Rituximab for the Treatment of Refractory Adult and Juvenile Dermatomyositis (DM) and Adult Polymyositis (PM): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Rituximab for the Treatment of Refractory Adult and Juvenile Dermatomyositis (DM) and Adult Polymyositis (PM) (NCT00106184), a randomized phase 2 study. The primary outcome is Comparison Between the Time to Improvement Between the Two Groups of IIM (Idiopathic Inflammatory Myopathy) Patients.

2. Clinical Question

The clinical question is therefore about the timing of a defined improvement event during the treatment phase. It is not simply whether participants ever improved, nor does the primary comparison directly estimate a long-term contrast between rituximab and no rituximab. Because the delayed-start group also received rituximab, the early between-group contrast is specifically about when treatment began.

The record also lists response rates at Week 8 and improvement in manual muscle testing at Week 44 as secondary outcomes. These address different aspects or time points of the study question. They should be interpreted alongside, not substituted for, the primary time-to-improvement analysis.

3. Trial Design

The registry describes a Phase 2, randomized, triple-masked, parallel-group study. Its listed enrollment is 200, and its status is completed. The record identifies the University of Pittsburgh as sponsor. The recorded start is 2006-03 and primary completion is 2010-02.

There are six arm descriptions, representing the treatment-timing groups within adult polymyositis, adult dermatomyositis, and juvenile dermatomyositis. Across these descriptions, Treatment Group A received rituximab at Weeks 0 and 1 and placebo at Weeks 8 and 9; Treatment Group B received placebo at Weeks 0 and 1 and rituximab at Weeks 8 and 9. The posted primary analysis combines the two timing groups for idiopathic inflammatory myopathy, rather than reporting separate primary results for each condition.

Randomization can help make assigned groups comparable at the outset, while masking is intended to limit the influence of treatment knowledge on care, assessment, or reporting. The registry labels the study triple-masked, but which individuals or roles were masked is not reported in the registry record. Details of randomization procedures, allocation concealment, stratification, and adherence are also not reported in the registry record.

The delayed-start design means the groups differ in treatment timing during the initial comparison. It does not create a continuing untreated control group for the entire treatment phase. After the scheduled switch, both groups were assigned rituximab. This matters when interpreting results over time: later observations do not represent the same contrast as the initial comparison between earlier and delayed treatment.

The record lists an intention-to-treat population for the primary outcome. It does not report in the registry record how that population was operationalized, how missing observations were handled, or whether any additional analysis populations were used. Those details are important for understanding exactly which participants and observations contributed to the reported comparison.

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.

Before assignment: In an effort to exclude IBM (Inclusion Body Myositis) and other myositis mimics, the medical records and muscle biopsy results (if available) of adults with PM were reviewed by a 3-member Adjudication Committee before enrollment.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intention to Treat (ITT)

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: 3 subjects in Group A and 2 subjects in Group B were excluded from analysis due to withdrawing from the study. These subjects either did not receive a full dose of study drug at week 0 or no study drug at all.

6. Primary Endpoint

The registry lists the primary outcome as Comparison Between the Time to Improvement Between the Two Groups of IIM (Idiopathic Inflammatory Myopathy) Patients, measured over Week 44 of treatment phase.

The primary endpoint is the time to improvement between the two idiopathic inflammatory myopathy groups, reported as a median in weeks with a full range. The registry gives the timeframe as Week 44 of the treatment phase. The event is not defined by one measure alone: a participant had to meet the study’s definition of improvement at two consecutive visits.

That definition required 3 of the 6 core set measures to improve by at least 20%, while no more than 2 measures could worsen by 25% or more. Manual muscle testing could not be counted among the worsening measures. The core set covered muscle strength, physician and patient global disease activity, physical function, extramuscular activity, and elevated muscle enzymes.

A time-to-event endpoint uses both whether an event occurred and when it occurred. This can distinguish groups whose participants reach improvement at different times, even if a simple final count would not capture that timing. The median gives the time by which half of the observed event-time distribution has reached the event, subject to the analysis and censoring used. The full range describes the reported spread of times, but does not show how observations are distributed within that range.

The registry does not provide the participant-level event times, a confidence interval for the median comparison, or details about censoring. The endpoint therefore has a clinically structured definition, but the posted summaries alone do not show the precision of the estimated time difference in the way a confidence interval would.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a time-to-event outcome: the event is meeting the study’s definition of improvement, and the outcome includes when that event occurs. The registry reports group medians and full ranges, then gives a log-rank test for the comparison. The secondary binary comparisons are accompanied by posted event counts and recalculated unadjusted measures. These methods answer related but not identical questions.

