Ph II Study of Azacitidine in Myelofibrosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Ph II Study of Azacitidine in Myelofibrosis (NCT00569660), a non-randomized phase 2 study. The primary outcome is Number of Participants With Objective Clinical Response.

2. Clinical Question

The statistical question is whether the registry record reports an observed response count that can be interpreted for this treated group. It does not. Although the study is marked completed and one outcome measure is listed, the record reports no analyses and no response estimate. The clinical question is therefore clear at the level of the planned endpoint, but its answer cannot be established from the posted results. The absence of a reported response result is not evidence that participants did or did not respond.

3. Trial Design

This was a Phase II study of azacitidine in people with myelofibrosis, sponsored by M.D. Anderson Cancer Center. The registry describes the allocation as non-randomized, the masking as none, and the model as single group. Enrollment is listed as 34, and the study status is completed. The recorded study start is 2005-06, and primary completion is recorded as 2008-04.

All participants were assigned to the azacitidine group; no control arm is described. The regimen was 75 mg/m^2 given subcutaneously daily for 7 days every 4 weeks. Because the trial had no randomized comparator, it can describe what was recorded for the treated group but cannot isolate a treatment effect by comparing outcomes with a concurrent control group. Changes over time, background care, participant selection, and other influences cannot be separated from treatment using this design alone.

The primary outcome was the number of participants with objective clinical response. The response assessment timeframe was every 2 courses of 4-week therapy, described as each 8 weeks. The outcome description also mentions bone marrow aspiration and biopsy with cytogenetics every 2 to 4 courses. The registry record does not provide further detail on how the response categories were adjudicated, how many assessments were planned for each participant, or how long participants were followed.

The record identifies the response analysis population as all enrolled patients who received at least one dose of study medication. It does not report participant flow, the number who received treatment, withdrawals, or the number included in any outcome analysis. Although completion and enrollment are recorded, these details do not substitute for an outcome analysis. The registry record does not report a protocol-based statistical plan, a sample-size justification, or the rationale for the selected design.

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: The maximum accrual of 34 was met.

Overall Study

Analysis population for the primary outcome, as defined in the registry: The statistical analysis for response rates determined on the intent-to-treat (ITT) populations. This is defined as all enrolled patients who received at least one dose of study medication.

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 Number of Participants With Objective Clinical Response, measured over Every 2 courses of 4 week therapy = each 8 weeks.

The primary endpoint is titled Number of Participants With Objective Clinical Response, with participants as its unit and a number as its measurement type. It is intended to count participants meeting a clinical response definition rather than to measure a continuous quantity such as a laboratory value. The endpoint description names complete response, partial response, hematologic improvement, and no response. Because that wording includes both response categories and no response, the precise rule for counting the primary outcome is not clear from the registry record.

Response was to be assessed every 2 courses of 4-week therapy, described as each 8 weeks. The record also mentions bone marrow aspiration and biopsy with cytogenetics every 2 to 4 courses. These assessments offer a structured way to describe clinical status over treatment, but the registry record does not say which assessment time was used for a single primary response count, how differing assessments were reconciled, or whether response had to be confirmed.

In general, a binary response endpoint can be useful because it expresses whether each participant met a prespecified criterion. It is interpretable only when the criterion, assessment timing, analysis population, and handling of missing assessments are sufficiently clear. Here, the registry specifies an intended response population and assessment schedule, but does not report the response count or the detailed operational criteria. The endpoint’s result therefore cannot be estimated from this record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Clinical trial results require both an outcome definition and an analysis that applies it to participant data. The registry record defines an intended primary response endpoint and its assessment schedule, but reports no analysis for that endpoint. The methods below explain what such results generally mean and what cannot be inferred when the underlying result is not reported.

What does a response count represent?

For a binary endpoint, each participant is classified according to whether a specified condition was met. A count records how many participants meet that condition in the analysis population. A response proportion relates that count to the population assessed. General interpretation depends on a clear response rule, a defined assessment time, and consistent handling of participants without evaluable assessments. The registry record gives a response population definition and assessment timing, but does not give the response count, a proportion, or the operational details needed to resolve all classification questions.

The outcome description lists complete response, partial response, hematologic improvement, and no response. Because no response is named alongside response categories, the exact composition of the counted outcome is ambiguous in the registry wording. It would be inappropriate to decide from that wording alone whether the intended count includes only response categories or another classification. The record also does not specify whether response needed confirmation or how assessments at different visits were combined.

How should uncertainty be assessed?

In general, an estimate such as a response proportion is subject to sampling uncertainty because results from a study group need not match the outcomes that would be seen in another group of similar participants. A confidence interval expresses a range of values compatible with the data under the statistical method used. Its width reflects both the amount of information and the variability in the observed data. A narrow interval indicates greater precision than a wide one, but neither establishes causation or removes design limitations.

