Study of Decitabine for Treatment of Older Patients With Acute Myeloid Leukemia (AML): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Decitabine for Treatment of Older Patients With Acute Myeloid Leukemia (AML) (NCT00358644), a non-randomized phase 2 study. The primary outcome is Complete Response = Morphologic Complete Remission (mCR).

2. Clinical Question

In practical terms, the primary question is whether and how often participants in this enrolled group were recorded as reaching the specified morphologic remission endpoint during the stated timeframe. The registry description frames the study as front-line treatment for older patients with AML. It does not provide the response count or a response rate in the registry record recorded for the primary outcome, so the magnitude of the observed response cannot be described from this record.

The design matters to the question. A single-group study can describe outcomes among people who received the study intervention, but without a concurrent control group it cannot establish how those outcomes compare with another treatment under the same study conditions. The record also does not describe the eligibility details or the clinical criteria used to determine mCR. Those details are not reported in the registry record, and should not be inferred from the endpoint label alone.

3. Trial Design

This is a completed Phase II study listed as non-randomized, with no masking and a single-group model. The registry records 55 participants enrolled. The condition is acute myeloid leukemia, and the official title describes a study of decitabine for front-line treatment of older patients with AML. The sponsor is Eisai Inc. The recorded study start is 2005-03 and primary completion is 2008-02.

The intervention is decitabine. The safety group is named Decitabine 20 mg/m2 Intravenous, but the registry entries do not state the dosing schedule, treatment duration, or other administration rules. Such protocol details are not reported in the registry record. The arm description is blank, so additional treatment-arm details cannot be established from this record.

Because the study is single-group and non-randomized, there is no randomized control arm against which to compare outcomes. The registry explicitly flags that its arms are not a like-for-like control comparison. The study can therefore be read as an evaluation of outcomes and adverse events in the enrolled group, rather than as a randomized comparative estimate of the effect of decitabine relative to another intervention.

The primary endpoint is the number of ITT participants with complete response, defined as morphologic complete remission, within 1 year. Although the brief study summary also describes an aim to estimate overall survival, an overall-survival outcome measure is not reported in the registry record. The record lists one outcome measure and one analysis, but it does not describe the statistical analysis procedure, missing-data approach, or any interim analyses. These details are not reported in the registry record.

The registry status is completed. That status describes the study record, not the result of the primary endpoint. The registry entries do not give the number of participants achieving mCR or a corresponding response estimate. Consequently, the design and stated objective can be explained, but the primary outcome result cannot be reconstructed from the information recorded.

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:

Analysis population for the primary outcome, as defined in the registry: Intent-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.

6. Primary Endpoint

The registry lists the primary outcome as Complete Response = Morphologic Complete Remission (mCR), measured over 1 year.

The primary endpoint is recorded as “Complete Response = Morphologic Complete Remission (mCR).” It is a participant count, with the ITT population specified and a timeframe of 1 year. The endpoint therefore concerns how many participants in the stated analysis population were classified as having the defined response during that timeframe. The registry does not provide the count in the registry record recorded for this endpoint, so a response rate or estimate of its precision cannot be stated.

In general, a remission endpoint is used to describe whether a specified treatment outcome was observed, rather than to summarize a continuous measurement. A morphologic response label refers to a classification based on morphology, but the specific assessment procedures and criteria used in this study are not reported in the registry record. The endpoint name should not be taken as a substitute for those operational details.

The stated timeframe gives context for when the response is considered, but the record does not explain how assessments were scheduled, how response timing was handled, or how participants without an assessment were classified. Those details are not reported in the registry record. ITT identifies the named population for the primary outcome, but by itself does not explain how missing outcome information was treated.

The brief summary also names overall survival as an aim. However, the primary measure recorded is mCR, and an overall-survival measure or result is not reported in the registry record. These are distinct outcomes and should not be conflated when interpreting the study’s stated purpose and its posted primary endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a single-group design estimate?