What does a median time convey?

A median is the point at which half of the event-time distribution lies on either side, subject to how the analysis treats participants who have not yet had the event. It is a compact description of timing, not a complete description of the distribution. Two groups can have similar medians but differ elsewhere in their event-time patterns. Conversely, a difference between medians does not by itself establish that the overall distributions differ.

A full range reports the smallest and largest observed times in the summary. It does not reveal how many observations lie near either endpoint, whether observations cluster around the median, or how uncertainty around the median was estimated. The registry does not post confidence intervals for the primary medians or their comparison.

What does the log-rank test assess?

In general, the log-rank test compares event-time experience between groups over the observed follow-up, taking account of event timing and censoring. Censoring occurs when the event time is not observed within the available follow-up; it is not the same as observing that the event never occurs. The test does not, by itself, express the size of a difference. An effect estimate such as a hazard ratio, together with a confidence interval, would provide complementary information about magnitude and precision. Those primary estimates are not reported in the registry record.

The log-rank test relies on assumptions about how censoring relates to event risk and is most straightforward to interpret when the relative event patterns are reasonably compatible with the test’s structure. The registry does not report diagnostics or additional model details, so the extent to which those assumptions were assessed is not reported in the registry record. The test result should therefore be read as the sponsor’s reported comparison, not as a complete account of the event-time data.

How should the secondary risk difference be read?

For a binary outcome, the risk difference is the difference between the proportions meeting the outcome in the two groups. A positive estimate in the direction coded for the comparison means the recorded proportion was higher in the named comparison group. The recalculated estimate is unadjusted: it uses the posted summary counts without adjustment for other variables. Its confidence interval represents a range of values compatible with the calculation and its assumptions, rather than a guarantee that the true value lies within that particular interval.

The posted recalculation includes a confidence interval that spans zero, and its unadjusted p-value is 0.297. In general, a confidence interval spanning zero for a difference is consistent with no difference as well as with differences in either direction at the interval’s stated confidence level. A p-value is not the probability that the null hypothesis is true, and it does not measure clinical importance. The sponsor’s reported test is a separate analysis; its result and the recalculated summary should not be conflated.

Relative risks and odds ratios express comparisons on different scales from a risk difference. They can appear larger or smaller depending on the baseline event frequency and the measure used. None of these measures alone establishes causation beyond the randomized comparison, and none resolves limitations in endpoint definition, missingness, or analysis reporting.

Why use planning illustrations cautiously?

The planning illustrations use the observed effect as a starting point and show hypothetical sample sizes under stated assumptions. The record describes a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect set to a stated fraction of the observed effect. These are planning illustrations, not evidence that the completed study had a particular probability of detecting an effect. They are not observed power and should not be interpreted as a retrospective verdict on whether the trial was adequately powered.

Planning from an observed effect is especially uncertain because the estimate from one study can be imprecise. A smaller assumed effect leads to a different planning requirement, but no single illustration establishes the sample size needed for a future study without a justified target effect and design assumptions. The registry does not report the trial’s own sample-size justification in the record summarized for these results.

What is not recoverable from summary statistics?

Aggregate counts and summary times cannot reproduce an analysis that depends on individual event times, censoring patterns, or covariates. They also cannot establish how outcome assessments were adjudicated or how protocol deviations were handled. A recalculation from posted summaries is useful as a transparent check on a simple unadjusted comparison, but it is not a replacement for the sponsor’s full statistical analysis. For the primary time-to-event endpoint, the registry does not post the underlying information needed to reconstruct a separate event-time analysis.

10. Analyses Reported by the Sponsor

For the primary endpoint, the registry reports median times and a log-rank p-value of 0.74. No recalculated primary comparison is available from the posted medians and ranges alone. A time-to-event recalculation would require information such as event times and censoring that is not reported in the registry record. The sponsor’s own primary analysis is therefore the relevant inferential result in the posted record.

For the secondary comparisons, the registry reports sponsor analyses and also includes unadjusted recalculations based on summary counts. These are not necessarily the same statistical analysis: the recalculation uses a simple comparison of the posted counts, whereas the sponsor lists a chi-squared test for one secondary outcome and a log-rank test for the other. The registry does not explain the full analysis specifications or the reason for the difference between the reported p-values and the recalculated unadjusted result.

