Eltrombopag Olamine in Treating Thrombocytopenia in Patients With Chronic Myeloid Leukemia or Myelofibrosis Receiving Tyrosine Kinase Therapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Eltrombopag Olamine in Treating Thrombocytopenia in Patients With Chronic Myeloid Leukemia or Myelofibrosis Receiving Tyrosine Kinase Therapy (NCT01428635), a na phase 2, phase 3 study. The primary outcome is Number of Participants With a Platelet Response.

2. Clinical Question

The clinical question is whether platelet response could be observed during eltrombopag treatment in this clinical setting, not whether eltrombopag outperformed another treatment. Since the trial has no concurrent control group, any observed response could not, by itself, establish that the treatment caused the response. The registry names response to tyrosine kinase inhibitor therapy after eltrombopag as a secondary outcome, but does not give its result. The registered enrollment was 21, and the record is marked completed. The criteria used to select participants, the distribution of disease subtypes, and the extent to which participants completed treatment are not reported in the registry record.

3. Trial Design

The registry describes a phase II/III, completed, single-group study sponsored by M.D. Anderson Cancer Center. Its stated purpose was to study eltrombopag olamine for thrombocytopenia in people with CML or MF who were receiving tyrosine kinase inhibitor therapy. The listed enrollment was 21. The recorded start date is 2012-01-13, and the primary completion date is 2022-01-03.

The intervention arm is called supportive care with eltrombopag olamine. Participants received the drug orally once daily, with treatment described as continuing in the absence of disease progression or unacceptable toxicity. The registry lists masking as none and allocation as not applicable. No separate untreated, placebo, or active-comparator arm is described, so treatment comparisons between randomized groups are not available.

The disease categories listed include chronic, accelerated, and blast phase CML, primary MF, and thrombocytopenia. This indicates that the registered scope covered more than one disease category, but the record does not report detailed eligibility criteria or participant counts by disease subtype. The safety listings distinguish CML and MF groups; their serious-event risk counts are 15 and 6, respectively. These group-specific listings do not supply the primary platelet-response results.

The primary endpoint is a count of participants with a platelet response, assessed over a timeframe of up to 9 years. A response is defined using disease-specific platelet thresholds, sustained for 3 months, plus an increase from baseline. A secondary outcome concerns response to tyrosine kinase inhibitor therapy after eltrombopag. The registry indicates 2 outcome measures, but the second measure’s identity is not reported in the registry record. Details of the protocol, treatment duration for individual participants, and the analysis plan are also not reported.

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 Number of Participants With a Platelet Response, measured over Up to 9 years.

The primary endpoint counts participants classified as having a complete platelet response. The registry defines response as a sustained platelet count for 3 months above 50 x 109/L for CML or above 100 x 109/L for MF, together with at least a 20% increase from baseline. The endpoint is assessed over a timeframe of up to 9 years. The thresholds differ by disease group, so interpretation depends on which condition applies to each participant.

A count of participants meeting a prespecified response definition can describe how often that outcome occurred in the treated group. Requiring a sustained platelet level, rather than a single measurement, is intended to distinguish a continuing response from a transient change. Requiring an increase from baseline also makes the definition depend on participants’ starting platelet counts. These are features of the registered endpoint; the record does not explain the rationale for choosing the thresholds or the duration requirement.

The registry record does not report the number of participants who achieved a complete platelet response, the number assessed for this endpoint, or the endpoint’s statistical analysis. It therefore does not allow readers to determine the observed response frequency or its precision. The registered definition alone describes how response was to be classified, not what result was obtained.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a single-group design estimate?

In a single-group study, outcomes are observed among participants who receive the intervention, without a concurrent comparison group. If a response result is reported, it can describe the response observed in the treated group. It cannot, on its own, separate the effect of treatment from the natural course of illness, changes in other care, or other influences. This is a general limitation of the design, not a finding about what happened to participants in this trial.

The registry identifies the model as single group, allocation as not applicable, and masking as none. There is no separately described control arm. Thus, the natural interpretation is descriptive rather than a randomized treatment comparison. The record does not report an analysis that adjusts for other factors or a design feature intended to estimate a counterfactual outcome.

How should the platelet-response definition be read?

The primary endpoint is a binary classification: each participant either meets the complete platelet-response definition or does not. The definition requires a platelet count sustained for 3 months above 50 x 109/L for CML or above 100 x 109/L for MF, and at least a 20% increase from baseline. The group-specific thresholds mean response classification is tied to the participant’s disease group. The registry does not report how baseline values, missing measurements, or partial follow-up were handled.

In general, a binary endpoint makes the outcome rule explicit, but discards detail about the amount and pattern of change once participants are classified. A participant just meeting a threshold and one exceeding it by a larger amount both count as responses under the same binary measure. The registry record does not post continuous platelet results that would allow readers to examine the underlying measurement changes.

What does a confidence interval convey?

