Low-Dose Decitabine in Myelodysplastic Syndrome Post Azacytidine Failure: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Low-Dose Decitabine in Myelodysplastic Syndrome Post Azacytidine Failure (NCT00113321), a non-randomized phase 2 study. The primary outcome is Overall Response.

2. Clinical Question

The clinical question concerns response to decitabine among patients whose MDS had failed azacytidine therapy. This is a question about outcomes in a selected treatment setting, not a randomized comparison of decitabine with another option. The registry’s brief summary describes the aim as studying whether decitabine can help control MDS in this setting; that statement is an aim, not a reported finding.

The endpoint description defines Complete Response as remission and No Complete Response as lack of remission. It says response was evaluated after completing one course, with bone marrow aspiration to document remission every 1-3 courses. The record describes blood tests at baseline and after completing 8-12 weeks of therapy. It does not report an observed count of participants in either response category, so the recorded clinical question cannot be answered from the posted primary-outcome result.

3. Trial Design

The registry lists this as a Phase II study with a non-randomized allocation, no masking, and a single-group model. The study evaluated decitabine in MDS after azacytidine failure. Chronic myelomonocytic leukemia is also listed among the conditions. The planned enrollment was 16, and the primary-outcome population is described as the treated population of 16 patients.

The decitabine regimen is described as 20 mg/m2 by vein (IV) over 1 hour daily x 5 days. The primary response assessment was scheduled after one course of therapy, described as 8-12 weeks. The endpoint description also says bone marrow aspiration was used to document remission every 1-3 courses. These are registry descriptions of the regimen and assessment plan; the record does not provide further details on treatment delivery or adherence.

The study started in 2005-03, had a primary completion date of 2008-11, and is listed as terminated. The registry states that it was terminated early due to low accrual and that no further analysis was done. No reason beyond low accrual is reported in the registry record. The registry does not report a protocol rationale for the planned enrollment, or further information about operational changes over the course of the study.

Because there was no randomized comparator and no masking, the design does not support a direct treatment contrast with a control group. A single-group response description can convey what was recorded among treated participants, but it cannot separate treatment-associated outcomes from other explanations through a randomized comparison. In this record, even a descriptive response estimate is unavailable: the primary endpoint definition is present, but a response count or analyzed response result 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.

Before assignment: Study was terminated due to low accrual.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Treated population (16 patients); No further analysis done since study terminated early due to low accrual.

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 Overall Response, measured over Blood tests baseline and after completing 8-12 weeks of therapy.

The primary endpoint is titled Overall Response and is measured in participants. Its stated categories are Complete Response, representing remission, and No Complete Response, representing lack of remission. The endpoint is binary in the sense that each participant is classified into one of the two described categories. The assessment was planned after completing one course of therapy, with the timeframe given as blood tests at baseline and after completing 8-12 weeks of therapy. The description additionally says bone marrow aspiration was used to document remission every 1-3 courses.

A remission endpoint can provide a clinically interpretable description of whether participants met the study’s stated response criterion after treatment. The use of a defined assessment time and an explicit response classification helps clarify what outcome the investigators intended to record. Those features do not, by themselves, establish how response was determined in practice or how missing assessments were handled; those details are not reported in the registry record.

The registry records the endpoint definition but not the number classified as Complete Response or No Complete Response, nor an effect estimate or uncertainty interval. It says no further analysis was done because the study terminated early due to low accrual. Therefore, the endpoint’s intended meaning can be explained, but the registry record does not establish the observed response frequency.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical interpretation begins with the distinction between an endpoint definition and an endpoint result. An endpoint definition specifies what investigators intended to measure and how the outcome categories are described. A result requires observed data, such as the number of participants meeting each category, and may also include an estimate of uncertainty. The registry provides the former for Overall Response, but not the latter. It states that no further analysis was done after early termination due to low accrual.

What does a single-group design permit?

In a single-group study, the observed outcomes describe the participants treated in the study. Without a concurrent comparison group, there is no randomized contrast that estimates how outcomes would differ under another treatment or under a control condition. This design feature limits causal interpretation independently of the size or direction of any observed response. Here, the registry does not provide a primary response count, so even a descriptive response frequency cannot be reconstructed from the record.

Randomization is useful in comparative trials because it aims to balance measured and unmeasured prognostic factors between treatment groups, on average. This study is listed as non-randomized and single-group, so no such balance across treatment arms applies. Masking is listed as none. The registry does not report additional procedures to reduce assessment bias, nor does it describe how response assessments were adjudicated.

How should a binary endpoint be read?

A binary endpoint assigns each assessed participant to one of two categories. In this study, those categories are Complete Response and No Complete Response, with remission and lack of remission as their stated meanings. If the category counts and an appropriate denominator were reported, a response proportion could describe the observed frequency in the analyzed population. No such primary endpoint counts are reported in the registry record. No calculation of a response proportion or confidence interval can therefore be made from the record.

