Study of Azacytidine Followed by GM-CSF in Patients With Myelodysplastic Syndrome (MDS): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Azacytidine Followed by GM-CSF in Patients With Myelodysplastic Syndrome (MDS) (NCT01542684), a na phase 2 study. The primary outcome is Overall Response Rate (ORR).

2. Clinical Question

Statistically, this is a question about the proportion of participants in a treated group who met the specified response criteria during a defined assessment period. It is not a randomized comparison of two regimens, and the study cannot by itself estimate a treatment difference against an untreated or alternative-treatment group. The response criteria are based on modified International Working Group criteria, with descriptions for complete response, partial response, stable disease, and failure or progression. The primary endpoint counts complete and partial responses as responses; stable disease is not included in that numerator.

The record names the target MDS risk groups in the official title, while its condition field says leukemia. That difference is part of the registry record and should be kept in mind when interpreting the listed population. The record does not provide a detailed participant description, baseline characteristics, or an explanation of the termination. Those details are not reported in the registry record.

3. Trial Design

This was a Phase II, single-group study of azacytidine followed by GM-CSF. The registry lists allocation as NA and masking as none. There was no concurrent control arm, so the study was designed as a description of outcomes in participants receiving the regimen rather than as a randomized comparison. Enrollment is listed as 8. The sponsor is M.D. Anderson Cancer Center. The record lists the study as terminated, with a start date of 2012-03 and primary completion date of 2013-04.

The official title identifies patients with low- or intermediate-1-risk MDS. The brief summary describes a goal of learning whether the combination could help control MDS and also studying safety. Those statements describe the research aims; they are not evidence of benefit or harm. No reason for termination is reported in the registry record, and the record does not explain whether termination changed enrollment, follow-up, or the planned analyses.

The treatment description specifies azacytidine administered intravenously or subcutaneously at a starting dose of 40 mg/m² daily for 4 days. GM-CSF was specified at 250 mcg/m², given intravenously or subcutaneously on the next day after azacytidine treatment, for 3 consecutive days. The record says that each treatment cycle will last at, but the remaining text is absent. The complete cycle schedule and any dose changes or treatment-discontinuation rules are not reported in the registry record.

The registry lists one outcome measure and no analyses. It identifies ORR as the primary endpoint, measured from baseline up to 2 treatment cycles, described as 8 weeks. The registry record does not give the number of participants assessed for that endpoint, the response count, or the analysis population. It also does not report a protocol, a statistical analysis plan, or a sample-size justification. Any planning calculations associated with the posted results are illustrations under stated assumptions, not evidence of the original design plan or observed statistical power.

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

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 Rate (ORR), measured over Baseline up to 2 treatment cycles (8 weeks).

The primary endpoint is overall response rate (ORR), expressed as the percentage of participants with complete response (CR) or partial response (PR) within two treatment cycles. The timeframe is described as baseline up to 2 treatment cycles (8 weeks). Thus, the endpoint classifies participants according to whether their recorded disease response met either of those two response categories during the specified period.

The registry describes response using modified International Working Group criteria. Its complete-response description includes bone marrow 5% myeloblasts with normal maturation of all cell lines, persistent dysplasia noted, peripheral blood Hgb 11 g/dL, platelets 100x109/L, neutrophils 1.0x109/L, and blasts 0%. The partial-response description refers to meeting complete-response criteria if abnormal before treatment, except that bone marrow blasts decreased by 50% over pretreatment but still > 5%; it also says cellularity and morphology are not relevant. The record describes stable disease as failure to achieve at least partial response without evidence of progression for > 8 weeks. These are registry descriptions; the detailed criteria should not be silently revised or completed beyond what the record states.

ORR is used to summarize the proportion meeting a prespecified response definition over a defined observation period. In general, a binary response endpoint gives a straightforward count-based summary, but it does not show the duration of response or distinguish outcomes beyond the categories used. Here, the registry record does not report the observed ORR, its uncertainty, or the number evaluated. Consequently, the endpoint’s definition is available, but its result and precision cannot be assessed from the registry record.

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 can describe the observed outcomes among participants who receive the study regimen. For a binary endpoint such as response, the basic quantity is the proportion classified as having the event of interest within the defined assessment period. That proportion is meaningful only alongside a clear numerator, denominator, endpoint definition, and account of missing assessments. The registry record names the endpoint and criteria but does not report the response count, evaluated denominator, or estimate.

Because this study has no concurrent control group, the observed response proportion, if reported, would not by itself estimate a contrast between treatments. The group’s outcome could reflect the regimen, the natural course of disease, selection of participants, other care, measurement choices, or a combination of factors. This is a general limitation of non-comparative designs, not a claim about what happened in this study. The registry lists allocation as NA and the model as single-group.

Why does endpoint definition matter?

Binary outcomes depend on how categories are assigned. Here, ORR includes complete response or partial response, while the record separately defines stable disease and no response or failure. The criteria and observation window determine who counts as a responder. In general, if assessments are not made consistently, or if participants are classified differently across time, the estimated proportion may not represent a uniform outcome. The registry record does not describe the assessment procedures, adjudication, or handling of ambiguous response classifications.