The recalculated risk difference and its confidence interval are descriptive checks based on the posted counts; they should not be presented as the sponsor’s result. Conversely, the sponsor’s reported p-values do not supply an effect estimate or confidence interval for the secondary comparisons. The record marks no discrepancy, but the statistical details are insufficient to treat the different analyses as interchangeable or to resolve their differences independently. The sponsor’s analysis should be interpreted as the analysis reported by the study, while the recalculation remains a distinct, unadjusted summary.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome and secondary outcomes, but a complete statistical plan is not reported in the registry record. In particular, the record does not state whether secondary comparisons were prespecified in a formal testing hierarchy, whether any multiplicity adjustment was planned, or how interpretation would change across multiple tests. Without that information, the secondary p-values should not be treated as if they were the sole confirmatory test.

Multiplicity matters because testing several outcomes can make an apparently small p-value more likely to occur by chance across the set of tests. A prespecified strategy can control how evidence is evaluated across outcomes. The absence of a reported strategy does not prove that none was used; it means the strategy is not reported in the registry record. The primary result remains the main inferential focus, while secondary results warrant more caution.

The registry identifies the primary analysis population as intention to treat. It does not report the handling of missing outcome measurements, withdrawals, protocol deviations, or participants who did not complete follow-up. For the time-to-improvement analysis, details of censoring and the number of events are not reported in the registry record. These omissions matter because the amount and pattern of missingness can influence both the event-time distribution and the estimated comparison.

The record also does not report whether missing data were imputed, whether sensitivity analyses were conducted, or whether the analysis population differed for any secondary outcome. It is therefore not possible to assess from the posted summaries whether conclusions were robust to alternative assumptions about unavailable observations. The intention-to-treat label is informative, but does not by itself describe how participants with incomplete data were handled.

Finally, the registry does not report the trial’s detailed statistical analysis plan, including any interim analyses or sample-size justification. These elements should not be inferred from the completed status or from the posted p-values. Their absence limits assessment of prespecification and analytical flexibility, without establishing that any particular procedure 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: surgical procedure (6.7%), pneumonia (3.8%), gastritis (2.9%), malignancy (2.9%), cellulitis (2.1%).

The registry reports serious adverse events affecting 24 participants in Treatment Group A and 41 in Treatment Group B. It also reports other adverse events affecting 45 participants in Treatment Group A and 55 in Treatment Group B. These are group-level counts; the posted record does not provide enough detail to determine event timing, duration, severity beyond the serious-event classification, or whether an individual experienced more than one event.

The serious-event terms listed most often include surgical procedure, pneumonia, gastritis, malignancy, cellulitis, and muscle weakness associated with disease flare. The percentages attached to these terms are reported in the registry, but the record does not clearly specify the denominator or whether categories can overlap. They should not be interpreted as mutually exclusive shares or used to infer a comparative risk without that context.

Deaths are not reported in the registry record. The group counts alone do not establish that one treatment sequence is safer than the other. A safety comparison would need clear definitions, exposure time, event-level information, and an analysis that accounts for the study design. These figures describe what the registry posted; they do not support medical advice or a definitive safety conclusion.

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

The central statistical finding is that the posted median times to improvement are close and the sponsor’s primary log-rank test does not identify a clear difference between the earlier- and delayed-rituximab groups. The results describe this study’s randomized comparison of treatment timing in refractory adult and juvenile myositis populations. They do not directly compare rituximab with no rituximab over the full treatment phase, because both groups were assigned rituximab at different times.

A non-significant test does not prove that the treatment sequences are equivalent. Equivalence or non-inferiority requires a design and analysis that define an acceptable margin and assess uncertainty against that margin. Such a margin and analysis are not reported in the registry record. The primary result also lacks a posted confidence interval for the between-group effect, so the range of plausible differences cannot be judged from the record’s primary summaries alone.

The result does not establish that no participant improved, nor does it imply that the groups had identical event-time distributions. It addresses the comparative timing endpoint as defined in the study. The secondary findings add context but are limited by differences in outcome definitions, unexplained matching counts, and the absence of a reported multiplicity plan.

Nor can these results establish the effect of treatment in populations or settings beyond those represented by the study’s eligibility and follow-up, details of which are not fully. The registry record does not provide all participant-level data or analysis specifications needed to assess subgroup effects, robustness to missingness, or clinical importance in detail.

