Prevention of Pegfilgrastim-Induced Bone Pain (PIBP): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Prevention of Pegfilgrastim-Induced Bone Pain (PIBP) (NCT00602420), a randomized phase 3 study. The primary outcome is Area Under Curve (AUC) of Average Pain From Diary vs. Day (1-5), Calculated by the Trapezoidal Rule..

2. Clinical Question

The trial addressed whether naproxen could prevent or lessen bone pain associated with pegfilgrastim in this chemotherapy setting. It was randomized, triple masked, and used a parallel-group design. The registry record describes the study as a phase III clinical trial and records an enrollment of 510. The outcome analysis includes 232 participants in the naproxen group and 229 in the placebo group; the reasons those counts differ from the enrollment figure are not reported in the registry record.

The question is specific to this short-term diary-based pain outcome. It does not, by itself, answer broader questions about other pain measures, later outcomes, or other patient groups. Such additional outcomes or populations are not reported in the registry record.

3. Trial Design

This was a randomized, triple-masked, parallel-group phase III trial comparing naproxen with placebo. The registry identifies 510 enrolled participants and records the study as completed. Its recorded start was 2008-06 and its primary completion was 2012-01. The registry identifies Gary Morrow as sponsor.

Participants in the naproxen group received oral naproxen twice daily, starting on the day pegfilgrastim was administered, identified as day 2, 3, or 4, and continuing for 5-8 days. Participants in the placebo group received an oral placebo on the same schedule. The registry record does not report the naproxen dose, the chemotherapy regimens, the pegfilgrastim dose, or how the pegfilgrastim administration day was selected.

The primary outcome was the area under the curve of average diary pain from baseline through day 5. The diary asked participants to record the severity of pain over the last 24 hours on a scale from 0, no pain, to 10, pain as bad as you can imagine. The AUC was calculated by the trapezoidal rule and had a stated range of 0-40, with units of pain scale score multiplied by days.

The outcome results list 232 participants for naproxen and 229 for placebo. These are the analysis-group counts reported with the primary outcome and are not the same as the overall enrollment figure. The registry record does not report whether the outcome analysis followed an intention-to-treat approach, how analysis populations were defined, or why participant counts differ across the record. Additional eligibility, allocation implementation, masking procedures, adherence, and follow-up details 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

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.

Population: At least 145 evaluable patients in each treatment group, thus 290 evaluable patients in total, required for an 80% power analysis.

6. Primary Endpoint

The registry lists the primary outcome as Area Under Curve (AUC) of Average Pain From Diary vs. Day (1-5), Calculated by the Trapezoidal Rule., measured over From baseline through day 5.

The primary endpoint was the area under the curve (AUC) of average daily pain from baseline through day 5, calculated using the trapezoidal rule. Day 1 was the day pegfilgrastim was administered. Participants recorded pain severity for the last 24 hours each day on a scale from 0, no pain, to 10, pain as bad as you can imagine. The reported AUC range was 0-40, with units of pain scale score multiplied by days.

An AUC combines information about pain severity across the observation period. In general, an accumulated measure can reflect both the level of a repeated measurement and how that level changes over time. That makes it different from a single pain rating on one day: two participants with different day-to-day patterns may have different accumulated scores even if one daily rating is similar.

For this trial, the endpoint was intended to represent the severity and duration of bone pain during the early period after pegfilgrastim. Lower values correspond to less accumulated diary pain on the defined scale and timeframe. The registry reports a mean AUC for each group and gives the primary endpoint timeframe as baseline through day 5. It does not report a threshold for a clinically important difference or a separate interpretation rule for the score.

The endpoint captures only the diary measure and period described in the registry. It does not establish how pain affected functioning, whether symptoms continued beyond day 5, or how participants who did not complete diary entries were handled. Those details are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is continuous: it is a numerical AUC calculated from repeated diary pain scores. The registry reports the mean for each randomized group, together with a 95% confidence interval for each mean. A mean gives a measure of the group's average endpoint value. It does not show the full distribution, such as whether values were concentrated near the mean or widely spread. The registry does not report a standard deviation or another measure of within-group spread for these arm summaries.