The timeframe is described as baseline up to 2 treatment cycles (8 weeks). A fixed assessment window gives a common period over which response is considered, but it does not answer questions about response durability or later outcomes. The record does not report follow-up beyond that endpoint timeframe or provide a separate duration-of-response outcome.

What does uncertainty mean for a proportion?

General statistical explanation: a confidence interval communicates uncertainty around an estimate. It reflects that the observed proportion is based on a finite group and that another sample from a similar population could yield a different estimate. A narrower interval indicates greater precision under the method’s assumptions; a wider interval indicates less precision. The interval does not automatically account for bias, missing outcomes, or lack of a comparator. No ORR interval or method for calculating one is reported in the registry record.

For binary outcomes, methods typically rely on the number classified with the event and the number assessed. Different interval methods can behave differently, especially when data are sparse or an estimate is near a boundary. Which method was used is not reported in the registry record. Because neither the response count nor denominator is posted, there is no basis to reconstruct an endpoint estimate or its interval.

Why can over-dispersion matter?

General statistical explanation: over-dispersion occurs when observed variation is greater than a model assumes. It is important in analyses where outcomes are modeled as counts or proportions, because an overly simple variance assumption can make standard errors too small and confidence intervals too narrow. Depending on the data and model, analysts may use methods that account for extra variation. The registry record does not report an ORR model, an over-dispersion assessment, or an adjustment for it. For a simple binary summary, the key issue remains the response count, denominator, and uncertainty method; none is given for the primary endpoint.

How should planning illustrations be read?

General statistical explanation: sample-size calculations estimate how much information a future study might need under specified assumptions. They are planning tools, not observations from participants and not proof that a completed study had adequate power. The record describes planning illustrations based on the observed effect, a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; it says the target effect is a stated fraction of the observed effect. It also distinguishes totals of events for time-to-event settings from participants per arm for other settings.

Those assumptions are not a description of this trial’s design: this trial is listed as single-group, not as an equal-allocation two-arm study. The registry record does not report a sample-size justification for the trial or identify an original target enrollment. Sample-size illustrations should therefore be read only as hypothetical planning scenarios under their stated assumptions. They are not observed power, and they do not remedy the absence of a reported ORR result or its uncertainty.

What can be inferred from the analysis listing?

The registry lists zero analyses. This indicates that no analysis results are listed in the record; it does not establish why they are absent or whether analyses were conducted outside the registry. The statistical analysis plan, any interim analysis, adjustment strategy, and exact approach to missing data are not reported in the registry record. Consequently, the posted information does not support reconstruction of a treatment effect, hypothesis test, or adjusted estimate.

10. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, comparisons, or repeated statistical tests are considered. In general, testing many questions can increase the chance of obtaining an apparently notable result by chance, so a statistical plan may specify a primary endpoint and approaches to interpreting additional tests. The registry lists one outcome measure and zero analyses. It does not report a multiplicity adjustment, a hierarchy of endpoints, or a plan for secondary outcomes. With no posted analysis results, there are no reported tests to interpret or adjust.

The primary endpoint is identified as ORR, but the record does not provide an analysis population definition. It does not say whether the endpoint was to be analyzed among all enrolled participants, only those who completed treatment, or another defined group. The number evaluated for response is not reported in the registry record. Nor does the record describe how deaths, withdrawals, incomplete assessments, or other missing response information were to be classified or handled.

In general, missing outcome data can affect a proportion because the estimate depends on which participants are included and how unobserved outcomes are treated. Excluding participants with missing assessments, classifying them as nonresponders, or using another method can lead to different estimates. These are general possibilities, not descriptions of what this study did. No missing-data method, imputation approach, sensitivity analysis, or assessment of missingness is reported in the registry record.

The record also does not report an interim analysis plan or statistical analysis plan. The absence of these details does not show that a particular method was or was not used; it means the registry record does not document it. With the response count, endpoint denominator, analysis population, and missing-data handling unreported, the primary endpoint cannot be independently reconstructed from the posted information.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: fever (12.5%), injection site reaction (12.5%), urinary tract infection (12.5%), bone pain (12.5%).

The safety record lists one group, Azacytidine + GM-CSF, with 8 participants at risk for serious adverse events and 3 affected. For other adverse events, it lists 8 at risk and 8 affected. These are event-category counts as recorded; they do not describe the number of events per participant, timing, duration, or whether events were considered related to treatment. No death count is listed.

The most frequently listed serious adverse event terms each have a reported value of 12.5%: fever, injection site reaction, urinary tract infection, and bone pain. These percentages should be read as registry-reported values. The record does not specify further detail about each term’s severity, recurrence, timing, clinical course, or attribution. The terms should not be added together to infer a total, because the registry record does not establish whether the same participant experienced more than one term.

In general, safety counts in a small single-group study are descriptive. They do not establish how event rates compare with another regimen or with background rates in a similar population. The registry record does not report a control group, exposure-adjusted rates, or a detailed adverse-event analysis. The listed safety figures therefore describe recorded events in this group, not a comparative estimate of treatment-associated risk.

