This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for A Phase II Dose-escalation Study Characterizing the PK of Eltrombopag in Pediatric Patients With Previously Untreated or Relapsed Severe Aplastic Anemia or Recurrent Aplastic Anemia (NCT03025698), a non-randomized phase 2 study. The primary outcome is Eltrombopag PK Parameters: AUCtau, AUClast.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03025698 |
| Phase | Phase 2 |
| Condition | Aplastic Anemia |
| Allocation | Non-Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 51 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Novartis Pharmaceuticals |
| Results posted | Yes |
2. Clinical Question
Population
Pediatric patients with previously untreated, refractory or relapsed severe aplastic anemia, or recurrent aplastic anemia. The registry reports enrollment of 51 participants. The pharmacokinetic analysis set included participants who provided an evaluable pharmacokinetic profile; the number analyzed for each outcome is the subset with available and evaluable data.
Intervention
Oral eltrombopag with intra-patient dose escalation, in combination with immunosuppressive therapy as described for the cohorts. Cohort A, Regimen 1 began hATG, CsA and eltrombopag on Day 1. Cohort A, Regimen 2 began CsA and eltrombopag on Day 1. Cohort B began hATG, CsA and eltrombopag on Day 1.
Comparator
There was no controlled comparator. The study was open-label and non-randomized. Its cohorts and regimens describe study treatment groups, not randomized arms that can be interpreted as like-for-like controls.
Primary question
The primary objective was to characterize eltrombopag pharmacokinetics by measuring AUCtau and AUClast. Measurements were planned at least 11 weeks after dose initiation, or later when participants were taking the highest dose, up to Week 78. Numerical pharmacokinetic results are not reported in the registry record.
This is a pharmacokinetic characterization question: how much drug exposure was measured under the study’s dosing conditions, as represented by area-under-the-curve measures. It is not, on the basis of the primary endpoint, a test of whether eltrombopag improves clinical outcomes. The registry also lists response and blood-count outcomes, along with transfusion-related outcomes, as secondary measures. Their numerical results are not reported in the registry record.
Because there was no controlled comparator, the study does not provide a randomized comparison of eltrombopag against another treatment or against no eltrombopag. The cohort descriptions also differ in whether hATG was included from Day 1. Any differences between cohort results, if available, could reflect those treatment arrangements and other differences rather than an isolated eltrombopag effect. The registry record does not report a causal comparison between the cohorts.
3. Trial Design
The registry describes a Phase II, open-label, non-controlled, non-randomized study using intra-patient dose escalation. Its stated purpose was to characterize pharmacokinetics after oral eltrombopag administration in pediatric patients with severe or recurrent aplastic anemia. The study used a parallel model, but the record does not describe randomized assignment to parallel treatment arms. The trial is listed as completed, with enrollment of 51 participants.
The treatment descriptions distinguish Cohort A Regimen 1, Cohort A Regimen 2 and Cohort B. In Regimen 1 and Cohort B, hATG, CsA and eltrombopag began on Day 1. In Regimen 2, CsA and eltrombopag began on Day 1. The registry identifies the drugs as eltrombopag, hATG and CsA. Details of how patients entered each cohort or regimen are not reported in the registry record.
The primary pharmacokinetic assessment was scheduled at least 11 weeks after dose initiation, or later when participants were taking the highest dose, up to Week 78. The registry gives this timeframe for the AUCtau and AUClast measures. It does not report the actual timing of each participant’s pharmacokinetic sampling or the numerical values obtained.
The study started on 2017-09-30 and reached primary completion on 2022-04-22. Because it was non-controlled and non-randomized, its treatment groups should not be read as if they were exchangeable groups created by random allocation. The design can describe pharmacokinetic measurements and observed outcomes within the studied treatment settings, but by itself it does not isolate the effect of eltrombopag from the effects of concomitant treatment, patient differences or time.
