Selumetinib in Treating Patients With Neurofibromatosis Type 1 and Cutaneous Neurofibroma: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Selumetinib in Treating Patients With Neurofibromatosis Type 1 and Cutaneous Neurofibroma (NCT02839720), a na phase 2 study. The primary outcome is Median Best Response of Cutaneous Neurofibromas in Participants With at Least One Restaging Evaluation.

2. Clinical Question

3. Trial Design

This was a pilot phase II, single-group study sponsored by the National Cancer Institute. The registry record lists 11 enrolled participants and gives the study status as completed. The recorded start date is 2017-08-26, and the primary completion date is 2022-11-05. These dates and the completed status describe the trial record; they do not by themselves establish that every planned assessment was completed or that every participant contributed to the primary endpoint.

Participants received selumetinib sulfate orally twice daily on days 1–28, with treatment repeating every 28 days for up to 24 cycles if disease progression or unacceptable toxicity did not occur. The treatment description includes an additional, truncated statement about continuing treatment for participants who experienced a volume decrease in target cutaneous neurofibromas. The rest of that statement is not reported in the registry record, so its duration and conditions cannot be fully described.

Masking is recorded as none. A laboratory biomarker analysis is also listed as an intervention, but its procedures and role in the analysis are not reported in the registry record. The listed secondary outcome is change in the number of cutaneous neurofibromas. Five outcome measures are recorded in total, but the names and definitions of the other measures are not reported in the registry record.

Because the study has one treatment group and no described concurrent comparator, its design supports a description of outcomes observed during treatment, not an estimate of a randomized treatment contrast. The registry record does not report a protocol rationale for the sample size, a statistical analysis plan, or details of interim analyses. Those design features should not be inferred from the phase label or the completed status.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Includes all participants who had at least one restaging evaluation on treatment with selumetinib

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 Median Best Response of Cutaneous Neurofibromas in Participants With at Least One Restaging Evaluation, measured over Up to 24 cycles of treatment (1 cycle = 28 days).

The primary endpoint is the median best response of cutaneous neurofibromas among participants with at least one restaging evaluation on selumetinib. Response is expressed as percent change in the volume of target cutaneous neurofibromas from baseline. The registry describes measuring each target lesion with calipers, calculating its volume from length, width, and height, and combining the target-lesion volumes for assessment. Evaluations occurred at baseline and after every 4 cycles. For each participant, the most decreased average percent change from baseline across the on-treatment evaluations was designated the best response; the median and full range of those participant-level best responses were intended for reporting.

In general, a best-response endpoint captures the most favorable measured value during the specified follow-up, rather than the value at a single fixed assessment. It can describe the observed range of response across participants, but does not show how long a response lasted unless duration is separately measured and reported. A median gives a central summary, while a full range describes the lowest and highest observed values; neither alone shows the distribution between those extremes.

Using tumor-volume change is relevant to the stated question because the endpoint focuses on the size of target cutaneous neurofibromas over treatment. However, the registry record does not report the numerical median or full range. It therefore gives no numerical estimate of the best-response distribution, and the endpoint definition should not be mistaken for an observed treatment result.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the primary endpoint summarize?

The endpoint is based on each participant’s best measured percent change in target cutaneous neurofibroma volume from baseline, followed by a median across participants with at least one restaging evaluation. In general, selecting the best observed response emphasizes the most decreased value during the observation period. It is not the same as summarizing the change at a single prespecified final visit, and it does not describe response duration by itself.

The registry describes volume measurement using calipers and a calculation based on length, width, and height for each target neurofibroma. It then describes using the sum of target-lesion volumes at evaluations to calculate overall percent change for each participant. Because a volume estimate is built from several measured dimensions, measurement variability in those dimensions can affect the calculated volume. This is a general measurement consideration, not a reported estimate of measurement error in this trial. The registry record does not report a reproducibility assessment or other measurement-variation analysis.

What do the median and full range mean?

In general, a median is a measure of the center of a set of observations: it identifies the middle value when observations are ordered. A full range reports the smallest and largest observations. These summaries can be useful when a small study has a spread of responses that is not well represented by an average alone. However, the median and range do not show the pattern of values between the endpoints of the range, and a range can be sensitive to an especially low or high observation.

For this trial, the registry record specifies the median and full range as the intended summaries, but reports neither numerical value. Consequently, no description of the observed central response or spread can be made. The number of participants contributing to the endpoint is also not reported in the registry record. The enrolled count is not a substitute for the endpoint analysis count because eligibility for the endpoint required at least one restaging evaluation.

Why does the single-group design matter?

In general, a concurrent comparator helps estimate how outcomes differ between groups under the conditions of a study. Random allocation, when used, can support a like-for-like comparison by distributing measured and unmeasured participant characteristics across treatment groups. This study is recorded as single-group, with allocation not applicable and no masking. No comparator group is described, so the design does not estimate a randomized contrast.