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

The registry record establishes that a Phase II, single-group study of azacytidine followed by GM-CSF was registered for patients with low- or intermediate-1-risk MDS. Its primary endpoint was ORR, defined using complete or partial response within two treatment cycles, described as 8 weeks. The record also lists safety counts for the treatment group and reports that the study was terminated.

It does not establish the observed response proportion. Although the endpoint definition is detailed, no number of participants with complete or partial response, denominator for the endpoint, or uncertainty estimate is reported. The record also does not report a statistical analysis of the response endpoint. Therefore, the available information cannot show how often response occurred or how precisely any estimate was measured.

The safety entries show that serious and other adverse events were recorded in the group, with the counts and serious-event terms listed in the registry. They do not establish causation, comparative risk, or the overall balance of benefits and harms. There is no concurrent control group against which to compare response or safety outcomes.

In general, a single-group study can provide descriptive information, but it cannot by itself separate the contribution of a regimen from other factors. The status of terminated does not explain why the study stopped, and the reason is not reported in the registry record. It should not be treated as a result about the treatment. The record likewise does not report whether the planned endpoint assessment was completed or whether additional results exist elsewhere.

The responsible interpretation is therefore limited: the registry describes the intended population, regimen, endpoint, and certain adverse-event counts, but the primary response result and its statistical precision are not reported in the registry record. No conclusion that the regimen works or does not work is warranted from these entries alone.

13. Why This Trial Matters Statistically

This study illustrates why a clearly defined endpoint and a reported endpoint result are both necessary for statistical interpretation. ORR is specified with response categories and a time window, but the registry does not report the response count or the number assessed. Without those basic data, readers cannot recover the estimate, evaluate its precision, or compare it with another estimate.

It also illustrates the limits of a single-group design. Describing responses among people who received a regimen is different from estimating the effect of that regimen relative to an alternative. A concurrent comparator can support a direct comparison; this study has none. The design information therefore determines which questions can be answered, regardless of how carefully the endpoint is defined.

Small enrollment makes transparent reporting particularly important. A binary proportion can be sensitive to the outcome of individual participants, and uncertainty should be communicated alongside the estimate. In this record, neither the estimate nor an interval is available. This is not a basis for calculating or assuming an outcome; it is a reason to distinguish the study’s stated objective from a result.

The safety entries also show why adverse-event counts need context. The registry gives counts at risk and affected and lists serious-event terms, but does not describe timing, attribution, recurrence, or a control group. These limitations constrain interpretation of the safety observations.

Finally, the record identifies the study as terminated without reporting why. Statistical interpretation should not turn status labels into evidence of treatment effect or study quality. The case highlights the value of reporting the analysis population, missing-data approach, response denominator, estimate, uncertainty, and analysis plan so that readers can understand both what the data show and what they cannot establish.

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 primary limitation is that the registry record does not report the primary endpoint result. It names ORR and describes response criteria, but gives no response count, denominator, estimate, confidence interval, or p-value. The number assessed for response is also not reported in the registry record. This prevents independent assessment of the magnitude and precision of the endpoint.

The study is single-group, with no concurrent comparator. Any outcome observed in the group would be descriptive rather than a randomized estimate of a treatment difference. The record lists allocation as NA and masking as none. It does not report procedures for response assessment or whether response classifications were independently reviewed.

Enrollment is listed as 8, but the registry does not establish how many participants completed treatment or had an evaluable response assessment. The analysis population and missing-data handling are not reported in the registry record. The planned cycle description is truncated, so the full treatment-cycle schedule is unavailable. The record also does not provide a statistical analysis plan, sample-size justification, or explanation of termination.

The official title describes low- or intermediate-1-risk MDS, while the condition field lists leukemia. The registry does not clarify this difference. The adverse-event information is limited to counts and listed terms; it does not provide event timing, duration, attribution, recurrence, or comparative rates. These constraints limit interpretation of both response and safety, and they cannot be resolved by inference from the study’s aims or design.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Study of Azacytidine Followed by GM-CSF in Patients With Myelodysplastic Syndrome (MDS) — NCT01542684.
• Registry record last updated 2014-04-30; status last verified 2014-03.
• Results first posted 2014-04-30.
• 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

NCT01542684 is a Phase II, single-group study of azacytidine followed by GM-CSF in patients identified in the official title as having low- or intermediate-1-risk MDS. The registry lists enrollment as 8, allocation as NA, masking as none, and status as terminated. The sponsor is M.D. Anderson Cancer Center. The listed dates are start 2012-03 and primary completion 2013-04.

The primary outcome is ORR, defined as complete or partial response within two treatment cycles, described as 8 weeks. The registry lists one outcome measure and zero analyses; the ORR result is not reported in the registry record. Safety entries list 3 of 8 affected by serious adverse events and 8 of 8 affected by other adverse events. Four serious adverse event terms are listed at 12.5% each.

The record does not report the reason for termination, the response denominator, an analysis population, missing-data methods, or a statistical analysis plan. The available information supports a description of the trial and listed safety reporting, but not a quantitative conclusion about response.