Sorafenib-induced Hand- Foot Skin Reaction Treatment: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Sorafenib-induced Hand- Foot Skin Reaction Treatment (NCT00667589), a randomized phase 2 study. The primary outcome is Change in Skindex-16 Total Score Between Baseline and 8 Weeks.

2. Clinical Question

3. Trial Design

The registry describes NCT00667589 as a phase 2, randomized, double-blind, parallel study sponsored by Northwestern University. Its title identifies a four-arm evaluation of urea 40% cream, fluocinonide 0.05% cream, tazarotene 0.1% cream, and an emollient cream for hand-foot skin reaction associated with sorafenib. The brief summary likewise describes the study as evaluating treatments for a rash caused by sorafenib.

The registry reports enrollment of 6. It records a start date of 2008-06 and a primary completion date of 2010-12, and the study status is terminated. A reason for termination is not reported in the registry record. The arm descriptions are empty, so details about how the creams were applied, treatment duration, and other arm-specific procedures are not reported in the registry record.

The primary endpoint is change in Skindex-16 total score from baseline to 8 weeks. A secondary outcome is change in the same score between baseline and 2 weeks. The record lists 2 outcome measures but lists no analyses. Although the study is described as having four treatment arms, the record says no subjects were randomized to the emollient arm. It also says that several participants did not contribute an 8-week questionnaire to the primary outcome analysis. These features leave the posted record without a reported multi-arm comparison for the primary endpoint.

Randomization and double masking describe aspects of the intended study design, but the registry record does not report the allocation procedure, who was masked, or how masking was maintained. A detailed statistical analysis plan, sample-size justification, and interim-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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Results for subjects from the fluocinonide 0.05% cream arm, subjects from the Tazarotene 0.1% cream arm, and one subject from the Urea 40% cream arm are not included in analysis because these subjects did not complete a Skindex-16 questionnaire at 8 weeks. No subjects were randomized to the Udderly Smooth Emollient arm.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Change in Skindex-16 Total Score Between Baseline and 8 Weeks, measured over baseline and 8 weeks.

The primary endpoint is change in Skindex-16 total score between baseline and 8 weeks. The questionnaire has 16 questions about quality of life in people with skin disease. Its total score ranges from 0 to 96; the registry description associates 0 with better quality of life and 96 with worse quality of life. A change score is intended to capture how a participant’s reported skin-related quality of life differs from their baseline assessment over the stated follow-up period.

Using a patient-reported quality-of-life measure can complement clinical descriptions of a skin condition: symptoms and their effects on daily life may be important even when a study also evaluates clinical features. That is a general reason such measures are used, not a claim about the results of this trial. The record does not report a threshold for a clinically meaningful Skindex-16 change, nor does it report the observed change by treatment arm. It therefore does not establish how large a change would be important to participants or whether any arm’s score changed.

The registry describes the primary analysis population as excluding participants in the fluocinonide and tazarotene arms and one participant in the urea arm who did not complete the questionnaire at 8 weeks. The resulting endpoint information is not reported in the registry record as a numerical estimate or comparison.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The record identifies a randomized, double-blind, parallel design with four named cream interventions. General statistical interpretation starts by distinguishing the design from the analysis: randomization describes how treatment assignment was intended to occur, while an analysis describes how outcomes were compared. The registry record does not describe the randomization procedure, the masking arrangements, or a statistical analysis plan. It lists no analyses, so the planned or performed comparison method is not reported in the registry record.

What does a change score represent?

For the primary endpoint, the stated quantity is the difference between an individual’s Skindex-16 score at baseline and the score at 8 weeks. In general, a change score summarizes within-person movement over time. A group summary of those changes can describe what happened among participants with observed assessments, but a comparison of group summaries is needed to estimate a difference between treatments. The registry record provides neither the numerical group changes nor a treatment contrast. It also does not state whether the analysis used a particular model, adjustment, or method for handling incomplete questionnaires.

The score’s direction matters when interpreting any future estimate: the record associates lower values with better quality of life and higher values with worse quality of life. The record does not define the sign convention for reporting change, such as whether the calculation subtracts baseline from follow-up or follow-up from baseline. Without that convention and the results, a direction of change cannot be assigned from the posted record.

Why does missing outcome information matter?

General statistical reasoning distinguishes participants who were randomized from those who have an observed endpoint. A randomized participant without an 8-week questionnaire cannot contribute a directly observed 8-week change score. Omitting such observations can reduce the information available and can affect interpretation if participants with missing questionnaires differ from those with completed questionnaires. The registry gives a reason for some exclusions—questionnaires were not completed at 8 weeks—but does not report why the questionnaires were missing, whether other endpoint data were available, or whether any method was used to address missingness.