A confidence interval communicates estimation uncertainty. In general, it gives a range generated by a method that, over repeated use under its assumptions, would cover the quantity of interest at the stated confidence level at the corresponding frequency. It is not a prediction interval for individual outcomes and does not mean that a particular participant's value is likely to fall inside it. The arm-specific intervals and the interval for the reported contrast address different quantities, so they should not be substituted for one another.

The registry's primary comparison is a two-sided t-test of the mean difference in final values. In general, a t-test assesses whether observed differences in group means are compatible with a null hypothesis of no mean difference, while accounting for variability and sample size under the test's assumptions. A two-sided test considers departures in either direction. The registry reports the contrast as 1.67, a 95% confidence interval from 0.10 to 3.24, and p = 0.037. The analysis description does not specify the subtraction order or report further details of the model. The arm means provide the context for the direction of the observed difference.

The interval for the contrast is useful because it shows both the estimate and its precision. A narrow interval would indicate more precise estimation than a wide interval, for the same kind of analysis and scale. Here, the confidence limits run from 0.10 to 3.24. Their interpretation depends on the contrast definition and test assumptions. The registry does not report an adjustment for baseline values or other covariates, so whether the analysis was adjusted is not reported in the registry record.

The registry also lists a Wilcoxon (Mann-Whitney) test, with p = 0.007. In general, this rank-based method compares the distributions of observations between groups using the ordering of values rather than directly testing a difference in means. It can be useful when a comparison based only on means is not the sole analytic interest or when distributional features make a rank-based comparison informative. It should not automatically be interpreted as a test of medians: that interpretation requires additional assumptions about the shapes of the group distributions. No rank-based effect estimate or confidence interval is reported in the registry record.

The presence of two methods gives readers more than one statistical perspective on the same outcome. It does not show that one method was primary and the other a sensitivity analysis, because that role is not reported in the registry record. Nor does the record state whether the analyses were prespecified, whether assumptions were assessed, or how the trial handled multiple testing. These matters affect how the results should be understood, particularly when more than one test is reported.

The AUC itself was calculated by the trapezoidal rule. In general, this rule approximates the area under a curve by joining successive observed values with straight-line segments and adding the resulting areas. For repeated pain scores, the resulting summary reflects scores across the defined time interval rather than a single day's reading. The registry gives the AUC range as 0-40 and identifies the endpoint timeframe as baseline through day 5. It does not give a separate rule for missing daily entries or explain whether interpolation or another procedure was used when a score was absent.

Over-dispersion is a general concern in statistical models that assume a particular level of variability: if real observations vary more than the model allows, uncertainty can be understated and inference may be misleading. The registry describes a t-test and a rank test, not a count model, so over-dispersion is not a reported feature of these analyses. More generally, distributional shape and variability can matter for continuous outcomes too; the arm-level spread statistics needed to assess those features are not reported in the registry record.

The reported p-values are not probabilities that the null hypothesis is true, nor do they measure the probability that the result will replicate. They summarize how unusual the test statistic would be under a specified null model and its assumptions. Statistical evidence and practical importance are separate questions. The registry does not report a clinically important difference threshold, so the clinical meaning of the numerical AUC contrast cannot be decided from the p-values alone.

The record contains an enrollment count and outcome-specific group counts, but does not state the analysis population, treatment of missing outcome data, or reasons participants might be absent from the endpoint analysis. It also does not report a sample-size justification or a detailed statistical analysis plan. Those features are necessary for a fuller assessment of precision planning, potential bias, and the intended role of each test.

10. Analyses Reported by the Sponsor

The registry reports the sponsor's two analyses as a two-sided t-test of the mean difference in final values and a Wilcoxon (Mann-Whitney) test. The t-test result is a mean difference of 1.67, with a 95% confidence interval from 0.10 to 3.24 and p = 0.037. The Wilcoxon test is reported with p = 0.007, without an effect estimate or confidence interval. These are the results to use when describing the registry's reported analysis.

The record does not give a separate recalculated estimate or any recalculated p-value, so a numerical comparison between the sponsor's analysis and an independent calculation is not reported in the registry record. No discrepancy is flagged in the registry entries. The arm means, 6.04 for naproxen and 7.71 for placebo, provide the group-level context for the direction of the reported difference, but the subtraction order for the contrast is not specified in the analysis description.

