Azacitidine in Treating Patients With Myelofibrosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Azacitidine in Treating Patients With Myelofibrosis (NCT00381693), a registered phase 2 study. The primary outcome is Patients With Confirmed Response (Complete Remission or Partial Remission on 2 Consecutive Evaluation at Least 4 Weeks Apart) During the First 4 Months of Treatment.

2. Clinical Question

The study asked whether patients with myelofibrosis treated with azacitidine had a confirmed complete or partial response during the first 4 months of treatment. A response had to be documented on 2 consecutive evaluations at least 4 weeks apart. The record defines the response categories but does not report how many participants met the primary endpoint.

The question is clinically framed around whether the treatment period was associated with a confirmed change meeting specified disease-response criteria. Complete remission required resolution of disease-related symptoms, ultrasound-documented resolution of hepato-splenomegaly, normalization of peripheral blood findings and bone marrow histology, including disappearance of fibrosis and osteosclerosis. Residual cytogenetic abnormalities were allowed. Partial remission required a major response in any applicable baseline criterion other than constitutional symptoms, without progression in another category.

This is a question about response observed in the treated group, not a direct estimate of how azacitidine compares with another treatment or with no treatment. Without a reported response count and without a comparator, the registry record cannot establish the size or cause of any treatment-associated change.

3. Trial Design

This was a phase II study of azacitidine in myelofibrosis, sponsored by Mayo Clinic. The registry lists enrollment of 10 and an azacitidine intervention. The status is terminated. The recorded start was 2006-08, and primary completion was 2008-03.

Masking is listed as none. The record does not state the allocation method, and it does not report a randomized comparison. It lists an azacitidine group for adverse-event reporting but does not give a comparator group. Consequently, the design information does not support a like-for-like treatment comparison. A comparison between patients receiving azacitidine and patients receiving another intervention is not described.

The primary endpoint was a participant count: patients with confirmed complete or partial remission during the first 4 months of treatment. Confirmation required the response to be present at 2 consecutive evaluations at least 4 weeks apart. The record also lists overall survival, time to progression, and the number of participants with treatment-related adverse events as secondary outcomes. It reports 4 outcome measures, but does not provide results for those measures in the information available in the registry record.

The record does not explain why the study was terminated, how participants were recruited, how treatment was administered, or how long individual participants were followed. It also does not report a protocol-based sample-size justification or a statistical analysis plan. Those details should not be inferred from the phase designation or the enrollment figure.

With no comparator and no reported primary outcome result, the study design can describe the intended response assessment and the treatment group’s recorded safety counts, but it cannot by itself identify a comparative treatment effect. Termination is a status label; the reason for termination is 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.

6. Primary Endpoint

The registry lists the primary outcome as Patients With Confirmed Response (Complete Remission or Partial Remission on 2 Consecutive Evaluation at Least 4 Weeks Apart) During the First 4 Months of Treatment, measured over 4 months.

The primary endpoint counts participants who achieved a confirmed complete or partial remission during the first 4 months of treatment. It is a binary outcome at the participant level: each participant either meets the stated response definition or does not. The outcome unit is participants, and the registry identifies the parameter as a number rather than a continuous measurement.

Confirmation is part of the endpoint definition. A response must appear at 2 consecutive evaluations at least 4 weeks apart. In general, requiring confirmation can help distinguish a sustained finding from a transient observation, although the registry record does not explain the rationale for choosing this particular schedule.

The complete-remission definition combines symptom, imaging, blood, and bone-marrow criteria. The partial-remission definition requires a major response in an applicable baseline criterion, excluding constitutional symptoms, and no progression in another category. These rules specify what qualifies as a response; they do not state how many participants qualified.

A participant count is straightforward to interpret as the number meeting a prespecified definition, but its meaning depends on complete and consistently assessed outcome data. The registry record does not report a primary response count, a denominator for the primary endpoint, or an estimate of uncertainty. It therefore does not establish the observed response frequency or its precision.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

A primary endpoint translates the study question into an outcome that can be assessed consistently. In this study, the endpoint is a count of participants who satisfy the response definition during the first 4 months of treatment. The definition combines complete and partial remission, so both categories contribute to the stated endpoint. The record does not give the observed count or an analysis of that endpoint.

What does a participant count represent?

For a binary endpoint, each participant is classified according to whether the defined event occurred. Here, the event is a confirmed response. In general, binary outcomes are often described using a count and a proportion. A proportion requires both the count meeting the definition and a clearly specified denominator. The registry record gives the study enrollment but does not report the primary endpoint denominator or response count. It is therefore not possible to infer a response proportion from the record.

A count also does not, on its own, show how precisely the underlying response frequency has been estimated. In general, a confidence interval describes a range of values compatible with the observed data under the statistical method used. Its width reflects uncertainty, among other things. No confidence interval for the primary response outcome is reported in the registry record, and no interval can be reconstructed from the posted information without the outcome data and analysis details.