The description says results from subjects in the fluocinonide and tazarotene arms and one subject in the urea arm are excluded from the primary analysis for this reason. It also says no subjects were randomized to the emollient arm. The number of participants contributing to each primary endpoint group is not fully reported in the registry record. No imputation or other missing-data procedure is reported in the registry record. These omissions prevent assessment of how the missing observations were handled beyond the stated exclusions.

What can a confidence interval add?

In general, a confidence interval gives a range of values that are compatible with an estimate under the method used to calculate it. Its width conveys precision: a wider interval indicates less precise information than a narrower interval. Such an interval does not eliminate uncertainty or automatically determine whether an observed difference is important to patients. The registry record reports no confidence interval for the primary endpoint, and it does not report the estimate to which one could attach an interval. The endpoint’s precision therefore cannot be assessed.

What does a significance test answer?

In general, a significance test evaluates how compatible data are with a specified null hypothesis under stated assumptions. A p-value alone does not measure the size or clinical importance of an effect. The registry record does not report a primary-outcome p-value or the null hypothesis and test used. Its general sample-size illustration assumptions mention a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, but those assumptions do not establish that this was the trial’s own analysis plan. The illustration is described as based on an observed effect, not as observed power. No sample-size illustration values are reported in the registry record.

Why does over-dispersion matter in general?

Over-dispersion occurs when observed variability is greater than a statistical model assumes. In general, failing to account for it can make uncertainty estimates too small and tests too optimistic. Whether over-dispersion is relevant depends on the outcome and model. The primary endpoint is a questionnaire score, but the registry record does not report a model or analysis from which to assess over-dispersion. It would therefore be inappropriate to claim that this trial used a method to address it or that it affected the results.

The safest statistical reading is to keep design details, endpoint definitions, and results separate. The design and outcome are described, while the primary estimates, uncertainty measures, and analysis method are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The study lists four treatment arms and two outcome measures: change in Skindex-16 total score between baseline and 8 weeks, and change between baseline and 2 weeks. Considering multiple treatments and outcome time points can create multiplicity: in general, the more comparisons examined, the greater the chance of finding an apparently notable result by chance if no adjustment or prespecified hierarchy is used. Whether and how to address multiplicity depends on the analysis plan and the questions intended to be primary. A multiplicity strategy is not reported in the registry record.

The record identifies the 8-week change as the primary outcome and the 2-week change as a secondary outcome. It does not report results for either measure, identify planned pairwise comparisons, or explain how the secondary endpoint would be interpreted relative to the primary endpoint. No analysis is listed in the registry record. It is therefore not possible to determine whether comparisons were prespecified, adjusted for multiplicity, or conducted at all.

Missing 8-week questionnaires are explicitly noted. The primary description excludes results from the fluocinonide and tazarotene arms and one subject from the urea arm because those subjects did not complete the questionnaire at 8 weeks. No subjects were randomized to the emollient arm. The registry record does not report the reasons for missing questionnaires, an analysis population definition, or a missing-data method such as imputation. It also does not report whether the 2-week endpoint was complete.

In general, an intention-to-treat analysis seeks to preserve the randomized group assignments, while analyses restricted to participants with observed outcomes answer a question about the analyzed subset and may be vulnerable to bias when missingness is related to outcomes. The registry record does not identify an intention-to-treat analysis or another analysis population. With no primary estimates and no documented method for handling missingness, the extent to which the posted results would represent randomized participants cannot be assessed.

11. 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 5% reporting threshold.

The registry reports serious and other adverse-event counts by group. For fluocinonide 0.05% cream, 1 was listed at risk for each category, with 0 affected by serious adverse events and 0 affected by other adverse events. For Udderly Smooth® emollient, 0 was listed at risk for each category, with 0 affected in both categories. For tazarotene 0.1% cream, 3 was listed at risk for each category, with 0 affected in both categories. For urea 40% cream, 2 was listed at risk for each category, with 0 affected in both categories.

These are counts in the registry’s reported risk sets, not estimates of comparative safety. The emollient group has no participants listed at risk, consistent with the statement that no subjects were randomized to that arm; its zero affected count should not be read as observed evidence of safety in treated participants. The record’s small enrollment and limited reported risk sets also mean that zero reported events do not establish that an event could not occur. The registry does not report deaths, and the reason for termination is not reported in the registry record.