No response estimate or confidence interval is reported in the registry record, so precision cannot be assessed for the primary endpoint. No statistical test or p-value for response is reported either. It is not appropriate to calculate or infer any of these quantities without the outcome data and a clear analysis definition. In particular, enrollment alone does not determine the observed response result.

What does a single-group design allow?

In general, a single-group study can describe outcomes observed among participants who receive the study intervention. A concurrent randomized control group helps distinguish outcomes associated with the intervention from outcomes that might otherwise occur. Without such a comparator, differences from an external benchmark can be affected by differences in eligibility, baseline health, assessment, follow-up, and care. The study is explicitly non-randomized and single group, so its design does not support a randomized between-group treatment comparison.

Masking is listed as none. In general, unmasked assessment can be more vulnerable to expectations influencing subjective judgments than assessment conducted without knowledge of treatment assignment. The registry record does not describe the response adjudication process or whether any independent review was used. This uncertainty is relevant to interpreting a clinical response endpoint, but it does not establish that assessment was biased.

What can planning calculations tell readers?

In general, sample-size planning asks how many participants would be needed to detect a chosen effect under specified assumptions. The registry record does not report the trial’s own sample-size justification or statistical design assumptions. Any planning illustration described with the results is separate from the trial’s original design: the stated illustration assumptions use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. The record does not provide an observed response effect or the illustration’s numerical results in the information available for interpretation.

Such planning figures are hypothetical design illustrations, not observed power. Observed power is not a substitute for an estimate and its confidence interval; it does not make an unreported response outcome interpretable. The stated two-arm equal-allocation assumption also differs from this study’s single-group design. A planning illustration therefore should not be mistaken for an analysis of the trial or for evidence about its response outcome.

How should adverse-event counts be read?

Safety summaries commonly distinguish the number of participants with at least one event from the number of event occurrences. The registry lists participants at risk and participants affected for serious and other adverse events, but does not clarify whether repeated events in one participant are counted once or how events were coded and attributed. A participant count should not be interpreted as an event count unless the record explicitly says so. The listed percentages for selected serious-event terms are also registry summaries; they do not establish cause and do not provide details about timing or severity.

These general statistical principles help define what a complete interpretation would require. For this study, the registry record does not report the primary response analysis, its precision, missing-outcome handling, or a detailed analysis plan. The appropriate conclusion is limited to the design and descriptive information actually recorded.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists one primary outcome measure and zero analyses. It does not report a statistical analysis plan, prespecified hypothesis tests, or a multiplicity adjustment strategy. In general, testing several outcomes or several versions of an outcome can increase the chance of a false-positive finding if the testing is not accounted for. A prespecified hierarchy or adjustment can address that concern, but the record does not say whether either was planned. Because no endpoint analyses are reported, multiplicity cannot be evaluated from results.

The response population is described as all enrolled patients who received at least one dose of study medication. This definition clarifies an intended analysis population, but the record does not state how many participants met it or how many had evaluable response data. It does not report the number of withdrawals, missed assessments, or participants excluded from an analysis.

Missing-data methods are not reported in the registry record. In general, the treatment of missing outcomes can affect a binary response estimate, particularly if missingness is related to health status or response. Possible approaches rely on assumptions that need to be stated and justified; no approach can be attributed to this study from the available record. The endpoint description also does not say how a participant without an assessment was classified.

The record does not report an intention-to-treat analysis in the broader sense of retaining every randomized participant, and the design is not randomized. It instead defines the response population as enrolled patients who received at least one dose. The distinction matters because the enrolled count alone does not establish the size or completeness of the outcome analysis. Without participant flow, missing-data rules, and endpoint results, readers cannot assess whether the observed analysis population differed from enrollment or how that might affect interpretation.

11. Safety Results

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

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: nausea (8.8%), vomiting (8.8%), pneumonia (8.8%), hyperuricemia (5.9%), neutropenic fever (5.9%).

For the Azacitidine group, the registry records 34 participants at risk for serious adverse events and 17 participants affected. For other adverse events, it records 34 at risk and 10 affected. These are participant counts in separate categories; the record does not explain whether a participant could appear in both categories. The counts should not be added together to infer a unique number of affected participants.

The selected serious-event terms and percentages are nausea, 8.8%; vomiting, 8.8%; pneumonia, 8.8%; hyperuricemia, 5.9%; neutropenic fever, 5.9%; and non-neutropenic fever, 5.9%. These terms are summaries, not a complete description of each participant’s experience. The record does not report event timing, severity grades, duration, recurrence, or whether investigators considered events related to treatment. It also does not report deaths in this adverse-event record.

In general, an adverse event occurring during a study is not automatically caused by the study intervention. The registry counts describe what was recorded in the treated group; they do not establish causality or allow a comparison of safety rates against a control group. The absence of additional detail should not be interpreted as evidence that other events did not occur.

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

The record establishes that a completed Phase II, non-randomized, single-group study evaluated azacitidine in myelofibrosis and identified objective clinical response as its primary outcome. It also records the treatment schedule, the intended response assessment timeframe, the response analysis population definition, and selected adverse-event counts. These registry entries describe the study and the kinds of outcomes it aimed to assess.