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: Participants participated in a treatment period (26 weeks), follow-up period (additional 52 weeks) and a Long-term Follow-up period for 3 additional years. A dose of 25mg daily for 1 to 6 years and 50 mg daily for 6 to \<18 years was administered. Doses could be modified every two weeks based on blood platelet counts and dosing guidelines until targeted platelet count or maximum dose of 150 mg was achieved, whichever occurred first. .
Overall Study
| Stage | Cohort A: Regimen 1 | Cohort A: Regimen 2 | Cohort B |
|---|---|---|---|
| STARTED | 10 | 4 | 37 |
| Participants Who Entered 26-week Treatment Phase | 10 | 4 | 37 |
| Completed 26-week (Wk) Treatment | 7 | 4 | 25 |
| Entered 52-wk Post-treatment Follow up Phase 1 (Wk 78) | 9 | 4 | 26 |
| Did Not Enter 52-wk Post-treatment f/u Phase (Wk 78) | 1 | 0 | 11 |
| Entered 3-year Post-treatment Follow-up 2 | 6 | 3 | 19 |
| Did Not Complete PTFU2 | 1 | 1 | 9 |
| COMPLETED | 5 | 2 | 10 |
| NOT COMPLETED | 5 | 2 | 27 |
Reasons for not completing this period:
| Reason | Cohort A: Regimen 1 | Cohort A: Regimen 2 | Cohort B |
|---|---|---|---|
| Adverse Event | 0 | 0 | 2 |
| Physician Decision | 3 | 0 | 8 |
| Progressive Disease | 0 | 0 | 3 |
| Participant/Guardian Decision | 2 | 1 | 11 |
| Lost to Follow-up | 0 | 0 | 1 |
| Death | 0 | 1 | 0 |
| Completed study treatment/discontinued study for unspecified reasons | 0 | 0 | 2 |
Analysis population for the primary outcome, as defined in the registry: The Number of Participants Analyzed reflects the Pharmacokinetic Analysis Set 1 (PAS1), which comprised all participants who provided an evaluable PK profile. For each outcome, the Number Analyzed represents the subset of participants in PAS1 with available and evaluable data.
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.
Population: Full analysis Set (FAS): FAS comprised all patients to whom study treatment had been assigned.
| Characteristic | Unit | Cohort A: Regimen 1 | Cohort A: Regimen 2 | Cohort B | Total |
|---|---|---|---|---|---|
| Participants analysed | 10 | 4 | 37 | 51 | |
| Age, Continuous | Years | 12.0 (4.0 to 17.0) | 10.0 (7.0 to 12.0) | 10.0 (2.0 to 17.0) | 10.0 (2.0 to 17.0) |
| Sex: Female, Male: Female | Participants | 5 | 1 | 17 | 23 |
| Sex: Female, Male: Male | Participants | 5 | 3 | 20 | 28 |
| Race/Ethnicity, Customized: Caucasian | Participants | 4 | 2 | 24 | 30 |
| Race/Ethnicity, Customized: Black | Participants | 2 | 1 | 4 | 7 |
| Race/Ethnicity, Customized: Asian | Participants | 3 | 1 | 9 | 13 |
| Race/Ethnicity, Customized: Unknown | Participants | 1 | 0 | 0 | 1 |
6. Primary Endpoint
The registry lists the primary outcome as Eltrombopag PK Parameters: AUCtau, AUClast, measured over at least 11 weeks after dose initiation or later when patients are taking the highest dose, up to Week 78.
| Element | Registry entry |
|---|---|
| Result type | Geometric Mean |
| Measure of spread | Geometric Coefficient of Variation |
| Unit | hr*ng/mL |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is listed as eltrombopag pharmacokinetic parameters AUCtau and AUClast, with units of hr*ng/mL. AUCtau is the area under the concentration curve through the end of a dosing interval. AUClast is the area under the curve through the last quantifiable concentration point, Tlast. Both measures represent drug exposure over time rather than a clinical response such as a blood-count change.