Why does the endpoint require confirmation?

The response must be present at 2 consecutive evaluations at least 4 weeks apart. That requirement is part of the endpoint, not an analysis result. In general, confirmation rules aim to make an outcome definition more reproducible and less dependent on a single assessment. They can also affect who qualifies as a responder: a participant must meet the criteria at the required evaluations. The record does not specify the assessment schedule beyond the confirmation rule, nor does it explain how unavailable or discordant evaluations were handled.

The complete- and partial-remission definitions use multiple clinical and laboratory features. A composite definition can capture different kinds of improvement under one endpoint, but it depends on clear rules for determining which criteria apply at baseline and how each component is assessed. The registry describes the criteria, but does not report assessment procedures, adjudication, or how missing components were treated.

What can be concluded from the analysis information?

The registry lists 4 outcome measures and 0 analyses. This describes what is recorded in the results information; it does not establish whether analyses were conducted outside the registry. The statistical model, hypothesis test, confidence-interval method, covariate adjustment, and any prespecified decision threshold are not reported in the registry record. Without these details and without outcome values, there is no reported statistical test to interpret.

In general, an unadjusted calculation from summary statistics is distinct from a sponsor’s own analysis, which may use participant-level data, prespecified methods, or additional adjustments. No recalculated result or sponsor analysis is reported in the registry record, so there is no numerical discrepancy to assess. A missing analysis should not be replaced by an assumed test or reconstructed result.

How should planning illustrations be understood?

The planning assumptions describe an illustration based on an observed effect: a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. For time-to-event outcomes, the stated convention is a total number of events; for other outcomes, it is participants per arm. These are planning conventions, not evidence of the study’s original design or its achieved statistical power.

No numerical planning results or observed primary effect are reported in the registry record. Consequently, the assumptions do not establish a protocol sample-size justification, and they cannot be used to infer observed power. In general, sample-size calculations estimate what may be needed under stated assumptions; they do not measure the probability that a completed study has found a true effect. A planning illustration is also not a substitute for the estimate and uncertainty from observed outcomes.

Why does over-dispersion matter in general?

Over-dispersion occurs when variability in data is greater than a statistical model assumes. In general, a model that underestimates variability can make uncertainty appear smaller than it is. This is relevant to some count-data models, but the primary endpoint is described as a participant count for a binary event, and the registry does not report a count model or an over-dispersion assessment. It would therefore be inappropriate to claim that over-dispersion affected this study’s analysis.

The statistical information is limited to the endpoint definition, outcome-measure listings, and planning assumptions. The response count, denominator, uncertainty estimate, statistical test, and missing-data procedures are not reported in the registry record. These omissions prevent an assessment of the primary endpoint’s magnitude, precision, and robustness.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Azacitidine")
n      <- c(10)
events <- c(0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Azacitidine  10  0  0.0  0.0  30.8  0.0  27.8  0.2  28.5

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 4 outcome measures, including the primary response measure and secondary outcomes for overall survival, time to progression, and treatment-related adverse events. It lists 0 analyses. No statistical analysis plan, hierarchy among outcomes, multiplicity adjustment, or strategy for choosing which results to emphasize is reported in the registry record. It is therefore not possible to determine whether or how the possibility of multiple statistical tests was addressed.

The registry record does not report the analysis population for the primary endpoint. It does not specify whether the endpoint was evaluated among all enrolled participants, those who began treatment, or another defined group. These populations can differ, and the choice can affect a response summary. The record also does not state how participants with incomplete assessments were handled, whether missing outcomes were excluded or imputed, or whether sensitivity analyses were conducted.

Missing information in a registry entry is not evidence that the corresponding observations were missing in the study. It means the record does not describe their availability or handling. Since neither the response count nor its denominator is posted, the extent and impact of missing primary-endpoint data cannot be assessed. The same caution applies to the secondary outcomes: their measure names are listed, but outcome results and analysis details are not reported in the registry record.

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: platelet count decreased (20.0%), anemia (10.0%), neutrophil count decreased (10.0%), hematoma (10.0%).

The safety information lists 10 participants at risk for serious adverse events and 4 participants affected. It separately lists 10 participants at risk for other adverse events and 10 affected. These counts describe the posted safety group; they do not indicate how many events each participant experienced. The registry record does not give a definition of the event-at-risk period or further details about collection and attribution.

The listed serious adverse-event terms are platelet count decreased at 20.0%, anemia at 10.0%, neutrophil count decreased at 10.0%, and hematoma at 10.0%. These terms can describe events that occur in the same participant, so the percentages should not be added or treated as mutually exclusive categories. The record does not state the severity grades, timing, duration, or individual clinical circumstances for these terms.