In general, a single-group study describes outcomes observed among participants who received the study intervention. Its direct statistical target is a summary of outcomes within that group, such as a participant count or response proportion for a binary endpoint. It does not, by itself, estimate a randomized contrast between treatments. In this record, the study is listed as non-randomized and single-group, and the registry flags that its arms are not a like-for-like control comparison.

That distinction is important when interpreting causal claims. A change or outcome observed after treatment may be influenced by the underlying condition, participant characteristics, other care, or the way outcomes are assessed. A concurrent randomized comparison is designed to reduce some sources of systematic difference between groups. The registry entries recorded for this study do not describe such a comparison, so the primary outcome should be interpreted as an outcome in the enrolled group rather than a treatment effect against a control.

What does a binary endpoint represent?

In general, a binary endpoint classifies each participant into one of two outcome states for the specified analysis, such as having or not having a defined event. A participant count can be reported directly; a proportion expresses that count relative to the analysis population. The registry labels the primary outcome as a participant count and identifies complete response as morphologic complete remission. The response count itself is not reported in the registry record, so neither a proportion nor an uncertainty interval can be stated from the recorded details.

For a binary outcome, the denominator and the outcome definition are central to interpretation. The record names the ITT population and the 1 year timeframe. It does not state the specific response assessment procedures, how missing assessments were classified, or whether the outcome required a particular confirmation process. These methodological details are not reported in the registry record.

How should confidence intervals be understood?

In general, a confidence interval expresses uncertainty around an estimated quantity. It is calculated from the observed data using a specified statistical method. A narrower interval indicates greater precision under that method, while a wider interval indicates less precision. An interval does not eliminate design limitations, such as the absence of a concurrent control group, and it cannot be interpreted without knowing which estimate it accompanies and how it was calculated. No confidence interval or primary response estimate is reported in the registry record.

Why do analysis populations matter?

The registry identifies the ITT population for the primary endpoint. In general, an ITT analysis aims to retain participants in the groups to which they were assigned, though its application depends on the study design and the availability of outcome data. Here the study is single-group, and the registry entries do not specify how the ITT set was defined beyond naming it, how many participants had evaluable endpoint data, or how missing outcomes were handled. Those details are not reported in the registry record. The label alone cannot establish that every enrolled participant contributed an observed endpoint value.

What can a planning calculation tell readers?

In general, sample-size planning asks how many participants may be needed to detect a prespecified effect under stated assumptions. A planning illustration based on an observed effect is not the same as the study’s original design justification, and it is not observed power. The registry entries describe an illustration using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect set as a stated fraction of the observed effect. The assumptions for the study’s own design and its sample-size justification are not reported in the registry record.

The illustration’s equal-allocation, two-arm setup should not be mistaken for the actual study design: the registry lists this study as single-group and non-randomized. The registry entries also do not include a numerical observed effect for the primary endpoint. Thus no planning result can be independently interpreted as a trial result or as evidence of achieved power. Planning numbers describe hypothetical design conditions, not what occurred among the enrolled participants.

How should adverse-event summaries be read?

In general, the number of participants affected by an adverse-event category differs from the number of event episodes. A participant may experience repeated episodes or more than one type of event. The registry gives at-risk and affected participant counts for serious and other adverse events, and it lists percentages for selected serious terms. It does not provide the term-specific counts in the registry record recorded here, or clarify overlap among terms. Consequently, term percentages should not be summed or treated as mutually exclusive categories.

General statistical interpretation should keep descriptive summaries separate from comparative evidence. The safety counts describe what was recorded in this study group. Without a concurrent comparator, they do not establish how event frequency differs from another treatment group. The registry also does not report deaths in the safety details, so a death count cannot be inferred from the adverse-event summaries.

