A Clinical Trial of COX and EGFR Inhibition in Familial Polyposis Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Clinical Trial of COX and EGFR Inhibition in Familial Polyposis Patients (NCT01187901), a randomized phase 2 study. The primary outcome is Change in Duodenal Polyp Burden From Baseline to 6 Months.

2. Clinical Question

This is a question about a measured anatomical outcome over a stated treatment period. A negative change indicates that the measured burden at 6 months was lower than at baseline; a positive change indicates that it was higher. The registry reports a median change for each arm and a sponsor-reported comparison using a Wilcoxon (Mann-Whitney) method. The result therefore concerns the distributions of these measured changes, not a separate clinical outcome or a longer-term effect.

3. Trial Design

The study was a randomized, placebo-controlled, parallel-group phase II trial. It was listed as triple-masked. The registry records an enrollment of 92 participants, with 46 assigned to the sulindac-erlotinib arm and 46 to placebo in the posted primary outcome results. The sponsor was the University of Utah. The recorded start was 2010-04, the primary completion was 2014-06, and the trial status is completed.

The active arm received erlotinib 75 mg per day together with sulindac 150 mg twice daily for 6 months. The placebo arm received capsules matching each active drug, using the corresponding once-daily and twice-daily schedules for 6 months. The registry describes the study as randomized and triple-masked, but the procedures used to generate the random sequence, conceal assignment, maintain masking, and assess whether masking was successful are not reported in the registry record.

A parallel-group design assigns participants to one study arm and compares outcomes across the arms. In this trial, the primary comparison is between the combination regimen and placebo, not between erlotinib and sulindac separately. Because both active drugs were given together, the result cannot isolate the contribution of either drug from the other or establish whether their effects interact.

The registry lists six outcome measures and six analyses, including the primary duodenal polyp-burden outcome and secondary outcomes describing polyp burden or polyp number, with some outcomes specified separately for classic FAP and AFAP participants. Five secondary outcomes are listed as posted. The record does not report a full protocol-level description of the allocation procedure, the planned analysis populations, or the handling of deviations from assigned treatment.

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:

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 Duodenal Polyp Burden From Baseline to 6 Months, measured over Baseline and 6 months.

The primary endpoint is change in duodenal polyp burden from baseline to 6 months. Burden is the sum of the diameters of polyps found in a 10-centimeter segment of the duodenum, expressed in millimeters. Change is defined as the 6-month measurement minus the baseline measurement. With that definition, a negative value represents a reduction in the measured burden, while a positive value represents an increase.

The endpoint turns observations of multiple polyps into one continuous measurement for each participant. This can capture changes in the overall size of the polyp burden, rather than simply counting polyps. A count and a summed-diameter measure describe different features: a count records how many polyps are present, while the burden measure also reflects their diameters. The registry lists changes in polyp number as secondary outcomes.

The endpoint is specific to the duodenal segment and the 6-month observation period. It does not, by itself, describe colorectal polyp burden, progression beyond that period, or other health outcomes. The primary outcome is summarized using a median and an interquartile range. These summaries emphasize the center and spread of observed changes without assuming that the outcome follows a symmetric distribution. The registry record does not report further details about how polyp measurements were obtained or how measurement variation was addressed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous change score. For each participant, the 6-month polyp burden is compared with baseline by subtracting baseline burden from the 6-month measurement. The resulting value can be below or above zero. A value below zero means the measured burden decreased over the observation period; a value above zero means it increased. The registry reports each arm’s median change and interquartile range, rather than a mean and standard deviation.

Why use a median and interquartile range?

General statistical explanation: the median is the middle of an ordered set of observations. The interquartile range describes the interval between the lower and upper quartiles, containing the middle half of the observations. These summaries can be useful when outcomes are skewed or include values far from the center, because the median is less affected by extreme observations than the mean. Here, the registry identifies the median as the summary measure and gives the interquartile limits for each arm. It does not report the full distribution of individual changes.

