconstructed

The registry lists one primary outcome measure and zero analyses. It does not report the endpoint estimates, confidence interval limits, p-value, analysis population count, missing-data approach, or model specification. General statistical methods cannot fill in those absent study results. Without those elements, readers cannot verify the analysis, compare the reported uncertainty with a clinically meaningful threshold, or determine how strongly the data support a particular direction of change.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and zero analyses. This means the record does not provide a posted statistical analysis for the primary endpoint. It does not establish whether additional outcomes were collected or whether analyses were performed outside the registry record. A multiplicity strategy, including any adjustment for multiple comparisons or outcomes, is not reported in the registry record. With no posted endpoint analysis, it is not possible to assess how multiplicity was handled or whether the reported endpoint was selected from a larger set of measurements.

The primary endpoint population consisted only of cytochrome P450 extensive metabolizers, and the registry states that this population was smaller than the number who completed the study. The number included in the analysis, the number who completed the study, and the reasons for the difference are not reported in the registry record. The number of participants with paired endpoint measurements is also not reported in the registry record.

A missing-data method, including any imputation, exclusion rule, or sensitivity analysis, is not reported in the registry record. General: Missingness can affect both the estimate and its uncertainty, especially when the likelihood of missing data is related to participant characteristics or outcomes. Without counts and reasons for missing measurements, readers cannot judge whether the analyzed participants may differ systematically from those not analyzed.

The record does not report a protocol-defined analysis population, exclusions beyond the extensive-metabolizer restriction, or a plan for handling protocol deviations. These omissions do not prove that such procedures were absent; they mean they are not reported in the registry record.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry lists safety figures for two groups: Pre-isotretinoin and With Isotretinoin. In each group, 33 participants were at risk for serious adverse events and 0 were affected. In each group, 33 participants were at risk for other adverse events and 0 were affected. Deaths were 0 in each group. These are the posted counts for the categories recorded.

These figures should be read as registry-reported event counts, not as a complete account of every possible safety outcome. The record does not provide event narratives, severity grading beyond the serious-versus-other grouping, timing details, or an explanation of how events were collected. It also does not report whether the same participants contributed to both group entries, though the study is described as a within-person pre–post design.

Zero recorded events in a category does not establish that the risk is zero. General: When no events are observed, uncertainty about the underlying event frequency remains, and the strength of a safety conclusion depends on the amount and duration of observation and on how events were ascertained. The registry record does not provide enough detail to quantify that uncertainty. The safety counts do not provide evidence about the primary CYP2D6 activity endpoint.

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

The study is designed to examine a specific pharmacologic question: whether CYP2D6 activity, measured through the DM/DX metabolic ratio, changes between the pre-isotretinoin and with-isotretinoin measurements in adolescents. The within-person structure makes each participant their own comparator, and the endpoint is restricted to cytochrome P450 extensive metabolizers. These features define the population and comparison to which any result would apply.

The registry record does not report numerical results for the primary endpoint. It does not give pre-isotretinoin or with-isotretinoin geometric means, the 90% confidence interval limits, a contrast between conditions, or a test result. Therefore, the posted record does not establish the direction, magnitude, or precision of any change in CYP2D6 activity. The stated research aim should not be mistaken for a finding.

Even if a pre–post difference were reported, the design would require careful interpretation. General: A paired comparison can account for stable differences between participants, but without random assignment it cannot rule out time-related changes or other factors that coincide with the exposure. The registry record does not provide enough protocol or analysis detail to evaluate these potential explanations.

The enrollment of 33 is not the endpoint analysis count. The record says that only extensive metabolizers were included and that fewer participants were analyzed than completed the study, but the exact counts are not reported. This leaves the size and precision of the endpoint analysis unknown. The safety entries show 0 affected for the listed adverse-event categories and 0 deaths in each group; those entries should not be extended into broader conclusions beyond the recorded categories.

The most defensible reading is that the registry describes the study question, endpoint, and pre–post design, but does not provide the primary numerical result needed to evaluate the question. Neither a positive nor a null finding can be inferred from the information recorded.