The reported t-test and rank-test results concern the same endpoint and groups. They are alternative analyses, not independent evidence from separate participant samples. The record does not state whether one was designated as primary or whether the other was intended as a sensitivity analysis.

11. Multiplicity, Missing Data and Analysis Populations

The registry reports one primary outcome measure and two analyses of that measure: a two-sided t-test and a Wilcoxon (Mann-Whitney) test. It reports no secondary outcome measures. The record does not explain whether both tests were planned in advance, whether one was considered the main analysis, or whether any adjustment for multiple testing was used. Without that information, the role of the second test and the intended interpretation of the pair cannot be determined.

Multiplicity is a general concern when several outcomes, comparisons, or analytical choices are tested. In general, the more opportunities there are to find a nominally unusual result, the more cautious interpretation may be warranted unless the testing strategy is specified. The registry record does not report a multiplicity plan, so it is not possible to establish how the reported p-values relate to any prespecified family of tests.

The outcome results list 232 participants in the naproxen group and 229 in the placebo group, while overall enrollment is recorded as 510. The registry does not report why these figures differ, whether every enrolled participant was randomized or treated, or which analysis population was used. It also does not report the number of missing diary observations, how missing daily scores were handled in calculating AUC, or whether missingness differed by group.

Missing outcome data can affect both precision and interpretation. In general, the impact depends on how much data are missing and why they are missing, as well as the analysis method used. The registry record does not report assumptions about missingness, imputation procedures, or sensitivity analyses. It therefore does not allow readers to evaluate the possible influence of missing diary data on the reported endpoint comparison.

The statistical analysis plan, sample-size justification, interim analyses, and analysis-population definitions are not reported in the registry record. This means the public record does not establish whether the reported analyses follow prespecified methods, how the trial handled deviations from treatment assignment, or whether any interim monitoring affected the analysis. These are reporting gaps, not evidence that a particular method was or was not used.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: death (1.9%).

The adverse-event figures list serious events in 5 of 257 participants at risk in the naproxen group and 3 of 253 at risk in the placebo group. Other, non-serious adverse events are listed for 4 of 257 at risk in the naproxen group and 3 of 253 at risk in the placebo group. The registry does not report death counts by group; the death-count fields are not populated.

The registry also lists Death as a top serious adverse event term at 1.9%. That term-level entry is not accompanied by group-specific counts in the record, and the relationship between the term listing and the group figures is not explained. It should therefore be read as a registry-reported term and percentage, not as a group-specific comparison.

These counts are descriptive. The record does not report an analysis comparing adverse-event rates between groups, the event definitions, the observation period for safety reporting, or whether participants could experience more than one event. The at-risk counts differ from the primary outcome analysis counts, and the reason is not reported. The figures do not by themselves establish a causal relationship between an intervention and an event, and they do not provide a complete account of safety outcomes.

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

The posted primary outcome results show a lower mean AUC in the naproxen group than in the placebo group over the diary period from baseline through day 5. The reported group means are 6.04 and 7.71, respectively. The registry's two-sided t-test reports a mean difference of 1.67, a 95% confidence interval from 0.10 to 3.24, and p = 0.037. The registry also reports a Wilcoxon (Mann-Whitney) p-value of 0.007. These figures describe the observed groups and the statistical tests recorded for this endpoint.

The result does not establish the size of benefit that an individual participant would experience. Group averages do not describe every participant's outcome, and the registry does not provide the full distribution of AUC values. It also does not report a threshold for a clinically important difference, so statistical evidence of a group difference cannot by itself establish how important the difference is in practical or clinical terms.

The result is limited to the specified diary measure and short observation period. It does not establish whether the groups differed after day 5, on other pain outcomes, or in broader aspects of health. Other outcome measures are not reported in the registry record. The trial population is described as patients with non-hematologic cancer undergoing chemotherapy, so the results should not be assumed to describe populations not represented by that description.

The record does not report the analysis population, missing-data handling, or reasons for the difference between overall enrollment and outcome-group counts. These omissions limit assessment of how missing or excluded data might affect interpretation. The results are therefore best understood as the reported comparison of the analyzed groups, with those reporting gaps kept in view.