In general, a confidence interval describes the statistical uncertainty around an estimated quantity under a specified method. For a response proportion, an interval would help show how precisely the observed data estimate the underlying response frequency. It would not correct the absence of a control group, remove bias from missing outcome data, or establish that treatment caused the observed responses. No primary response estimate or confidence interval is reported in the registry record, so precision cannot be assessed.

The same distinction applies to a significance test. In general, a p-value assesses how compatible data are with a stated statistical model or null hypothesis. It does not measure the size or clinical importance of an effect. Since the record reports 0 analyses and does not post a primary result, there is no registry-reported p-value to interpret.

Why can over-dispersion matter?

In general, over-dispersion occurs when variation in observed data exceeds what an analysis model assumes. If unaccounted for, it can make standard errors and uncertainty estimates unreliable. This concept is relevant to some count and proportion models, but the registry record does not state that an over-dispersion model was used, or report enough outcome data to assess whether over-dispersion was present. It should not be inferred that such a problem occurred in this trial.

What do the planning assumptions mean?

The stated planning illustration uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; its target effect is described as a fraction of the observed effect. Such assumptions are used in general to explore how many participants might be needed for a hypothetical comparison under specified conditions. The record explicitly characterizes the illustration as planning based on the observed effect, rather than as the trial’s own design. It is not observed power.

Those assumptions do not correspond to the registered single-group design, which has no two-arm allocation. The record does not report the trial’s own sample-size justification, the effect size used in planning the actual study, or a prospective power calculation. Sample-size illustrations should therefore not be mistaken for the enrollment rationale or evidence that a treatment effect was established.

What analysis details are unavailable?

The registry record does not report the statistical test or model for the primary endpoint, a confidence interval, adjustment variables, subgroup analyses, or handling of repeated measurements. It also does not describe a prespecified interim analysis, monitoring plan, or stopping rule. The absence of these details means the exact analytic approach cannot be reconstructed from the registry record. General statistical concepts explain how methods might work, but they do not establish which methods were used in this trial.

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("Supportive Care (Eltrombopag Olamine) CML Particip", "Supportive Care (Eltrombopag Olamine) Myelofibrosi")
n      <- c(15, 6)
events <- c(12, 0)
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  Supportive Care (Eltrombopag Olamine) CML Particip  15  12  80.0  51.9  95.7  54.8  93.0  54.4  92.7
ARM  Supportive Care (Eltrombopag Olamine) Myelofibrosi  6  0  0.0  0.0  45.9  0.0  39.0  0.4  41.0

11. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, comparisons, or subgroup analyses are tested. In general, conducting many tests can increase the chance of a nominally positive finding by chance, so a statistical plan may define a primary outcome and specify how other tests will be handled. The registry identifies a primary platelet-response endpoint and a secondary outcome concerning response to tyrosine kinase inhibitor therapy after eltrombopag. It also records 2 outcome measures, but the second measure’s identity and any multiplicity strategy are not reported in the registry record.

The record does not state whether statistical tests were adjusted for multiple outcomes, whether any subgroup comparisons were planned, or whether an outcome hierarchy governed interpretation. It also does not report interim analyses. These details should not be inferred from the phase designation or the completed status. The registry indicates 0 analyses, so there are no posted analysis results with which to evaluate multiplicity in practice.

Missing outcome data can affect a response count if participants lack the measurements needed to determine whether they met the sustained platelet criteria. In general, the implications depend on how much data are missing, why they are missing, and the analysis rule. The registry record does not report the number of participants with missing platelet measurements, withdrawals, exclusions, or the method used to handle incomplete follow-up. The analysis population is also not described. It is therefore not possible to determine whether the primary endpoint was assessed among all enrolled participants, only those with evaluable measurements, or another group.

These omissions limit assessment of both the completeness of outcome ascertainment and the reliability of any unposted result. They do not establish that data were missing or that an analysis was biased; they mean the registry record does not provide the information needed to assess those questions.

12. Safety Results

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

Deaths

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: infection (33.3%), bullous dermatitis (16.7%), fatigue (16.7%), flu like symptoms (16.7%), lung infection (16.7%).

The safety listings separate CML and MF participants. For serious adverse events, the registry lists 10 affected among 15 CML participants at risk and 4 affected among 6 MF participants at risk. For other adverse events, it lists 13 among 15 CML participants and 4 among 6 MF participants at risk. One death is listed in each group. These are counts in the registry’s safety record; they should not be interpreted as evidence that the intervention caused the events.

The listed serious-event terms include Infection at 33.3%, and Bullous dermatitis, Fatigue, Flu Like Symptoms, Lung Infection, and Myocardial Infarction at 16.7% each. The registry record does not explain the denominator used for these term percentages or whether a participant could appear under more than one term. The percentages therefore cannot be translated into participant counts from the registry record alone.