Pharmacokinetic measures are used to characterize how drug exposure varies under a dosing regimen. They can help describe exposure at the assessed dose and sampling period and can inform interpretation of dose escalation. As a general statistical point, concentration-derived exposure measures often have positive, potentially skewed distributions; a geometric mean can provide a summary on a multiplicative scale when that approach is appropriate. The registry specifies a geometric mean and geometric coefficient of variation for these endpoints.
The planned assessment was at least 11 weeks after dose initiation, or later when participants were taking the highest dose, up to Week 78. The registry record does not report numerical AUCtau or AUClast estimates, their geometric coefficients of variation, or other uncertainty measures. It therefore does not support a quantitative description of the exposure observed in this study.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The primary endpoint identifies what the investigators intended to characterize: eltrombopag exposure measured using AUCtau and AUClast. The registry lists the units as hr*ng/mL, the summary as a geometric mean and the dispersion measure as a geometric coefficient of variation. However, the numerical endpoint results are not reported in the registry record. Without those values, the magnitude and variability of measured exposure cannot be described, and no estimate of precision can be discussed.
The study enrolled 51 participants and is listed as completed. Enrollment does not indicate how many participants contributed evaluable pharmacokinetic profiles. The registry defines the pharmacokinetic analysis set as participants who provided an evaluable profile, with the number analyzed for an outcome limited to those with available and evaluable data. The counts analyzed for AUCtau and AUClast are not reported in the registry record.
The study was open-label, non-controlled and non-randomized. Its intra-patient dose-escalation approach and cohort treatment descriptions are relevant to describing the setting in which pharmacokinetic measurements were collected. They do not create a randomized comparison. In particular, the record does not support interpreting differences between cohorts as the effect of eltrombopag alone: treatment descriptions differ in whether hATG was included, and the registry does not report an adjusted or causal cohort comparison.
The listed secondary outcomes include overall response and platelet response, blood counts, red-cell and platelet transfusions, and duration of red-cell transfusion independence. These outcomes describe clinically relevant domains, but their numerical results are not reported in the registry record. They cannot be used to draw a quantitative conclusion about response or transfusion outcomes from the information available.
For safety, the registry reports on-treatment serious adverse events affecting 29 participants in the total-participant entry, and other adverse events affecting 51 participants in that entry. These are event-count summaries, not estimates of comparative risk. No on-treatment deaths are listed in the total-participant entry. The record lists one post-treatment death in Cohort A Regimen 2 and in the corresponding Cohort A and all-participant aggregate entries; the relationship of that death to treatment is not reported. The listed serious-event terms and percentages should be read with the denominator and event-level details not reported in the registry record.
Overall, the registry record describes the study’s aims, design and selected safety counts, but it does not report the primary pharmacokinetic results or numerical secondary efficacy outcomes. The available information therefore supports a description of what was measured and how the study was organized, not a quantitative conclusion about exposure or clinical benefit.
9. Statistical Methods Explained
What does an area-under-the-curve measure represent?
In general, an area-under-the-curve measure summarizes concentration over time. It combines information across the sampling interval rather than relying on a single concentration measurement. AUCtau uses the end of the dosing interval as its boundary, while AUClast uses the last quantifiable concentration point. Their units, hr*ng/mL, reflect concentration accumulated over time. Neither endpoint is itself a measure of clinical response.
The registry identifies the geometric mean as the summary for these pharmacokinetic parameters. In general, a geometric mean is calculated on a logarithmic scale and then expressed on the original scale. It is often used for positive measurements that vary multiplicatively or are skewed. The geometric coefficient of variation describes relative variability on a multiplicative scale; it is not the same as a confidence interval and does not, by itself, show the precision of the estimated mean. The numerical geometric means and coefficients of variation are not reported in the registry record.
What can be learned from the analysis population?
The registry defines Pharmacokinetic Analysis Set 1 as participants who provided an evaluable pharmacokinetic profile. For each outcome, the analyzed number is the subset of that set with available and evaluable data. This distinction matters because enrollment and analysis counts need not be the same: an enrolled participant may not contribute usable data for every endpoint. The count analyzed for each primary endpoint is not reported in the registry record, so the extent of missing or nonevaluable pharmacokinetic data cannot be quantified.