The most defensible reading is therefore narrow: the posted primary analysis did not demonstrate a statistically persuasive difference in time to improvement between the two assigned treatment sequences. That finding is not evidence that rituximab is effective or ineffective relative to no treatment, and it is not a definitive comparison of safety. It is an estimate from a specific timing design, with important limits in the information reported.

14. Why This Trial Matters Statistically

This trial is statistically instructive because its comparator is a delayed-start sequence rather than a group assigned to placebo throughout. That design changes the estimand: the primary comparison asks whether earlier versus later rituximab assignment changes the timing of a defined improvement event. It cannot be read as a sustained contrast between treatment and no treatment.

The endpoint also illustrates why outcome definition matters. Improvement required coordinated changes across core set measures and confirmation at consecutive visits. Such a rule can make an endpoint more structured than a single measurement, but it also makes the event dependent on multiple components and assessment occasions. The registry’s summary medians and log-rank test do not reveal how each component contributed to the event-time pattern.

The result highlights a common distinction between statistical significance and precision. A non-significant p-value does not show that effects are absent, and a pair of similar medians does not quantify the uncertainty around their difference. Without a primary effect estimate and confidence interval, the size of differences compatible with the data is difficult to evaluate.

The posted secondary outcomes further illustrate why analyses must be read in context. Recalculated unadjusted comparisons can serve as checks on aggregate counts, but they are not interchangeable with the sponsor’s listed tests. Multiple endpoints and incomplete analysis-plan reporting make it especially important to separate the primary inference from supplementary findings.

Finally, the available summaries show the value and limits of registry reporting. They make the design, endpoint definition, group summaries, and selected analyses accessible, while leaving important details about missing data, censoring, multiplicity, and analysis specifications unreported. Statistical interpretation depends not only on the numerical result, but also on what contrast was randomized and how completely its analysis is described.

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.

A primary limitation is the absence of a posted effect estimate and confidence interval for the time-to-improvement comparison. Medians, ranges, and a log-rank p-value summarize the result, but do not show the precision or magnitude of a between-group effect. The underlying event-time data and censoring details are not reported in the registry record, preventing an independent reconstruction of the primary test.

The treatment sequence also constrains interpretation. Group B began with placebo but was assigned rituximab later, so the design does not provide a continuing no-rituximab comparison. The primary result concerns timing between the assigned sequences, not the effect of receiving rituximab versus never receiving it.

The study includes adult polymyositis, adult dermatomyositis, and juvenile dermatomyositis, while the posted primary result combines the groups. Condition-specific primary estimates are not reported in the registry record. The combined result therefore does not show whether patterns were similar across these diagnoses or whether the overall estimate was influenced by differences among them.

The record does not report key details about randomization implementation, masking roles, missing-data handling, censoring, protocol deviations, or the statistical analysis plan. It also does not report a multiplicity strategy or the trial’s sample-size justification. These omissions limit assessment of robustness and prespecification.

For secondary outcomes, the same counts appear under differently described endpoints, and the registry does not explain that reporting. The sponsor’s tests and unadjusted recalculations are not fully aligned in their reported p-values, with insufficient methodological detail to reconcile them. The safety summaries likewise lack event-level detail and a clear denominator for the listed serious-event terms. These are limitations of what is reported, not proof that the underlying study procedures were inadequate.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Rituximab for the Treatment of Refractory Adult and Juvenile Dermatomyositis (DM) and Adult Polymyositis (PM) — NCT00106184.
• Registry record last updated 2015-03-04; status last verified 2015-03.
• Results first posted 2015-03-04.
• 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

The registry describes a completed Phase 2, randomized, triple-masked, parallel-group study of rituximab timing in refractory adult polymyositis, adult dermatomyositis, and juvenile dermatomyositis. Group A received rituximab first, while Group B began with placebo and received rituximab later.

The primary endpoint was time to improvement, defined through changes in core set measures confirmed at consecutive visits. The posted median times were 20.2 weeks for Group A and 20.0 weeks for Group B; the sponsor’s log-rank comparison reported p = 0.74. The posted secondary analyses do not show a clear between-group separation, but their interpretation is limited by reporting and multiplicity uncertainties.

The registry provides adverse-event counts but not enough detail for a definitive comparative safety assessment. Primary effect precision, missing-data handling, censoring, and several statistical-plan details are not reported in the registry record. The findings should be understood as a comparison of treatment timing, not as a long-term comparison with no rituximab.