Gemcitabine and Mitoxantrone in Treating Patients With Relapsed Acute Myeloid Leukemia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Gemcitabine and Mitoxantrone in Treating Patients With Relapsed Acute Myeloid Leukemia (NCT00268242), a na phase 2 study. The primary outcome is Complete Response Rate.

2. Clinical Question

3. Trial Design

This was a phase II, single-group study sponsored by The Cleveland Clinic. The registry records no allocation to comparative arms and no masking. It lists an enrollment of 24 and a status of terminated. The record gives a start date of 2006-01 and a primary completion date of 2010-08. It identifies the disease as acute myeloid leukemia in first relapse in the official title.

The study evaluated gemcitabine in combination with mitoxantrone. Gemcitabine was given as a continuous intravenous infusion at 10mg/m^2/minute for 12 hours, starting on Day 1. Mitoxantrone was administered intravenously at 12mg/m^2/day on days 1, 2, and 3. These are the regimen details recorded for the study.

The primary endpoint was complete response, with an anticipated observation period of four weeks. The registry describes a two-stage design: initially, 18 patients were to be enrolled, with stopping planned if 5 or fewer achieved complete response; otherwise, an additional 22 patients were to be accrued. It also states that accrual was not halted while follow-up of the first 18 evaluable patients was under way. Consequently, 24 patients were enrolled. The record reports 5 complete responses (21%) and says the study was terminated.

That account should be read as the registry’s description of the intended design and the recorded conduct, not as evidence of a randomized comparison. The response result is an observation from the treated group. With no concurrent comparator, it cannot by itself establish how outcomes would have differed under another treatment or a different care strategy. A more detailed statistical analysis plan, including how analyses were to be conducted, 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: If \</= 5 of the initial 18 patients enrolled achieved a CR, the study would be stopped. Accrual was not halted while follow-up of the first 18 evaluable patients was under way. Therefore, 24 patients were enrolled. Only 5 patients (21%) achieved a CR and therefore, the study was terminated.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: A total of 5 patients (21%) achieved a complete response.

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 Rate, measured over 4 Weeks.

The primary endpoint was the number of participants achieving a complete response, reported as a complete response rate, with assessment at four weeks. A complete response is a binary outcome for each participant: the response criteria are either met or not met. Such an endpoint gives a clinically interpretable count of patients meeting a prespecified disease-response definition over a stated assessment period.

The registry says response was assessed using International Working Group criteria. Its description refers to morphologic complete remission, the morphologic leukemia-free state, an absolute neutrophil count of more than 1,000/μL, platelets of at least 100,000/μL, cytogenic complete response, and morphologic complete response with incomplete blood count recovery (CRi). The wording in the registry record does not provide a separate detailed operational definition for each component or explain how each category was handled in the reported count.

The record reports 5 patients (21%) achieving complete response and identifies this endpoint as the basis for terminating the study. A response count is useful for describing activity within the treated group, but it does not express the duration of response or survival. Disease-free and overall survival are listed among the secondary outcomes, but their results are not included in the registry record described for the primary endpoint. The record also does not report an uncertainty interval for the response estimate.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a phase II study with a single treatment group and a binary primary outcome. This design determines what kinds of statistical conclusions are supported. A single-group study can describe outcomes observed after treatment, but it cannot estimate a treatment contrast against a concurrent control group. The recorded allocation is not applicable, and masking is listed as none. These features matter because treatment assignment and knowledge of treatment can influence how a study is conducted and how results are interpreted; the registry record does not describe methods intended to address those concerns.

What does a binary response count tell readers?

A binary endpoint classifies each participant according to whether a defined event occurred. For this study, the event was achieving complete response according to the registry’s criteria. The total response count communicates how many participants met that definition, while a response rate expresses that count relative to an analysis denominator. The registry reports 5 patients (21%) achieving complete response and states that 24 patients were enrolled. It does not fully explain the denominator and analysis population behind the percentage in the results description. In particular, enrollment and evaluability are not interchangeable: a participant may be enrolled without having all outcome information available. How missing or unevaluable responses were handled is not reported in the registry record.

A point estimate is only one part of an outcome summary. Its precision depends on the amount of information and the statistical assumptions used. A confidence interval is a way of expressing uncertainty around an estimate; it is not a guarantee that the true value lies within that interval. The registry record does not report a confidence interval for the response result or identify a method for calculating one. It also does not report a p-value for testing the result against a benchmark. Readers therefore have the reported response summary but not an accompanying inferential analysis.

How should the two-stage design be understood?

The registry describes a two-stage design with a stopping rule based on the number of complete responses among the initial 18 patients. If 5 or fewer achieved complete response, the study was to stop; otherwise, an additional 22 patients were to be accrued. This is a design feature intended to allow an early decision based on accumulating response outcomes. It is not the same as a randomized trial, and it does not create a control group.