What does the Wilcoxon (Mann-Whitney) test assess?

The sponsor reports a Wilcoxon (Mann-Whitney) comparison and p < 0.001. In general, this rank-based method compares the relative ordering of observations in two independent groups. It does not require the observations to follow a normal distribution. Its result is often described as a comparison of distributions; interpreting it specifically as a test of medians requires additional assumptions about the shapes of the group distributions. Those distributional details are not reported in the registry record.

A p-value describes how compatible the observed data are with a statistical null model for the test, under its assumptions. It is not the probability that the null hypothesis is true, and it does not quantify the size of the outcome difference. The reported p-value alone does not show the range of comparative effects compatible with the data. A confidence interval for an effect estimate would provide information about that uncertainty, but a between-arm confidence interval is not reported for the primary outcome.

Why does over-dispersion matter?

General statistical explanation: over-dispersion occurs when the observed variability is greater than a statistical model assumes. It can make standard errors too small and statistical evidence appear stronger than it should if the model does not account for the extra variation. This issue is particularly relevant to count outcomes, where observations may vary more than a simple count model expects. The primary outcome in this trial is a sum of polyp diameters, not a count, and the registry reports a rank-based test. The record does not report an over-dispersion assessment or a model-based analysis for this endpoint. The secondary polyp-number outcomes are count measures, but the registry entries given do not specify their modeling assumptions.

The reported medians and interquartile ranges are descriptive summaries. They show the center and middle spread within each arm, but they are not a direct estimate of the between-arm difference. A comparative estimate and its confidence interval would answer a different question: how large the arm difference may be and how uncertain that estimate is. No such recalculated estimate is available in the posted primary results. The recalculation cannot be made from these summaries alone using the simple unadjusted methods used for posted summary statistics.

How should sample-size illustrations be understood?

General statistical explanation: a sample-size calculation uses assumptions about the target effect, variability, significance level, allocation, and statistical method to estimate how many participants would be needed for a planned study. An illustration based on an observed effect is not the same as the trial’s original design calculation. It is not observed power, and it cannot establish that the trial was adequately designed or that a result is reliable. The registry record does not report the trial’s own sample-size justification. Any planning illustration based on the observed effect should therefore be read as conditional on its stated assumptions, not as evidence about the original planning process.

Likewise, a nominal p-value does not account automatically for testing multiple outcomes or multiple analyses. Whether and how multiplicity was handled must be determined from a prespecified analysis plan. Such an adjustment is not reported in the registry record. The record also does not report whether there were interim analyses, stopping rules, subgroup-selection rules, or prespecified sensitivity analyses. These details matter because they can affect how a reported test should be interpreted, but they cannot be inferred from the reported outcome values.

10. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary comparison as a Wilcoxon (Mann-Whitney) test with p < 0.001. The reported medians and interquartile ranges describe the observed changes separately in each arm: -8.5 mm in the sulindac-erlotinib arm and 8.0 mm in the placebo arm, with the respective interquartile ranges of -9.5 to -7.0 and 5.0 to 9.5. These summaries and the sponsor’s test are the available primary-outcome results.

No simple unadjusted recalculation of a between-arm estimate and confidence interval is available from the posted summary statistics. The registry record notes that values are missing or not estimable for a two-arm comparison for this other-endpoint calculation. Accordingly, there is no recalculated result to compare numerically with the sponsor’s analysis. The sponsor’s own analysis takes precedence when a recalculation is unavailable; in this case, the registry’s reported test and p-value are the comparative statistical result.

The p-value should not be mistaken for an effect estimate or a confidence interval. It gives neither a numerical between-arm difference nor a range of plausible effects. The posted arm summaries convey the directions and distributions of the observed changes, while the reported test gives the sponsor’s statistical comparison. Details of the test’s prespecified status, analysis population, and handling of missing observations are not reported in the registry record.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures and six analyses, including the primary duodenal polyp-burden outcome and five secondary outcomes. Those secondary measures include changes in duodenal polyp burden and polyp number, with some analyses specified for classic FAP and AFAP participants. When several outcomes or subgroup comparisons are examined, the chance of at least one apparently notable result can increase. General statistical practice may address this through a prespecified testing sequence or adjustment, but whether either approach was used is not reported in the registry record.