The record does not report the number of participants with objective clinical response or an estimate of response frequency. It therefore does not show how often response occurred, how precisely it was estimated, or how results varied among response categories. The absence of a response result is not a zero response and is not evidence of no response. It is an absence of a reported analysis.

Even if a response count were available, the single-group non-randomized design would limit causal interpretation. Outcomes in participants receiving azacitidine could reflect treatment, the course of myelofibrosis, other care, or participant characteristics. Without a concurrent comparator, this study alone cannot distinguish those explanations in a like-for-like comparison.

The adverse-event summaries document affected participant counts and selected serious-event terms, but do not establish that azacitidine caused them. The registry does not provide the detail needed to understand event timing, severity, attribution, or overlap among categories. These safety counts should therefore be interpreted descriptively and within the limits of the recorded information.

Overall, the registry record answers questions about the study’s structure and intended measurement more clearly than it answers the primary clinical outcome question. It gives no basis for a numerical conclusion about response or its precision. It also does not support a conclusion that the intervention does or does not work. The most defensible reading is that the outcome was planned and assessed on a schedule, but the primary analyzed result is not reported in the registry record.

13. Why This Trial Matters Statistically

This study illustrates why a registered endpoint and a reported endpoint result are not the same thing. The record identifies the outcome as the number of participants with objective clinical response and gives an assessment schedule. Yet it lists zero analyses and does not report the response count. Without that result, readers cannot estimate the outcome or judge its precision, regardless of the study’s completed status or recorded enrollment.

It also illustrates the difference between describing outcomes and estimating a treatment effect. A single-group study can provide descriptive information about participants receiving azacitidine, but a non-randomized design without a comparator cannot provide a within-trial contrast that separates treatment-associated outcomes from other influences. Statistical interpretation must respect that design rather than treating a response count as a causal estimate.

The definition of the analysis population is another important part of interpretation. The registry specifies that response rates were to be determined for enrolled patients who received at least one dose. However, participant flow, missing assessments, and the number analyzed are not reported. Those details matter because the population contributing outcome data may not be fully represented by the enrollment figure.

Finally, the response description includes several response categories as well as no response, while the exact rule for the primary count is not stated. Clear definitions and transparent handling of missing data are prerequisites for readers to understand what an endpoint estimate means. The record’s limitations make the proper statistical conclusion restrained: the design and intended endpoint are identifiable, but the primary result cannot be evaluated from the posted registry information.

14. Limitations

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

The most consequential limitation for interpreting the primary outcome is that the registry record does not report its analyzed result. There is no response count, estimate, confidence interval, or test result. The listing of one outcome measure does not indicate that its result is available, and completion status does not fill that gap.

The endpoint wording is not fully clear. It lists complete response, partial response, hematologic improvement, and no response in its description, without explaining the exact rule for counting participants with objective clinical response. The record does not clarify whether response needed confirmation, which assessment determined the endpoint, or how assessments across visits were combined. The scheduled assessment information does not resolve these questions.

The design is non-randomized and single group, with no masking. Accordingly, there is no concurrent control comparison, and the record does not describe an independent or blinded response review. These features limit causal interpretation and may leave clinical assessment open to expectations, although the record does not establish that bias occurred.

Participant flow and missing-data handling are also absent. The record defines the response population as enrolled patients who received at least one dose, but does not state how many met that definition, how many had evaluable assessments, or how withdrawals and missing assessments were handled. Baseline characteristics and other eligibility details are not reported in the available registry information, limiting assessment of how broadly the group represents people with myelofibrosis.

The safety information is limited to participant counts and selected serious-event terms. Deaths, event timing, severity, duration, attribution, and the handling of multiple events are not reported. The registry record also does not report the trial’s own sample-size justification, interim analyses, or detailed statistical plan. These absences constrain interpretation; they should not be filled by assumptions about what the protocol may have specified.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Ph II Study of Azacitidine in Myelofibrosis — NCT00569660.
• Registry record last updated 2012-08-07; status last verified 2012-08.
• Results first posted 2011-03-30.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

NCT00569660 is a completed Phase II, non-randomized, unmasked, single-group study of azacitidine in myelofibrosis, sponsored by M.D. Anderson Cancer Center. Enrollment is recorded as 34. The study start is 2005-06, and primary completion is 2008-04.

The regimen was 75 mg/m^2 subcutaneously daily for 7 days every 4 weeks. The primary endpoint was the number of participants with objective clinical response, assessed every 2 courses of 4-week therapy, described as each 8 weeks. The intended response population was enrolled patients who received at least one dose of study medication.

The registry lists one outcome measure and zero analyses. It does not report the primary response count or its precision. Safety entries record 17 participants with serious adverse events and 10 with other adverse events, with 34 at risk for each category; selected serious-event terms are listed with percentages. The design has no control arm, and the record does not provide the detail needed for a comparative or causal interpretation.