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

The registry record establishes that a randomized, double-blind, phase 2 parallel study was registered to evaluate four creams for hand-foot skin reaction associated with sorafenib. It names the primary quality-of-life outcome, defines the Skindex-16 score range and direction, and reports an enrollment of 6. It also provides limited adverse-event counts by group. These registry entries describe the study’s scope and some of its posted information.

The record does not report the primary outcome values, a difference between treatment groups, confidence intervals, p-values, or listed statistical analyses. It also states that no subjects were randomized to the emollient arm and that missing 8-week questionnaires led to exclusions from the primary endpoint analysis. As a result, the posted information does not quantify change in quality of life or establish a treatment comparison. It cannot show which cream had a larger or smaller effect.

The safety counts should likewise be interpreted narrowly. Zero affected participants are reported in the listed risk sets, but the counts do not demonstrate that an adverse event is impossible or allow a reliable comparative assessment. In particular, the emollient arm has no participants listed at risk.

In general, a trial’s design and its observed results answer different questions. Randomization and masking describe intended methods for reducing bias; they do not substitute for reported outcome estimates. A study status of terminated is not itself an explanation of the results, and the record does not give a reason for termination. Overall, the registry supports a description of the trial and its reporting gaps, but not a conclusion about the relative effects or safety of the creams.

13. Why This Trial Matters Statistically

This study illustrates why a registered treatment comparison cannot be interpreted from its design label alone. Randomization and double masking are relevant design features, but the record also needs outcome estimates, uncertainty measures, and a clear account of which participants contributed to each analysis. For the primary endpoint, those results are not reported in the registry record.

The case also highlights the distinction between an outcome being named and an outcome being evaluable as a comparison. The study lists a four-arm design, yet the record says no subjects were randomized to the emollient arm and describes exclusions due to missing 8-week questionnaires in other arms. Its primary outcome analysis is not a reported comparison across treatments. A treatment question needs observed outcome information from the groups being compared; listing interventions is not enough.

Patient-reported outcomes add an important dimension to clinical research because they capture participants’ assessment of quality of life. Here, the Skindex-16 offers a defined score range and direction, but no result is posted. That absence prevents readers from judging either the magnitude of change or its precision, and it demonstrates why transparent reporting should include estimates and uncertainty as well as endpoint definitions.

Finally, the adverse-event counts show how zero counts must be read in context. The reported risk sets are limited, and the emollient group has no participants at risk. A zero count in a small or empty risk set is not equivalent to evidence of no risk. The statistical importance of this trial lies less in a comparative conclusion than in the need to distinguish the registered question from the information actually 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 registry record reports enrollment of 6, but does not provide a complete account of participant flow by arm, the number contributing to each endpoint, or detailed participant characteristics. It notes that some 8-week questionnaires were not completed and that results from specified arms or participants were excluded, but does not explain the reasons for missing questionnaires or describe a method for addressing missing data. The primary outcome values themselves are not reported.

The emollient arm presents a particular limitation for comparison: the record says no subjects were randomized to it, and its adverse-event risk set is 0. The record therefore does not provide observed treatment or safety information for that arm. For the other arms, the primary endpoint description also indicates incomplete outcome information, and no arm-level result or comparison is posted.

Important design and analysis details are not reported in the registry record, including eligibility criteria, treatment application instructions, the randomization procedure, masking arrangements, a statistical analysis plan, and the rationale for the planned enrollment. The record lists no analyses and does not report a reason for termination. It also does not report a multiplicity strategy, a clinically meaningful change threshold for Skindex-16, confidence intervals, or p-values. These gaps prevent evaluation of the intended analysis, the precision of any effect, and the context for the termination status.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Sorafenib-induced Hand- Foot Skin Reaction Treatment — NCT00667589.
• Registry record last updated 2013-06-05; status last verified 2013-05.
• Results first posted 2013-05-27.
• 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

NCT00667589 is a terminated phase 2 randomized, double-blind, parallel study sponsored by Northwestern University, with enrollment reported as 6. It evaluated urea 40% cream, fluocinonide 0.05% cream, tazarotene 0.1% cream, and a bland emollient for hand-foot skin reaction associated with sorafenib. The primary endpoint was change in Skindex-16 total score from baseline to 8 weeks; change from baseline to 2 weeks was listed as a secondary outcome.

The registry says no subjects were randomized to the emollient arm and notes missing 8-week questionnaires affecting the primary analysis. It lists no analyses, and does not report primary outcome values or comparative estimates. Adverse-event counts are reported for the listed risk sets, with 0 affected by serious and other adverse events in each group; the emollient group has 0 listed at risk. A reason for termination, detailed analysis methods, and a sample-size justification are not reported in the registry record.