In general, an analysis population definition clarifies who contributes to a particular estimate. It does not explain why data are unavailable, whether missingness could be related to exposure or clinical status, or how such missingness was handled. Those details are not reported in the registry record.
Why does the design matter for interpretation?
Randomization is a general design method for making treatment groups more comparable, on average, with respect to measured and unmeasured characteristics. This study was non-randomized and non-controlled, so its cohort descriptions should not be treated as randomized comparator arms. The registry describes intra-patient dose escalation, but does not report a controlled estimate of treatment effect or a statistical comparison between regimens.
In general, repeated measurements within a participant can help describe how measurements vary across that participant’s treatment course. They do not automatically remove the influence of time, concurrent therapy or changes in a participant’s condition. The registry record does not report the specific model used to account for repeated observations, dose, time or concomitant treatment. It is therefore not possible to identify from the record how those sources of variation were handled analytically.
What does uncertainty mean for an estimate?
In general, a confidence interval describes a range of values compatible with the data and the method used, under the assumptions of that method. Its width conveys precision: a wider interval indicates less precise estimation than a narrower interval. A point estimate without an interval does not show the full uncertainty around the estimate. The registry record does not report confidence intervals or other uncertainty intervals for the primary pharmacokinetic endpoints.
General inferential methods also depend on assumptions about the data and model. For pharmacokinetic parameters, distributions may be skewed, and a logarithmic-scale analysis may be appropriate in some settings. The registry reports geometric summaries but does not describe a model, transformation, variance estimation method or diagnostic assessment. These methods are therefore not inferable from the endpoint summary alone.
What do planning illustrations mean?
In general, a sample-size calculation estimates how much information a planned study might need under stated assumptions. It is not an observed result and does not establish the precision achieved by a completed study. The planning assumptions associated with the posted illustrations specify a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation and an effect defined as a stated fraction of the observed effect. For time-to-event outcomes, the illustration uses a total number of events; other illustrations express participants per arm. These are planning conventions, not evidence that this trial used randomized equal allocation or had a particular observed power.
The record states that the planning illustration is based on the observed effect and is not the trial’s own design. Because this study was non-randomized and non-controlled, the illustration should not be read as changing the design or creating a like-for-like comparison. No planning result or numerical sample-size illustration is reported in the registry record.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists 28 outcome measures and 0 analyses. The primary outcomes are AUCtau and AUClast. Listed secondary outcomes include overall response and platelet response, hematologic counts, red-cell and platelet transfusions, and duration of red-cell transfusion independence. The registry record does not report a multiplicity adjustment, a hierarchy for testing outcomes, or a plan for selecting among multiple endpoints. Without that information, any interpretation of multiple statistical tests would require caution; the record does not establish that such tests were conducted.
The primary pharmacokinetic analysis population is described as participants with an evaluable profile, with each endpoint using the subset with available and evaluable data. This defines the intended source of observations but does not give the analyzed count for either primary endpoint. Counts of missing or nonevaluable data, reasons for missingness and methods for handling missing data are not reported in the registry record.
The registry does not report whether analyses used an intention-to-treat, per-protocol or other efficacy population for the secondary outcomes. Nor does it report an imputation method or sensitivity analysis for missing measurements. These omissions limit assessment of how exclusions and incomplete follow-up may have affected estimates. They do not establish that data were missing or that any particular method was used; those details are simply not reported in the registry record.