Observed change within a single group can be described, but it does not by itself identify the cause of that change. The registry record does not report a control condition, a comparison with another treatment, or a statistical adjustment that would create a controlled contrast. This is a design limitation rather than a judgment about the direction of any response; the numerical primary result is not reported.

How should the planning calculations be read?

The planning assumptions describe a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect to detect expressed as a stated fraction of the observed effect. In general, a sample-size calculation uses assumptions about the effect of interest, variability, error rates, and study design to illustrate how much information a planned comparison might require. Its output is a planning quantity, not a result from participants.

Those assumptions do not match the recorded trial structure, which is single-group and has no two-arm allocation. The planning illustration therefore should not be read as a description of the trial’s actual enrollment plan or as evidence that the study had a particular probability of detecting an effect. Sample-size calculations are not observed power. The numerical outputs and the observed effect value on which the illustration was based are not reported in the registry record, so the calculations cannot be evaluated numerically.

In general, a confidence interval describes a range of values compatible with an estimate under a stated statistical procedure, and its width conveys precision. A p-value addresses compatibility of data with a specified null model; it does not measure effect size or the probability that a treatment is effective. Neither a confidence interval nor a p-value is reported for the primary endpoint in the registry record. With no numerical endpoint estimate or uncertainty interval, the precision of the outcome cannot be assessed.

What role can over-dispersion play?

In general, over-dispersion occurs when count data vary more than a statistical model assumes. If present and ignored, uncertainty estimates from a model may be too small. This concept can matter for outcomes expressed as counts, such as a number of lesions, depending on the data and model. The registry lists change in the number of cutaneous neurofibromas as a secondary outcome, but does not report its numerical results or the model used to analyze it. Over-dispersion is therefore a general consideration, not a finding about this trial.

The registry record does not report a statistical analysis plan, model specifications, diagnostics, or details of any adjustment for baseline characteristics. It also records zero analyses. No particular model, variance assumption, hypothesis test, or adjustment procedure should be attributed to the investigators on the basis of the endpoint description alone.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists five outcome measures and identifies change in the number of cutaneous neurofibromas as a secondary outcome. It gives the primary endpoint definition, but does not report numerical results for that endpoint or for the named secondary outcome. The names and definitions of the other outcome measures are not reported in the registry record. The record also reports zero analyses.

When a study evaluates multiple outcomes or multiple analyses, the chance of a false-positive finding can increase if results are interpreted without accounting for the number of questions tested. In general, multiplicity methods can control or describe that issue, but the choice depends on the analysis plan and the intended claims. A multiplicity strategy is not reported in the registry record. No adjustment, hierarchy, or prespecified testing sequence can be inferred from the list of outcomes alone.

The primary endpoint population includes participants who had at least one restaging evaluation on treatment. The registry record does not report how many participants met that condition, how many lacked a restaging evaluation, or how missing assessments were handled. It does not report whether participants who discontinued treatment were included in any analysis, or how incomplete volume measurements were treated. These details are important because an analysis restricted to participants with post-baseline assessments can differ from one that includes all enrolled participants.

The enrollment count of 11 describes participants enrolled, not necessarily the number contributing to each outcome. An analysis population definition beyond the endpoint’s restaging requirement is not reported in the registry record. The record also does not report imputation methods, sensitivity analyses for missing data, or a participant flow accounting for exclusions from outcome analyses. Consequently, the impact of missing or unevaluable data cannot be assessed.

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 1% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: eye disorders - other, specify (9.1%), immune system disorders - other, specify (9.1%), neoplasms benign, malignant and unspecified (incl cysts and polyps) - other, specify (9.1%), tumor hemorrhage (9.1%), hypertension (9.1%).

The safety record lists 11 participants at risk for serious adverse events and reports five affected. For other adverse events, it lists 11 at risk and 11 affected. The recorded death count is zero. These are counts for the single treatment group; there is no comparator group described against which to judge relative event frequency.

The listed serious-event terms are eye disorders—other, specify; immune system disorders—other, specify; neoplasms benign, malignant and unspecified (including cysts and polyps)—other, specify; tumor hemorrhage; and hypertension. Each is listed with 9.1%. These terms are reported as categories, and the registry record does not state whether the same participant contributed to more than one category. They should not be added together or interpreted as distinct people without that information.

In general, a serious adverse event is a safety classification, not a statement about whether an event was caused by treatment. The registry record does not report event-level timing, duration, outcome, or attribution for these terms. It also does not provide a comparison that could establish whether event counts were higher or lower than under another condition. The zero recorded deaths is a count in this group; it does not establish that no other serious harms occurred. These figures describe the posted safety counts and should be read separately from the unreported tumor-volume endpoint result.

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

The trial record documents a pilot phase II study of selumetinib in adults with NF1 and cutaneous neurofibromas. It records 11 enrolled participants, a single treatment group, and a primary endpoint based on the best measured percent change in target cutaneous neurofibroma volume among participants with at least one restaging evaluation. It also records selected adverse-event counts and lists change in the number of cutaneous neurofibromas as a secondary outcome.