The conduct description adds that accrual was not halted while follow-up of the first 18 evaluable patients was under way. The registry says that 24 patients were therefore enrolled and reports 5 complete responses (21%), after which the study was terminated. The record does not provide a full patient-level timeline, the response status of each stage at each decision point, or enough detail to independently reconstruct the stopping decision. Those details are not reported in the registry record.

What is the role of the planning illustrations?

Planning illustrations use assumptions different from the actual single-group study. The recorded illustration assumptions include a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is described as a stated fraction of the observed effect. Such calculations are hypothetical planning aids, not observed outcomes and not observed power. In particular, equal allocation between two arms does not describe this single-group study. The registry does not report a separate sample-size calculation or justification for the trial’s own two-stage design beyond the response thresholds and enrollment stages.

More generally, a sample-size calculation depends on its assumed effect, variability or event rate, error criterion, allocation, and analysis method. If assumptions do not match the study design or the question of interest, an illustration may not represent the precision or operating characteristics of the actual trial. No new conclusion about the study’s power should be drawn from planning illustrations, and they do not change the interpretation of the observed response count.

What analysis details are absent?

The registry lists 6 outcome measures but no analyses. It does not report an inferential method for the primary endpoint, an approach to missing data, an analysis population definition, or a multiplicity procedure. The absence of those details means readers cannot determine from this record whether any particular model, adjustment, or sensitivity analysis was used. General statistical methods can explain what such choices mean, but they cannot fill in an undocumented trial analysis.

For example, over-dispersion is a general concern when count data vary more than a statistical model expects. It can affect uncertainty estimates in analyses of counts. The primary endpoint is reported as a binary response count, and no model or over-dispersion assessment is described. Likewise, adjustment for prognostic factors can sometimes improve precision or account for differences between groups, but this study has no comparative groups and the registry does not report an adjusted analysis. These concepts are relevant as general methods, not as claims about methods actually used in this trial.

The statistical reading is consequently bounded by what the record reports: a response count and percentage, a described two-stage design, and a termination statement. Further claims about inferential testing, uncertainty, or analysis choices would require details not reported in the registry record.

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("Gemcitabine + Mitoxantrone")
n      <- c(24)
events <- c(5)
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  Gemcitabine + Mitoxantrone  24  5  20.8  7.1  42.2  9.2  40.5  9.4  40.7

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 6 outcome measures. The primary endpoint is complete response rate; secondary outcomes include disease-free and overall survival, laboratory correlates by immunohistochemistry, white blood cell count at relapse, and the percentage of patients making it to bone marrow transplant. The registry entries available for the primary result do not include results for these secondary outcomes. The registry records no analyses, so a statistical plan for testing multiple outcomes is not reported in the registry record.

Multiplicity is the issue that arises when several outcomes or hypotheses are examined. If many tests are conducted, the chance of at least one apparently unusual result can increase unless the analysis plan accounts for the testing strategy. That is a general statistical consideration, not evidence that a particular correction was or was not applied in this trial. The registry does not report whether outcomes were tested, how any tests were prioritized, or whether any multiplicity adjustment was planned.

Missing-data and analysis-population details are also not reported. The record gives enrollment of 24 and describes follow-up of the first 18 evaluable patients during continued accrual, but it does not state how many patients were evaluable for the primary endpoint under a defined analysis rule or how unevaluable and missing responses were treated. The reported 5 patients (21%) should therefore be read as the registry’s response summary, without assuming an undocumented missing-data convention.

The record does not provide participant-level outcome completeness, reasons for missing assessments, or a sensitivity analysis. Those omissions limit assessment of how robust the response summary is to alternative assumptions about unavailable outcome data. They also prevent a complete reconstruction of which participants contributed to each secondary outcome.

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

Serious events with the highest arm-level rate: fever or infection (83.3%), rash (79.2%), nausea (66.7%), pain (66.7%), mucositis (58.3%).

The registry’s serious adverse event summary lists 24 participants at risk and 24 affected in the Gemcitabine + Mitoxantrone group. It also lists 24 participants at risk and 0 affected for other adverse events. The meaning attached to the event counts is participants with an event. These entries describe the registry’s recorded counts; they do not, by themselves, establish that treatment caused an event.

The listed serious adverse event terms and percentages are fever or infection, 83.3%; rash, 79.2%; nausea, 66.7%; pain, 66.7%; mucositis, 58.3%; and diarrhea, 58.3%. These are the top serious terms reported in the registry. They should be interpreted as term-specific summaries, not as mutually exclusive categories: the record does not say that a participant could contribute to only one term. The percentages should not be added together.

Deaths are not reported in the registry record. The available safety summary does not describe event timing, duration, severity grades, relatedness assessments, or how repeated events were handled. It also does not provide a comparison group. Consequently, these figures describe what the registry lists for the treated group but do not support a comparative safety estimate or a causal conclusion.