In general, a confidence interval expresses the uncertainty around an estimate under the assumptions of the method used. It does not correct for a non-comparative design, missing outcome data, or bias in outcome assessment. A narrow interval would not establish a causal treatment effect, and an interval cannot be interpreted when the underlying endpoint estimate is not reported. No confidence interval for Overall Response is reported in the registry record.

Why does the analysis population matter?

An analysis population defines which participants contribute to an endpoint estimate. The registry calls the primary-outcome population the treated population and gives its size as 16 patients. It does not describe further analysis populations, participant-level inclusion rules, or how participants without a completed response assessment would be treated analytically. These details matter because the estimate can depend on whether the analysis includes all treated participants, only those assessed, or another defined group. The registry does not report such a distinction beyond its stated treated population.

The record says blood tests were planned at baseline and after completing 8-12 weeks of therapy, and that response evaluation followed one course, with bone marrow aspiration to document remission every 1-3 courses. It does not report assessment completion, missing measurements, or the specific decision rules used for participants whose assessment was unavailable. General statistical practice distinguishes missing outcomes from observed failures; treating them as equivalent without a stated rule can change the interpretation. No such missing-data rule is reported in the registry record.

What do the planning assumptions mean?

The listed sample-size assumptions describe planning illustrations, not the original study’s design rationale. They specify a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the target effect set as a stated fraction of the observed effect. These assumptions relate to hypothetical planning for a comparison and should not be mistaken for the design of this single-group trial. The registry does not report the study’s own sample-size justification or a planned comparative allocation.

General sample-size planning asks what amount of information might be needed to detect a chosen effect under specified assumptions. Such illustrations are conditional on the assumed effect and model; they do not show the probability that a completed study actually detected an effect. In particular, they are not observed power. No additional planning result or numeric sample-size illustration is reported in the registry record’s trial details.

What can be concluded from absent analyses?

The registry lists one outcome measure and zero analyses. It also states that no further analysis was done due to early termination and low accrual. This means the posted record does not support a statistical test or an effect estimate for the primary endpoint. The absence of an analysis is not evidence that the endpoint was met or not met; it is a lack of reported analyzed outcome information. A conclusion should preserve that distinction rather than fill the gap with an assumed result.

For safety, the registry reports participant counts and selected serious adverse event terms. Those figures have a different role from the primary response endpoint: they describe reported safety events, not a randomized efficacy comparison. General interpretation requires attention to the reporting period, event definitions, and whether the entries are participant counts or event counts. The registry record does not provide further detail on these points, and the death count is not populated. The listed safety figures should therefore be reported as recorded, without deriving unreported rates or attributing events to treatment.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and zero analyses. The primary measure is Overall Response. A multiplicity adjustment addresses the risk of false-positive findings when multiple statistical hypotheses are tested; the need and method depend on the planned analyses. Here, the record does not report multiple primary comparisons or any completed analysis. A multiplicity procedure, or a rationale for not using one, is not reported in the registry record.

The endpoint population is described as the treated population of 16 patients. The registry says no further analysis was done because the study terminated early due to low accrual. It does not report the number with an evaluable response assessment, the number with missing outcome data, or rules for including participants whose assessment was unavailable. The stated assessment plan includes blood tests at baseline and after completing 8-12 weeks of therapy, with response evaluation after one course and bone marrow aspiration to document remission every 1-3 courses. Completion of these assessments is not reported in the registry record.

Missing-data handling is important because an observed response proportion depends on who is included and how unavailable outcomes are treated. General statistical methods may use explicit assumptions or sensitivity analyses, but neither a missing-data approach nor a sensitivity analysis is reported. The registry also does not report a separate intention-to-treat analysis, a per-protocol analysis, or additional analysis-population definitions. The only stated population is the treated population; the available record does not establish how every treated participant contributed to an endpoint calculation.

Because the primary endpoint has no reported analysis or result, neither the consequences of missing outcomes nor multiplicity can be evaluated from the registry record. These absences should not be taken as evidence that there were no missing assessments or no analytic decisions; they indicate that the relevant details are not reported.

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: death (18.8%), febrile neutropenia (12.5%), dyspnea (6.2%), elevated cardiac triponin (6.2%), blood/bone marrow (other) - spleen (6.2%).

The registry reports serious adverse event information for the Decitabine group, with 16 at risk and 6 affected. For other adverse events, it reports 16 at risk and 13 affected. These entries describe participants affected, as recorded; they do not by themselves provide event timing, duration, recurrence, severity beyond the serious classification, or a causal assessment.

The listed top serious adverse event terms are Death at 18.8%, Febrile Neutropenia at 12.5%, Dyspnea at 6.2%, Elevated Cardiac Triponin at 6.2%, and Blood/Bone Marrow (Other) - Spleen at 6.2%. The registry’s separate deaths field is not populated. Accordingly, the listed Death term should not be used to infer a death count from this record. The registry does not report further detail that would reconcile the term listing with the missing deaths entry.

These safety figures are descriptive reports from a single-group study. They do not compare risks against another treatment, and they do not establish that decitabine caused any listed event. The record does not report additional context needed to interpret exposure duration or event ascertainment. The figures should therefore be read as registry-reported safety information, distinct from the unreported primary response analysis.