The record does not report a multiplicity plan, the order in which outcomes were tested, or whether the primary p-value was adjusted for other analyses. Therefore, p < 0.001 should be read as the reported result for the stated Wilcoxon (Mann-Whitney) comparison, not as a multiplicity-adjusted result unless such an adjustment is documented elsewhere in the registry record. The listed number of outcomes alone does not establish how the sponsor handled multiple testing.

Details about missing data are also not reported in the registry record. The posted primary outcome gives 46 participants in each arm, but the record does not specify whether those counts represent all randomized participants, all participants with observed outcomes, or another analysis population. It does not describe how missing baseline or 6-month measurements were handled, whether any measurements were excluded, or whether sensitivity analyses assessed the effect of missingness.

The analysis population and missing-data approach can affect the interpretation of a randomized comparison. For example, a comparison retaining participants according to randomized assignment answers a different operational question from an analysis restricted to participants with complete measurements. This is a general distinction, not a claim about which approach was used in this trial. The registry does not provide enough detail to identify the approach or assess its consequences.

Interim analyses, stopping criteria, protocol deviations, and prespecified subgroup tests are also not reported in the registry record. Their absence from the record does not show that they did or did not occur. It means that the reported result cannot be assessed against those aspects of a complete statistical analysis plan from the registry information alone.

12. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent non-serious adverse event terms

The posted adverse-event counts show zero serious adverse events among 46 participants at risk in the sulindac-erlotinib arm and zero among 46 participants at risk in the placebo arm. The registry also reports other adverse events in 43 of 46 participants in the active arm and 33 of 46 in the placebo arm. These are counts of participants reported as affected, not a description of event types, severity beyond the serious/non-serious distinction, timing, duration, or relationship to study treatment.

No serious adverse-event terms are listed. Death counts are not reported in the registry record. The record does not provide enough detail to assess whether event collection or follow-up differed between the groups, whether a participant could contribute more than one event, or how adverse events were defined and coded. Zero recorded serious adverse events should therefore be read as the posted count during the reported safety assessment, not as proof that serious events could not occur.

These safety counts are descriptive. They do not, by themselves, quantify comparative risk with uncertainty or establish causality. They should not be treated as a complete safety profile, particularly because the registry does not list the specific non-serious events or provide further detail on safety monitoring and analysis.

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

The primary result describes a difference in the observed pattern of duodenal polyp-burden changes over 6 months between the randomized combination and placebo arms. The active-arm median change was negative and the placebo-arm median change was positive, under the registry’s definition of change as 6-month burden minus baseline burden. The sponsor reports a Wilcoxon (Mann-Whitney) comparison with p < 0.001. These findings characterize the posted outcome data and the statistical test reported for them.

The result does not establish the size of a between-arm effect with a confidence interval, because such an estimate is not reported for the primary comparison. It does not show that every participant experienced the median change, nor does it describe each participant’s response. It also does not establish how durable any observed change was beyond the 6-month measurement.

The intervention combined two drugs, so this comparison cannot identify the separate contribution of sulindac or erlotinib. It does not test whether either drug alone would produce the same outcome or whether the drugs interact. The outcome is a measurement of duodenal polyp burden, not a direct measure of colorectal polyp burden, symptoms, quality of life, or other outcomes not listed as the primary endpoint.

The p-value is not a measure of clinical importance, and the arm summaries do not provide all the information needed to judge precision. Interpretation is also constrained by the absence of registry entries about the analysis population, missing outcomes, multiplicity, and the trial’s original sample-size justification. These reporting limits do not negate the posted observations, but they narrow what can be concluded from them.