The adverse-event counts are also descriptive, and the registry does not report group-specific death counts or a comparative safety analysis. They cannot be used alone to establish the overall balance of benefits and harms. Taken together, the record provides a primary pain outcome comparison and limited adverse-event counts, but does not answer every question about clinical importance, longer-term outcomes, or safety.

14. Why This Trial Matters Statistically

This trial illustrates how a clinical question can be translated into a quantitative endpoint. Rather than comparing a single daily pain score, it used an AUC that combines diary pain scores over the defined period. That choice makes the outcome a summary of pain across time, but it also makes its interpretation dependent on the measurement schedule, the scale, and how incomplete diaries are handled.

It also illustrates why effect estimates and uncertainty should be read together. The registry reports group means, a mean-difference estimate, a confidence interval, and p-values. The estimate conveys the scale and direction of the observed contrast; the interval conveys uncertainty around it; and the p-values summarize evidence under the specified tests. None of these quantities alone determines clinical importance.

The record includes both a mean-based comparison and a rank-based comparison. These methods address the same two groups and outcome from different statistical perspectives. Their presence makes it especially important to know which analysis was prespecified as primary and whether multiplicity was considered. Those details are not reported in the registry record.

Finally, the different counts for enrollment, primary outcome groups, and safety groups show why population definitions and denominators matter. Without an account of analysis populations and missing data, readers cannot fully trace who contributed to each result. The trial is therefore a useful example of both the information a registry can provide and the additional reporting needed for a complete statistical interpretation.

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 does not report the detailed eligibility criteria, participant characteristics, chemotherapy regimens, or naproxen dose. These details would help define the population and treatment conditions to which the results apply. The record identifies a non-hematologic cancer population undergoing chemotherapy but does not provide enough information to assess how closely the participants represent a narrower clinical subgroup.

The primary outcome group counts are 232 for naproxen and 229 for placebo, while enrollment is 510. The record does not explain the difference or define the analysis population. Missing diary data, reasons for missingness, and the method for handling missing daily scores are also not reported. These gaps make it difficult to assess whether the outcome estimates could be affected by incomplete data or exclusions.

The registry gives arm means and confidence intervals but does not report standard deviations or the distributions of individual AUC values. The t-test assumptions and any diagnostic checks are not described. The analysis description also does not specify the contrast coding or report covariate adjustment. These details limit independent evaluation of the modeling choices and the precision of the summaries.

Although two tests are reported, the record does not say which was prespecified as the main test, whether the rank-based analysis was intended as a sensitivity analysis, or whether multiplicity was addressed. A statistical analysis plan and sample-size justification are not reported in the registry record. Secondary outcome measures are not reported.

The safety entries provide counts but no group-specific death counts, event definitions, safety observation period, or comparative analysis. The term-level listing of Death at 1.9% is not explained in relation to the group figures. Finally, the endpoint covers pain from baseline through day 5; results after that period and a threshold for clinical importance are not reported.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Prevention of Pegfilgrastim-Induced Bone Pain (PIBP) — NCT00602420.
• Registry record last updated 2015-11-09; status last verified 2015-10.
• Results first posted 2015-05-12.
• 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 phase III trial randomized participants to naproxen or placebo in a triple-masked, parallel-group design. It enrolled 510 participants and examined prevention of pegfilgrastim-induced bone pain among patients with non-hematologic cancer undergoing chemotherapy.

The primary endpoint was the AUC of average diary pain from baseline through day 5, calculated by the trapezoidal rule. The naproxen group mean was 6.04 (95% confidence interval 4.94 to 7.14), and the placebo group mean was 7.71 (95% confidence interval 6.59 to 8.84). The reported two-sided t-test gave a mean difference of 1.67 (95% confidence interval 0.10 to 3.24; p = 0.037). A Wilcoxon (Mann-Whitney) analysis had p = 0.007.

Outcome analysis counts, missing-data handling, analysis-population definitions, multiplicity procedures, and sample-size justification are not reported in the registry record. Safety entries give serious and other adverse-event counts, but group-specific death counts are not reported.