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("Decitabine 20 mg/m2 Intravenous")
n      <- c(55)
events <- c(13)
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  Decitabine 20 mg/m2 Intravenous  55  13  23.6  13.2  37.0  14.4  36.3  14.4  36.4

11. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, comparisons, or analyses are examined. In general, the more statistical questions tested, the greater the chance that at least one result may appear notable by chance if each is judged separately. Methods for addressing multiplicity can include prespecifying a primary question, adjusting significance thresholds, or using a hierarchy of tests. The registry lists one primary outcome measure and one analysis, but the statistical plan and any multiplicity procedure are not reported in the registry record. The number of measures listed alone does not establish the full scope of analyses performed.

The registry identifies the primary endpoint population as ITT. It does not state how missing response assessments were handled, whether participants who discontinued treatment remained in the endpoint analysis, or how incomplete follow-up was addressed. These details are not reported in the registry record. Missing-data methods can affect the interpretation of a participant count or proportion, especially if the probability of missingness is related to outcome status. No assumptions about the missingness mechanism can be made from the recorded details.

The study summary mentions overall survival, but an overall-survival outcome measure and its analysis population are not reported in the registry record. The record does not describe interim analyses, stopping rules, adjustment for repeated looks at accumulating data, or a statistical analysis plan. These matters are not reported in the registry record and should not be inferred from the Phase II designation.

The single-group, non-randomized design is also relevant to multiplicity and interpretation: a nominal analysis of the primary endpoint would not create a randomized treatment comparison. Since the primary response result and its uncertainty are not reported, there is no basis to evaluate statistical significance, multiplicity adjustment, or the sensitivity of the result to missing-data choices.

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: febrile neutropenia (23.6%), disease progression (16.4%), sepsis (14.5%), pneumonia (12.7%), bacteraemia (10.9%).

The registry lists a safety group titled Decitabine 20 mg/m2 Intravenous. For serious adverse events, 55 participants were at risk and 40 were recorded as affected. For other adverse events, 55 participants were at risk and 55 were recorded as affected. These are participant counts, not counts of all event episodes. The registry does not provide a death count in the safety details.

The listed serious-event terms and percentages are Febrile Neutropenia, 23.6%; Disease Progression, 16.4%; Sepsis, 14.5%; Pneumonia, 12.7%; Bacteraemia, 10.9%; and Neutropenia, 7.3%. These are term-level percentages as recorded. The registry entries do not specify whether a participant could be counted under multiple terms, nor do they give the term-specific participant counts. The percentages should therefore not be added together or interpreted as separate, non-overlapping shares of participants.

In general, safety summaries are descriptive: they report events observed and recorded in a study group. The absence of a concurrent control means these figures do not provide a like-for-like comparison of event frequency with another treatment. The registry entries also do not describe event collection procedures or attribution rules. Such details are not reported in the registry record.

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

The registry record identifies a completed Phase II study of decitabine in older patients with AML, a single-group non-randomized design, and 55 participants enrolled. It specifies a primary endpoint of ITT participants with complete response, defined as mCR, within 1 year. These details tell readers what the study set out to examine and how its primary outcome is labeled.

They do not establish the frequency of complete response in the enrolled group. The primary response count, response estimate, and confidence interval are not reported in the registry record. The stated study purpose also includes overall survival, but an overall-survival measure and result are not reported in the registry record. No quantitative statement about response or survival can therefore be made from these registry entries.

In general, a single-group result describes outcomes observed among participants in that group. It does not provide the same evidence as a randomized comparison with a concurrent control. The design cannot separate an observed outcome from other influences as directly as a randomized comparative study can. The registry explicitly cautions that the arms are not a like-for-like control comparison.

The safety summaries document participant counts affected by serious and other adverse events, along with percentages for selected serious-event terms. Those data are descriptive and do not show comparative safety relative to another treatment. The term percentages do not establish mutually exclusive event categories, and the record does not report a death count in the safety details.