The result is best understood as a randomized comparison of a specific combination regimen with placebo for a defined anatomical measure and observation period. It is not, by itself, a basis for conclusions about outcomes beyond those measured or about the independent effects of the individual drugs.

14. Why This Trial Matters Statistically

This trial illustrates how endpoint definition shapes statistical interpretation. The primary outcome is not simply a count of polyps: it is the sum of their diameters within a defined duodenal segment. A change score then compares that burden at baseline and 6 months. The sign of the change has a clear interpretation because the registry specifies the subtraction order, but the measure still captures only one aspect of polyp burden.

The trial also shows why descriptive summaries and comparative tests should be read together but not treated as interchangeable. The arm medians and interquartile ranges describe observed changes within each group. The Wilcoxon (Mann-Whitney) test supplies the sponsor’s reported comparison between groups. Neither the p-value nor the separate arm summaries provide a between-arm effect estimate with an associated confidence interval in the posted results.

Use of a combination regimen creates another important interpretive limit. Randomization supports comparison of the assigned combination with placebo, but it does not separate the effects of the two active components. That distinction matters when a trial’s question concerns a combined treatment while readers may also want to understand the components individually.

Finally, the record demonstrates the value of reporting analysis details alongside results. The test and p-value are stated, but the analysis population, missing-data handling, multiplicity plan, and original sample-size justification are not reported in the registry record. Without those details, readers can describe the primary comparison as posted while recognizing that some aspects of its statistical context remain unclear.

15. Limitations

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

The registry record gives the primary outcome summaries and the sponsor-reported test, but it does not report a confidence interval for a between-arm effect. This limits direct assessment of the precision and range of plausible comparative effects. A p-value alone cannot supply that information.

The record does not describe the primary analysis population, missing-data handling, or whether all randomized participants were included in the analysis. Although the arm counts are 46 and 46, the record does not clarify their relationship to enrollment and outcome completion. The effects of missing or excluded measurements therefore cannot be assessed from the registry information.

The approach to multiple outcomes and subgroup analyses is not reported. The record lists six outcomes and six analyses, including secondary outcomes in classic FAP and AFAP participants, but does not indicate whether multiplicity was controlled or whether subgroup analyses were prespecified. The trial’s own sample-size justification and any interim analyses are also not reported in the registry record.

The primary outcome concerns duodenal polyp burden over 6 months. It does not provide evidence about longer-term changes or other outcomes beyond those measured. The active regimen combines sulindac and erlotinib, so the comparison does not estimate the effects of each drug separately. Further measurement details, including how the polyps were assessed and how measurement variability was handled, are not reported in the registry record.

Safety reporting is limited to counts of participants with serious and other adverse events. Serious-event terms are not listed, deaths are not reported in the registry record, and the specific other events are not described. These limitations constrain interpretation of both the efficacy analysis and the safety summaries.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Clinical Trial of COX and EGFR Inhibition in Familial Polyposis Patients — NCT01187901.
• Registry record last updated 2016-06-17; status last verified 2016-05.
• Results first posted 2016-06-17.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

This completed randomized, placebo-controlled, triple-masked, parallel-group phase II trial enrolled 92 participants. It compared erlotinib 75 mg per day plus sulindac 150 mg twice daily with matched placebo for 6 months. The primary endpoint was change in duodenal polyp burden, defined as the sum of polyp diameters in a 10-centimeter segment and measured in millimeters.

The posted median change was -8.5 mm in the sulindac-erlotinib arm and 8.0 mm in the placebo arm. The sponsor reported a Wilcoxon (Mann-Whitney) comparison with p < 0.001. The primary outcome record does not provide a between-arm confidence interval. Serious adverse events were reported for zero of 46 participants in each arm; other adverse events were reported for 43 of 46 and 33 of 46 participants, respectively.

The registry does not report the original sample-size justification, analysis population details, missing-data handling, multiplicity plan, or deaths. The findings describe the posted primary outcome over the trial’s 6-month period.