In general, multiplicity concerns arise when several outcomes or statistical tests are considered, because the chance of at least one apparently notable result can increase across tests. A prespecified testing strategy or adjustment can address that issue, but neither is described in the registry record. With numerical outcome estimates and analysis details also not reported, the registry information does not permit evaluation of the statistical evidence across the listed outcomes.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Cohort A - Regimen 1 (On-treatment) | 4 | 10 | 40.0% |
| Cohort A - Regimen 2 (On-treatment) | 2 | 4 | 50.0% |
| Total Cohort A (Regiments 1 & 2) (On-treatment) | 6 | 14 | 42.9% |
| Cohort B (On-treatment) | 23 | 37 | 62.2% |
| Total Participants (On-treatment) | 29 | 51 | 56.9% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Cohort A - Regimen 1 (On-treatment) | 0 of 10 |
| Cohort A - Regimen 2 (On-treatment) | 0 of 4 |
| Total Cohort A (Regiments 1 & 2) (On-treatment) | 0 of 14 |
| Cohort B (On-treatment) | 0 of 37 |
| Total Participants (On-treatment) | 0 of 51 |
| Cohort A - Regimen 1 (Post--treatment) | 0 of 10 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Cohort A - Regimen 1 (On-treatment) | 10 of 10 |
| Cohort A - Regimen 2 (On-treatment) | 4 of 4 |
| Total Cohort A (Regiments 1 & 2) (On-treatment) | 14 of 14 |
| Cohort B (On-treatment) | 37 of 37 |
| Total Participants (On-treatment) | 51 of 51 |
| Cohort A - Regimen 1 (Post--treatment) | 0 of 0 |
Most frequent serious adverse event terms
| Event term | Organ system | Cohort A - Regimen 1 (On-treatment) | Cohort A - Regimen 2 (On-treatment) | Total Cohort A (Regiments 1 & 2) (On-treatment) | Cohort B (On-treatment) | Total Participants (On-treatment) | Cohort A - Regimen 1 (Post--treatment) |
|---|---|---|---|---|---|---|---|
| Pyrexia | General disorders | 2 of 10 | 1 of 4 | 3 of 14 | 8 of 37 | 11 of 51 | n/a of 0 |
| Febrile neutropenia | Blood and lymphatic system disorders | 1 of 10 | 1 of 4 | 2 of 14 | 5 of 37 | 7 of 51 | n/a of 0 |
| Acute kidney injury | Renal and urinary disorders | 0 of 10 | 1 of 4 | 1 of 14 | 2 of 37 | 3 of 51 | n/a of 0 |
| Gastroenteritis | Infections and infestations | 0 of 10 | 1 of 4 | 1 of 14 | 1 of 37 | 2 of 51 | n/a of 0 |
| Staphylococcal infection | Infections and infestations | 0 of 10 | 1 of 4 | 1 of 14 | 1 of 37 | 2 of 51 | n/a of 0 |
| Fatigue | General disorders | 0 of 10 | 0 of 4 | 0 of 14 | 2 of 37 | 2 of 51 | n/a of 0 |
| Serum sickness | Immune system disorders | 0 of 10 | 0 of 4 | 0 of 14 | 2 of 37 | 2 of 51 | n/a of 0 |
| Stomatitis | Gastrointestinal disorders | 0 of 10 | 1 of 4 | 1 of 14 | 0 of 37 | 1 of 51 | n/a of 0 |
| Drug-induced liver injury | Hepatobiliary disorders | 0 of 10 | 1 of 4 | 1 of 14 | 0 of 37 | 1 of 51 | n/a of 0 |
| Clostridium difficile infection | Infections and infestations | 0 of 10 | 1 of 4 | 1 of 14 | 0 of 37 | 1 of 51 | n/a of 0 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Cohort A - Regimen 1 (On-treatment) | Cohort A - Regimen 2 (On-treatment) | Total Cohort A (Regiments 1 & 2) (On-treatment) | Cohort B (On-treatment) | Total Participants (On-treatment) | Cohort A - Regimen 1 (Post--treatment) |
|---|---|---|---|---|---|---|---|
| Alanine aminotransferase increased | Investigations | 7 of 10 | 1 of 4 | 8 of 14 | 16 of 37 | 24 of 51 | n/a of 0 |
| Blood bilirubin increased | Investigations | 6 of 10 | 0 of 4 | 6 of 14 | 18 of 37 | 24 of 51 | n/a of 0 |
| Aspartate aminotransferase increased | Investigations | 6 of 10 | 1 of 4 | 7 of 14 | 13 of 37 | 20 of 51 | n/a of 0 |
| Blood creatinine increased | Investigations | 5 of 10 | 2 of 4 | 7 of 14 | 13 of 37 | 20 of 51 | n/a of 0 |
| Vomiting | Gastrointestinal disorders | 1 of 10 | 2 of 4 | 3 of 14 | 18 of 37 | 21 of 51 | n/a of 0 |