The primary endpoint’s numerical median and full range are not reported in the registry record. No numerical result for the listed secondary outcome is reported either. This means the posted result information does not show the observed size or spread of tumor-volume changes, and it does not permit an assessment of the precision of those changes. A well-defined measurement approach is not itself evidence of a particular observed outcome.

Because the study is single-group and has no comparator, any observed within-group changes would not provide a randomized comparison with another treatment or a control condition. In general, a single-group result can describe what was measured during treatment, but it cannot by itself establish that treatment caused a change. The available record does not include the numerical primary result needed even for that descriptive assessment.

The safety counts show that serious adverse events were recorded among participants and that other adverse events were recorded for all participants at risk; zero deaths were recorded. Those figures do not establish whether events were caused by treatment, and no comparative safety estimate is available. They are distinct from evidence about tumor response.

Accordingly, the registry record supports a description of the study’s design, measurement plan, enrollment, and selected safety reporting. It does not support a numerical conclusion about the primary outcome, its uncertainty, or a treatment effect relative to a comparator. The missing result information should not be replaced by assumptions based on the trial title, planned endpoint, or completion status.

13. Why This Trial Matters Statistically

This trial is statistically informative as an example of the difference between a clear endpoint definition and a reported endpoint result. The registry explains how the investigators intended to measure tumor-volume change and how they planned to summarize best response. Without the median, range, or analysis population count, however, readers cannot assess the observed response distribution or its precision. Transparent reporting of estimates and their uncertainty is needed to interpret a clinical result, not just a description of the measurement process.

The single-group design also makes the estimand—the quantity the study can directly describe—important. The record describes outcomes among people treated with selumetinib, not a difference between selumetinib and a control. In general, this distinction prevents a within-group observation from being mistaken for a randomized treatment effect. The design should shape the strength and type of statistical claims made from any reported outcomes.

The endpoint uses each participant’s best observed change across repeated restaging assessments. Statistically, that choice answers a different question from change at one fixed visit or duration of response. It can be useful for describing the most decreased measured value, while leaving persistence and timing as separate questions. The registry notes that the timepoint of best response was recorded, but does not report its numerical results.

Finally, enrollment and endpoint analysis populations are not interchangeable. The endpoint required at least one restaging evaluation, but the number of participants meeting that criterion is not reported. Clear accounting of who contributed data, how missing assessments were handled, and how multiple outcomes were treated would help readers evaluate the result if numerical estimates were available. In this record, those details and the primary numerical result are not reported.

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 principal limitation for statistical interpretation is that the registry record defines the primary endpoint without reporting its numerical median or full range. The number of participants with at least one restaging evaluation is also not reported. Although enrollment is recorded as 11, the number contributing to the primary endpoint cannot be determined from that count alone.

The single-group design has no described concurrent comparator, and masking is recorded as none. Therefore, the record does not offer a randomized between-group estimate. The primary endpoint is based on the best observed response during treatment, which is not the same summary as a result at one fixed follow-up assessment. The record states that the timepoint of best response was noted, but does not report those timepoint results.

Outcome reporting is incomplete for statistical assessment: five outcome measures are recorded, but only the primary endpoint and one secondary outcome are described in the available record. Numerical results for both are not reported, and the record gives zero analyses. A statistical analysis plan, multiplicity approach, and model details are not reported in the registry record. Neither the handling of missing assessments nor a detailed analysis-population flow is reported.

The treatment description is truncated after a statement about participants whose target neurofibromas decreased in volume. The remaining continuation criteria are not reported in the registry record. The record also lists a laboratory biomarker analysis without describing its methods or analytical role. These gaps limit reconstruction of the complete treatment and measurement procedures from the registry record. They do not establish that the procedures were absent; they indicate that the relevant details are not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Selumetinib in Treating Patients With Neurofibromatosis Type 1 and Cutaneous Neurofibroma — NCT02839720.
• Registry record last updated 2023-12-06; status last verified 2023-12.
• Results first posted 2023-12-05.
• 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

NCT02839720 is a completed pilot phase II, single-group study sponsored by the National Cancer Institute. It enrolled 11 participants and evaluated selumetinib sulfate in adults with NF1 and cutaneous neurofibromas. The recorded start date is 2017-08-26, and the primary completion date is 2022-11-05.

The primary endpoint is the median best percent change in target cutaneous neurofibroma volume among participants with at least one restaging evaluation. The registry describes measurements at baseline and after every 4 cycles, but does not report the numerical median, full range, or endpoint analysis count. Change in the number of cutaneous neurofibromas is listed as a secondary outcome; its numerical result is not reported.

The safety record reports five participants with serious adverse events among 11 at risk, 11 with other adverse events among 11 at risk, and zero deaths. The study has no described comparator group, and the registry record does not report a statistical analysis plan or numerical primary outcome results.