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

The result does mean that the registry reports 5 complete responses (21%) among patients treated in this single-group study and states that the study was terminated. The response was assessed at four weeks using the International Working Group criteria described in the record. The result can be used as a descriptive account of the response outcome recorded for this study.

It does not establish a comparative treatment effect. There was no concurrent control group, so the record cannot show how response would have differed under another treatment. It also does not establish that the combination caused the responses, or that the result would be the same in other patient groups or settings. Those questions require evidence beyond a single-group response summary.

The reported percentage does not communicate its statistical precision on its own. No confidence interval or p-value for the primary endpoint is reported in the registry record. The response count also does not tell readers how long responses lasted, whether patients survived longer, or how outcomes varied across participants. Survival is listed as a secondary outcome, but its result is not available in the registry record.

The registry’s benchmarks are part of the design description: a complete response rate of 30% or less was considered unacceptable, while 50% or more was considered promising. The record reports an outcome between those stated thresholds and says that the study was terminated. The registry does not provide a formal statistical comparison to either threshold, so the thresholds should not be mistaken for a reported hypothesis test or interval estimate.

Safety information is also descriptive. The registry reports serious adverse event counts and selected term percentages, but the registry entries do not establish causality or provide comparative risks. Overall, the findings characterize the outcomes recorded in this study; they do not settle comparative efficacy, durability, survival, or causal safety questions.

14. Why This Trial Matters Statistically

This trial illustrates why design and analysis must be read together. A response count can summarize what happened in a treated group, but without a concurrent comparator it cannot estimate a difference between treatment strategies. The distinction is important even when a study includes a predefined response threshold and a stopping design. A decision rule can organize enrollment and monitoring, but it does not supply the missing comparative information.

The registry’s description of the two-stage design also shows why accrual and outcome assessment timing matter. It states that accrual continued while follow-up of the first 18 evaluable patients was under way, leading to enrollment of 24. That operational detail is relevant to interpreting the relationship between the planned stopping rule and the recorded enrollment. The registry does not provide enough detail to recreate all decision points, which limits independent review of how the design was implemented.

The result is reported as both a count and a percentage, but no confidence interval or inferential test is reported. Statistical interpretation requires more than a point estimate: uncertainty and analysis rules help readers judge how much information the result contains and what claims it supports. The absence of analysis details is especially relevant for a small, single-group phase II study, where readers should avoid treating a descriptive percentage as a precise or comparative effect estimate.

Finally, the record distinguishes the primary response outcome from additional outcomes, including survival and laboratory measures. Without reported analyses and results for those measures, the primary response summary cannot stand in for a comprehensive account of benefit or harm. This trial is therefore a useful example of the boundary between an observed clinical outcome and the broader conclusions that require comparative design, complete reporting, and an explicit statistical analysis.

15. Limitations

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

The main limitation for interpretation is the single-group design. The study has no concurrent comparator, so the registry cannot estimate a treatment difference or separate the observed response from other influences on outcome. Allocation is recorded as not applicable and masking as none. These design features should be kept in view when interpreting both the response and safety summaries.

The enrollment was 24, and the primary result is reported as 5 responses (21%). The record does not report a confidence interval or other measure of uncertainty for this result. Nor does it provide a p-value or a listed statistical analysis. This leaves the precision and inferential basis of the response percentage uncharacterized in the registry record.

The registry describes a two-stage plan but also reports that accrual continued while follow-up of the first 18 evaluable patients was under way. It states that 24 patients were enrolled and that the study was terminated after 5 patients (21%) achieved complete response. It does not give a complete account of stage-specific results or decision timing, so the stopping process cannot be independently reconstructed from the record.

Information about analysis populations, missing outcome data, and the handling of unevaluable assessments is not reported in the registry record. The record also does not report an approach to multiplicity despite listing 6 outcome measures, or provide the results for the secondary outcomes in the registry record. Details on participant characteristics, response duration, and adverse event timing or relatedness are likewise not reported in the registry record. These limits mean the findings should be read as a narrow registry summary rather than a full account of treatment effects or clinical outcomes.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Gemcitabine and Mitoxantrone in Treating Patients With Relapsed Acute Myeloid Leukemia — NCT00268242.
• Registry record last updated 2018-02-22; status last verified 2015-08.
• Results first posted 2015-10-05.
• 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 record describes a phase II, single-group study of gemcitabine plus mitoxantrone in patients with acute myeloid leukemia in first relapse. The study enrolled 24 patients, used no masking, and is listed as terminated. The primary endpoint was complete response at four weeks. The registry reports 5 patients (21%) achieved a complete response and states that the study was terminated.

The record describes a two-stage design and reports that accrual continued during follow-up of the first 18 evaluable patients, resulting in enrollment of 24. No concurrent comparator is listed. The registry lists 6 outcome measures and no analyses. Confidence intervals, p-values, missing-data methods, analysis populations, and multiplicity procedures are not reported in the registry record. Serious adverse event counts and selected term percentages are reported, while deaths are not reported in the registry record.