13. Why This Trial Matters Statistically

This study raises a useful statistical design question: how to compare a pharmacologic measure before and during an exposure when participants serve as their own controls. Within-person data can remove some between-person variation from a comparison, but the gain in precision depends on the relationship between paired measurements and on the analysis correctly accounting for that pairing. The registry describes the design but does not report the analysis method or its numerical results.

The endpoint also illustrates why the summary scale matters. The DM/DX ratio is unit-less, and the registry specifies a geometric mean rather than an arithmetic mean. General: When a biological measure is positive and relative changes are meaningful, a geometric summary can be informative. Its interpretation still depends on the distribution, calculation method, and uncertainty interval. Those details cannot be fully evaluated without the endpoint estimates and analysis description.

Another important issue is the population restriction. Only cytochrome P450 extensive metabolizers were included in the primary endpoint analysis, and the number analyzed is not reported. Clear reporting of the analysis population and exclusions is essential for understanding who contributes to an estimate and how broadly it may apply.

The record also demonstrates the distinction between a study aim and a study result. A stated hypothesis about a possible decrease does not show that a decrease was observed. An endpoint name and an intended confidence interval do not substitute for the estimate, its uncertainty, and a transparent description of the analysis.

Finally, the registry’s planning assumptions should be kept separate from evidence generated by the study. Sample-size illustrations based on an observed effect can help frame hypothetical planning, but they are not observed power and do not establish the adequacy of the actual analysis. The trial’s own design justification and numerical primary result are not reported in the registry record.

14. Limitations

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

The central limitation for interpretation is that the numerical primary endpoint result is not reported in the registry record. The geometric means, 90% confidence interval limits, and a direct pre–post contrast are absent. A p-value and detailed analysis output are also not reported, so readers cannot assess the magnitude, direction, or precision of the primary comparison.

The analysis population is incompletely described. The record states that only cytochrome P450 extensive metabolizers were included and that fewer participants were analyzed than completed the study, but it does not report either count or the reasons for the difference. The number with complete paired measurements is not reported. This makes it impossible to evaluate the extent of exclusions or the effect of missing measurements on the analysis.

The pre–post design has no separate randomized control group. A within-person comparison can address stable participant differences, but changes over time and other factors coinciding with isotretinoin exposure remain possible explanations for a difference. The registry record does not report enough detail about timing, dose, adherence, measurement procedures, concurrent exposures, or protocol conduct to assess these possibilities.

The endpoint timeframe is stated as 1 week to 6 months, but the exact schedule and the timing of measurements for individual participants are not reported. The statistical model, transformations, handling of outliers, missing-data methods, and multiplicity plan are also not reported in the registry record. These omissions limit reproducibility and make it difficult to judge whether the methods matched the paired design and endpoint scale.

The safety record gives counts for serious and other adverse events and deaths, but event details and ascertainment procedures are not reported. These limitations concern what can be concluded from the registry record; they do not establish that unreported procedures or analyses were not performed.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Effects of Isotretinoin on CYP2D6 Activity — NCT03076021.
• Registry record last updated 2025-01-28; status last verified 2025-01.
• Results first posted 2025-01-28.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

NCT03076021 is a completed Phase 4, single-group study sponsored by the University of Washington. It enrolled 33 adolescents and examined CYP2D6 activity before and with isotretinoin, using dextromethorphan and the DM/DX molar concentration ratio. The primary endpoint was specified as a geometric mean with a 90% confidence interval over 1 week to 6 months. Only cytochrome P450 extensive metabolizers were included in the endpoint analysis, but the analysis count is not reported in the registry record.

The design is a within-person pre–post comparison, not a randomized comparison between separate groups. The primary endpoint estimates, confidence interval limits, formal test result, analysis details, and missing-data approach are not reported in the registry record. The listed safety groups each report 0 serious adverse events, 0 other adverse events, and 0 deaths among 33 participants at risk for each adverse-event category.