| Hypertension | Vascular disorders | 4 of 10 | 1 of 4 | 5 of 14 | 15 of 37 | 20 of 51 | n/a of 0 |
| Nausea | Gastrointestinal disorders | 2 of 10 | 1 of 4 | 3 of 14 | 17 of 37 | 20 of 51 | n/a of 0 |
| Hypomagnesaemia | Metabolism and nutrition disorders | 3 of 10 | 1 of 4 | 4 of 14 | 15 of 37 | 19 of 51 | n/a of 0 |
| Abdominal pain | Gastrointestinal disorders | 1 of 10 | 2 of 4 | 3 of 14 | 13 of 37 | 16 of 51 | n/a of 0 |
| Pyrexia | General disorders | 3 of 10 | 1 of 4 | 4 of 14 | 10 of 37 | 14 of 51 | n/a of 0 |
Serious events with the highest arm-level rate: febrile neutropenia (25.0%), stomatitis (25.0%), pyrexia (25.0%), drug-induced liver injury (25.0%), clostridium difficile infection (25.0%).
For on-treatment safety, the total-participant entry lists 29 participants with serious adverse events among 51 at risk, and 51 participants with other adverse events among 51 at risk. These are counts of participants affected as recorded, not counts of individual event episodes, and they do not provide a comparison against a control group. The record lists 0 deaths during on-treatment follow-up in the total-participant entry.
The registry lists Febrile neutropenia, Stomatitis, Pyrexia, Drug-induced liver injury, Clostridium difficile infection and Gastroenteritis at 25.0% each among the serious adverse event terms. The denominator and event-level details underlying these percentages are not reported in the registry record. The terms should therefore be read as the registry’s listed serious-event summaries, not as evidence that each affected the same number of participants or occurred with the same frequency across treatment groups.
For post-treatment entries, the registry lists one death for Cohort A Regimen 2, Total Cohort A and all participants. It lists zero participants at risk in those post-treatment entries. The timing, circumstances and relationship of the death to study treatment are not reported in the registry record. A listed death does not by itself establish that treatment caused it. The safety counts describe reported events; they are not a controlled estimate of comparative safety.
12. What the Result Does, and Does Not, Mean
The record establishes that this completed Phase II study was designed to characterize eltrombopag pharmacokinetics in pediatric patients with severe or recurrent aplastic anemia. It specifies AUCtau and AUClast, their units and summary measures, and the planned assessment timeframe. It also describes the treatment regimens and the pharmacokinetic analysis population.
It does not report numerical AUCtau or AUClast results, their variability, confidence intervals or the number analyzed for each endpoint. Consequently, the registry information does not show the level of eltrombopag exposure measured or how precisely it was estimated. The record also lists secondary clinical outcomes but does not report their numerical results, so it cannot support a quantitative conclusion about blood counts, response or transfusion outcomes.
The non-randomized, non-controlled design means the cohorts are not like-for-like controls. A difference between cohort summaries, if such summaries were available, could not automatically be attributed to eltrombopag because cohort treatment descriptions differ and the record does not report a causal analysis. The design can describe observations under the study regimens, but it does not provide a randomized estimate of treatment effect.
The safety entries report on-treatment adverse-event counts and listed serious-event terms, along with a post-treatment death entry. Those observations describe what the registry records; without a controlled comparator, detailed event information or a reported causal assessment, they do not quantify comparative safety or establish treatment causation.