The summary identifies serious adverse events, but it does not report deaths for the group. A missing death count is not evidence that no deaths occurred; the record’s death value is not reported. These safety figures describe observed reporting in the listed group and do not establish a comparative risk because no comparator is reported.

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

The record establishes that a phase II study of azacitidine in myelofibrosis was registered, that the stated primary outcome concerned confirmed response during the first 4 months of treatment, and that the study status is terminated. It also records safety counts for an azacitidine group. These registry entries describe the study’s stated focus and some posted safety information.

The record does not report the primary response count, denominator, or estimate of uncertainty. It therefore does not show how many participants met the complete- or partial-remission definition, how precise any response estimate would be, or whether the response differed from a specified benchmark. Missing results cannot be interpreted as no responses or as evidence of an effect.

No comparator or randomized comparison is described, and allocation is not stated. Even if a response result were available, outcomes in a treated group alone would not establish what would have happened under another treatment or no treatment. A comparison is needed to separate a treatment-associated difference from other explanations for observed outcomes.

The serious and other adverse-event counts describe the registry’s safety reporting, not a comparison of risks. The listed serious-event terms do not provide detail on event timing, severity grades, or participant-level overlap. These limitations mean the safety information should be read as a descriptive record rather than a comparative estimate.

The termination status does not explain why the study ended, and the reason is not reported in the registry record. It should not be used to infer the study’s outcome. Overall, the posted record supports a narrow description of the intended endpoint and safety counts, but it does not support a conclusion about treatment benefit, lack of benefit, or comparative safety.

14. Why This Trial Matters Statistically

This study illustrates why registry outcomes need to be read together with their definitions and the results actually reported. A response label alone is not enough: the complete- and partial-remission criteria determine who qualifies, and the confirmation rule requires the response at 2 consecutive evaluations at least 4 weeks apart. Those details clarify the endpoint, but they cannot replace the missing response count and denominator.

The record also illustrates the difference between a study’s stated objective and an analyzable result. The primary question is identifiable, and several secondary outcomes are named, but 0 analyses are listed and the primary result is not reported. Without estimates and uncertainty, readers cannot assess the magnitude or precision of the response outcome.

The design information matters to interpretation as well. No comparator is reported, and allocation is not stated. Thus, the record does not describe a basis for estimating a treatment contrast. A descriptive outcome in an azacitidine group, even if posted, would answer a different question from a randomized comparison.

Finally, planning assumptions should not be confused with evidence from completed analyses. The stated planning framework uses an observed effect, a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Those conventions do not establish the trial’s original sample-size rationale or observed power, and no numerical planning results are reported in the registry record.

These distinctions are useful in clinical research interpretation: an endpoint definition, a design description, a safety count, and an effect estimate are different kinds of evidence. The registry record contains some of the former information but not the primary outcome estimate needed to quantify response.

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 main trial-specific limitation is that the primary response result is not reported in the registry record. The endpoint count, denominator, estimate of uncertainty, and analysis population are not given. As a result, neither the response frequency nor its precision can be evaluated.

The record does not describe a comparator or randomized comparison, and allocation is not stated. It therefore does not provide comparative evidence about azacitidine versus another treatment or no treatment. The reported enrollment of 10 cannot clarify the primary endpoint denominator or how many participants completed the required assessments.

The study is marked terminated, but the reason for termination is not reported in the registry record. Eligibility criteria beyond the listed conditions, treatment details, follow-up procedures, and the assessment schedule are also not reported. Their absence from the record limits evaluation of applicability and conduct; it does not establish what occurred in the study.

The response definition requires repeated evaluation and multiple disease-related criteria, but the record does not state how missing or inconsistent assessments were handled. It also does not report the statistical analysis plan, multiplicity strategy, or outcome-specific analyses. The safety information gives group counts and selected serious-event terms but does not provide event-level detail, grades, timing, or a comparator. These limitations preclude a full assessment of efficacy, uncertainty, and comparative safety.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Azacitidine in Treating Patients With Myelofibrosis — NCT00381693.
• Registry record last updated 2011-04-21; status last verified 2011-04.
• Results first posted 2011-01-24.
• 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

NCT00381693, titled “Azacitidine in Treating Patients With Myelofibrosis,” is a phase II study sponsored by Mayo Clinic. The registry lists enrollment of 10, no masking, and a status of terminated. The recorded start was 2006-08, with primary completion in 2008-03.

The primary outcome is the number of participants with confirmed complete or partial remission during the first 4 months of treatment. Confirmation requires 2 consecutive evaluations at least 4 weeks apart. The registry lists overall survival, time to progression, and treatment-related adverse events as secondary outcomes.

The primary response result and statistical analyses are not reported in the registry record. Safety reporting lists 4 participants with serious adverse events among 10 at risk, and 10 with other adverse events among 10 at risk. No comparator is reported, and the reason for termination is not reported.