In general, statistical precision depends on the observed estimate and the method used to quantify uncertainty. Since the primary endpoint estimate and its interval are absent, its precision cannot be assessed. Planning illustrations based on an observed effect are design exercises, not evidence that the study had a particular observed power or that an outcome was demonstrated. Overall, the record supports a description of the trial’s design and posted safety counts, but not a conclusion about the size or comparative effect of decitabine on response or survival.

14. Why This Trial Matters Statistically

This study is statistically informative as an example of why the design and the endpoint result must be considered together. The registry describes a Phase II, non-randomized, single-group study with 55 participants and a binary primary endpoint: participant count for mCR within 1 year. That structure defines the kind of quantity the study intended to estimate, but the primary outcome count is not reported in the registry record.

A single-group study can contribute descriptive evidence about outcomes recorded among participants treated in a particular study context. Its limitations become important when readers try to interpret an outcome as a comparative treatment effect. Without a concurrent randomized comparator, the study cannot provide a like-for-like estimate of how outcomes differ between treatment strategies. The registry’s explicit design warning makes that limitation central to interpretation, rather than a minor technical qualification.

The record also illustrates how an outcome label is not a complete analysis plan. It identifies the ITT population, endpoint definition, and timeframe, but does not state the response assessment rules, handling of missing outcomes, or statistical methods. These omissions limit the ability to evaluate how the endpoint was operationalized and how much uncertainty surrounds any estimate.

Finally, the safety summaries show why participant counts and event-term percentages require careful reading. Affected participants are not equivalent to total event episodes, and the record does not establish whether the listed terms overlap. Descriptive counts can be useful, but without a comparator they do not establish relative safety. The statistical lesson is to distinguish what the registry records directly from what a reader might wish to infer: design and safety summaries are available, while the primary response and survival results are not reported in the registry record.

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 most consequential limitation for interpreting the primary objective is that the registry entries do not report the mCR count, response rate, or uncertainty interval. Although the primary endpoint is clearly named and assigned a 1 year timeframe, the outcome result cannot be assessed from the recorded information. The brief summary mentions overall survival, but an overall-survival measure and result are not reported in the registry record.

The design is non-randomized and single-group. There is no concurrent control group described, and the registry warns that its arms are not a like-for-like control comparison. As a result, even if a response result were available, the design would not support a randomized comparative estimate against another treatment. The registry’s arm description is blank, and the treatment schedule is not reported in the registry record.

Important analysis details are also absent. The record does not describe the statistical analysis procedure, confidence-interval method, response assessment schedule, missing-data handling, or the operational criteria used to classify morphologic complete remission. It does not report the sample-size justification for the trial, interim analyses, stopping rules, or a statistical analysis plan. These omissions should not be filled by assumptions based on the Phase II designation.

The safety record gives group-level at-risk and affected counts and selected serious-event percentages, but does not provide the term-specific counts or explain whether terms overlap. A death count is not reported in the safety details. These limitations constrain detailed interpretation of the safety summaries and prevent a comparative safety assessment.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Study of Decitabine for Treatment of Older Patients With Acute Myeloid Leukemia (AML) — NCT00358644.
• Registry record last updated 2013-05-20; status last verified 2010-06.
• Results first posted 2009-09-07.
• 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 II study of decitabine for front-line treatment of older patients with acute myeloid leukemia. It is listed as non-randomized, unmasked, and single-group, with 55 participants enrolled. The recorded study start is 2005-03, and primary completion is 2008-02. The sponsor is Eisai Inc.

The primary endpoint is the number of ITT participants with complete response, defined as morphologic complete remission, within 1 year. The response count and estimate are not reported in the registry record. The brief summary also names overall survival as a study aim, but an overall-survival outcome measure and result are not reported in the registry record.

The safety group is titled Decitabine 20 mg/m2 Intravenous. The record lists 40 of 55 participants affected by serious adverse events and 55 of 55 affected by other adverse events, along with selected serious-event percentages. It does not report a death count in the safety details. The single-group design does not provide a randomized, like-for-like control comparison.