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

The registry record establishes that a Phase II, single-group study of decitabine after azacytidine failure was conducted and later terminated. It describes the intended primary endpoint, the treatment regimen, and selected safety information. It also states that no further analysis was done because of early termination due to low accrual. These registry entries describe the study and its reporting status; they do not answer whether participants achieved the stated response outcome.

Overall Response was defined using Complete Response, representing remission, and No Complete Response, representing lack of remission. The registry does not report how many participants met either category, an estimated response frequency, or an uncertainty interval. The primary outcome therefore cannot be quantified from the record. A planned endpoint is not an observed finding, and the study’s aim should not be read as evidence that the intended outcome occurred.

The design further limits what could be inferred even if a response count were available. A non-randomized single group has no concurrent randomized comparator. Its outcome description would not establish how the same clinical population would have fared under another treatment or no study treatment. The lack of masking is another design characteristic; the record does not report additional measures used to address possible assessment bias.

The reported serious and other adverse-event figures can be described as safety reporting, but they do not supply a comparative estimate of harm. Selected serious-event terms are listed, while the deaths field is unpopulated and additional context is not reported. No causal attribution should be inferred from the entries alone.

In short, the available registry information supports a description of the trial design, endpoint definition, and reported safety entries. It does not support a numerical conclusion about overall response, a treatment-effect estimate, or a comparison with another therapy. Missing analyses and missing outcome details are limits of what the record allows readers to interpret, not evidence for either a favorable or unfavorable response result.

13. Why This Trial Matters Statistically

This study illustrates why an endpoint definition and an endpoint analysis must be distinguished. The registry describes Overall Response in concrete clinical terms, but the record also states that no further analysis was done after early termination due to low accrual. Without category counts or an estimate, readers cannot assess the observed response frequency or its precision. Statistical interpretation must therefore begin by identifying what information is available, rather than treating a planned outcome as a result.

The single-group design is also instructive. A study can document outcomes among treated participants without establishing a comparative treatment effect. Randomization and a concurrent comparator address a different question from a descriptive response assessment. Because this study is listed as non-randomized and single-group, a hypothetical response result would still require cautious interpretation and would not be a direct randomized contrast.

The record also shows why analysis populations and missing outcome procedures matter. It identifies the treated population, but does not explain assessment completion or how unavailable response measurements would be handled. Such details affect the denominator and interpretation of a binary endpoint. Their absence from the registry record prevents readers from evaluating those choices.

Finally, safety listings require careful separation from efficacy results and causal claims. The registry reports affected-participant figures and selected serious-event terms, but does not provide a comparative safety analysis or a populated deaths field. Reporting the entries without deriving additional quantities preserves what the record actually supports. Statistically, the key lesson is that study design, endpoint definition, analysis completion, and outcome reporting jointly determine what conclusions can be drawn.

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.

A central limitation is that the registry reports no analyzed primary endpoint result. Although Overall Response and its categories are defined, the record does not give counts in either category, an effect estimate, or an uncertainty interval. It states that no further analysis was done because the study terminated early due to low accrual. Thus, the endpoint cannot be summarized numerically from the registry record.

The study is listed as non-randomized and single-group, with no masking. There is no concurrent control comparison. This means the design cannot provide a randomized estimate of treatment differences, and the record does not describe additional procedures for limiting assessment bias. The registry also does not report response adjudication details or further operational information about treatment and follow-up.

The primary-outcome population is described as the treated population of 16 patients, but the record does not state how many completed the planned response evaluation. It does not report missing-outcome counts, rules for handling unavailable assessments, or additional analysis populations. The response assessment timeframe and bone marrow assessment plan are described, but actual completion and timing are not reported.

The safety information is limited to affected-participant figures and selected serious-event terms. The deaths field is not populated, and further details on event ascertainment, timing, exposure, and attribution are not reported in the registry record. The safety entries should not be interpreted as a comparative risk estimate.

Finally, the stated planning assumptions are not a reported design justification for this single-group study. A rationale for the planned enrollment, interim analyses, and a detailed statistical plan are not reported in the registry record. These gaps limit assessment of both the intended analysis and the reasons the study ended.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Low-Dose Decitabine in Myelodysplastic Syndrome Post Azacytidine Failure — NCT00113321.
• Registry record last updated 2012-08-07; status last verified 2012-08.
• Results first posted 2011-02-21.
• 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

The registry record describes a Phase II, non-randomized, unmasked, single-group study of decitabine for MDS after azacytidine failure. Chronic myelomonocytic leukemia is also listed as a condition. The regimen is described as 20 mg/m2 by vein over 1 hour daily x 5 days. The study started in 2005-03, had a primary completion date of 2008-11, and is listed as terminated.

The primary endpoint was Overall Response, classified as Complete Response or No Complete Response after one course of therapy. The primary-outcome population is described as the treated population of 16 patients. The registry says the study terminated early due to low accrual and that no further analysis was done; it does not report the response result. Safety entries include 6 participants affected among 16 at risk for serious adverse events and 13 among 16 at risk for other adverse events. The deaths field is not populated.