Safety counts are not directly comparable with a control group because no control arm is described. The record also does not report event timing, duration of treatment, severity grading details, investigator attribution, or whether events led to treatment changes. These figures are a limited description of the safety information posted, not a comparative assessment.

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

The registry record describes a study of eltrombopag olamine in people with thrombocytopenia associated with CML or MF while receiving tyrosine kinase therapy. It defines a primary platelet-response outcome and lists safety counts for CML and MF groups. However, it does not report the primary response count, the number assessed for response, or a statistical analysis. The available registry information therefore does not establish how often the defined response occurred.

Even if a response count were available, this single-group design would support a description of outcomes among treated participants, not a controlled estimate of what treatment changed. There is no concurrent comparator arm in the registry record. Observed outcomes could not be attributed to eltrombopag alone on the basis of this design, and the record does not report an adjustment strategy that would address this limitation.

The safety listings identify participants affected by serious and other adverse events and include deaths, but they do not establish causation. Without a comparator and further information about timing, attribution, and exposure, the counts cannot show whether event occurrence differed because of the intervention.

The record also leaves key interpretive details unavailable, including the actual analysis population, missing-data handling, and statistical methods. The planning assumptions attached to sample-size illustrations are hypothetical and are not the trial’s own design or observed power. Thus, the registry record is informative about the question, intervention, endpoint definition, and some safety counts, but does not provide enough outcome or analysis information to assess the primary result or make a comparative treatment-effect claim.

14. Why This Trial Matters Statistically

This trial illustrates why a clearly defined endpoint and a reported outcome result are both necessary for statistical interpretation. The platelet-response rule specifies disease-specific thresholds, a duration requirement, and a change from baseline. That makes the intended classification understandable. Yet without the number of participants meeting that definition or the number evaluated, readers cannot assess the observed response frequency or its precision.

The design also matters. A single-group study can document what occurred among people receiving an intervention, but it does not provide a concurrent estimate of outcomes under an alternative. This distinction is especially important when describing a treatment response: an observed outcome and a treatment-caused outcome are not interchangeable claims. The registry’s lack of a comparator constrains what any posted count could establish.

The safety record shows another statistical issue: counts and percentages require context. Group-specific counts are listed, and selected serious-event terms have percentages, but the denominator and counting method for those term percentages are not reported. Without that context, precise comparative interpretation is not possible.

Finally, analysis transparency determines whether readers can evaluate how outcomes were handled. The registry does not report the primary analysis, analysis population, missing-data approach, or multiplicity strategy, and it indicates 0 analyses. The planning illustrations use assumptions that do not match the single-group trial design. Keeping planned calculations, observed results, and causal conclusions distinct is essential to a sound reading of the 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 main limitation for interpreting the registered clinical question is that the primary platelet-response result is not reported in the registry record. The record does not give the number achieving response, the number assessed, or an estimate with uncertainty. Although the study is marked completed, completion status does not fill that reporting gap.

The single-group design has no concurrent comparator. Consequently, it cannot provide a like-for-like estimate of outcomes under eltrombopag versus another treatment or no eltrombopag. The study also covers CML and MF, with CML phase categories listed, but detailed eligibility criteria and participant characteristics by subtype are not reported. The separate safety group counts do not resolve that lack of clinical detail for the efficacy endpoint.

Further limitations arise from unavailable analysis information. The registry record does not report the analysis population, handling of missing platelet measurements, statistical method, confidence interval, or multiplicity approach. It also does not describe interim analyses, a stopping rule, or the trial’s own sample-size justification. The identity of the second of the 2 outcome measures is not reported in the registry record.

Safety interpretation is limited by the absence of a control group and by missing context on event timing, exposure, attribution, and the denominators for term-specific percentages. The registry lists deaths, but does not describe their circumstances or relationship to treatment. These limitations mean the posted record supports a description of study structure and selected safety counts, but not a complete assessment of efficacy or comparative safety.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Eltrombopag Olamine in Treating Thrombocytopenia in Patients With Chronic Myeloid Leukemia or Myelofibrosis Receiving Tyrosine Kinase Therapy — NCT01428635.
• Registry record last updated 2023-09-26; status last verified 2023-09.
• Results first posted 2023-02-01.
• 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

NCT01428635 is a completed phase II/III, single-group study sponsored by M.D. Anderson Cancer Center, with listed enrollment of 21. It evaluated eltrombopag olamine as supportive care for thrombocytopenia in people with CML or MF receiving tyrosine kinase therapy. The intervention was taken orally once daily, with treatment described as continuing unless disease progressed or toxicity became unacceptable.

The primary endpoint was the number of participants with a complete platelet response, defined by sustained, disease-specific platelet thresholds and at least a 20% increase from baseline, over a timeframe of up to 9 years. A secondary outcome concerns response to tyrosine kinase inhibitor therapy after eltrombopag. The primary response result and statistical analyses are not reported in the registry record. Safety listings give serious and other adverse-event counts by CML and MF group, along with deaths and selected serious-event terms.