In short, the registry record clarifies the study question and endpoint framework, but the absence of numerical pharmacokinetic and secondary outcome results prevents assessment of the observed exposure or clinical outcomes. The record supports neither an estimate of treatment benefit nor a conclusion that the treatment had no effect.
13. Why This Trial Matters Statistically
Pharmacokinetic studies address a different statistical question from trials focused primarily on clinical outcomes. Measures such as AUCtau and AUClast describe drug exposure over time. Their interpretation depends on the measurement schedule, the participants with evaluable profiles and the way variability is summarized. This registry record makes those elements of the intended endpoint explicit, including a geometric mean and geometric coefficient of variation.
The study is also a useful example of why design labels matter. It is Phase II, open-label, non-controlled and non-randomized, with intra-patient dose escalation and different cohort treatment descriptions. These features define the scope of interpretation: the study is framed as pharmacokinetic characterization, not a randomized comparison. Distinguishing descriptive exposure measurements from causal treatment effects helps prevent the cohort structure from being read as a controlled efficacy comparison.
The registry defines the pharmacokinetic analysis set using evaluable profiles and available data for each outcome. This makes clear that enrollment and endpoint-specific analysis counts are conceptually different. Reporting those counts and the handling of unavailable measurements would be important for judging the precision and representativeness of pharmacokinetic estimates, but the counts and methods are not reported in the registry record.
Finally, the record illustrates that a stated endpoint and analysis summary are not substitutes for numerical results. Without the AUC estimates and uncertainty measures, readers cannot assess exposure magnitude or precision. The available information is still useful for understanding the study’s purpose, treatment setting and limits, while keeping conclusions within what the registry actually reports.
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 main limitation for statistical interpretation is that numerical results for the primary pharmacokinetic endpoints are not reported in the registry record. The geometric means, geometric coefficients of variation, analyzed counts and uncertainty intervals are unavailable. The secondary outcomes are listed, but their numerical results are not reported either. This prevents assessment of observed exposure, clinical outcomes and precision.
The study was non-randomized and non-controlled. Cohort descriptions differ in whether hATG began on Day 1, and the record does not report a controlled comparison or a method for separating the effects of eltrombopag from concomitant treatment and other group differences. The treatment groups should therefore not be interpreted as randomized comparators.
The registry record does not report detailed pharmacokinetic sampling procedures, the model used to analyze repeated measurements, adjustment variables, missing-data methods, confidence intervals or sensitivity analyses. It also does not report endpoint-specific analysis counts or reasons for unavailable and nonevaluable data. These omissions limit assessment of precision, assumptions and possible selection related to the evaluable-profile requirement.
Safety interpretation is also limited by the absence of a controlled comparison and by unreported denominator and event-level details for the listed serious adverse event percentages. The relationship of the post-treatment death to treatment is not reported in the registry record. These limitations do not imply that a particular bias or data problem occurred; they identify what cannot be assessed from the record.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: A Phase II Dose-escalation Study Characterizing the PK of Eltrombopag in Pediatric Patients With Previously Untreated or Relapsed Severe Aplastic Anemia or Recurrent Aplastic Anemia — NCT03025698.
- Registry record last updated 2026-08-10; status last verified 2026-07.
- Results first posted 2026-08-10.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT03025698 is a completed Phase II, open-label, non-controlled, non-randomized study of oral eltrombopag pharmacokinetics in pediatric patients with severe or recurrent aplastic anemia. Enrollment was 51. The study used intra-patient dose escalation and included Cohort A Regimen 1, Cohort A Regimen 2 and Cohort B, with the registry describing the use of hATG, CsA and eltrombopag in each regimen.
The primary endpoints were AUCtau and AUClast, summarized as geometric means with geometric coefficients of variation. Assessment was planned at least 11 weeks after dose initiation, or later at the highest dose, up to Week 78. Numerical primary and secondary outcome results are not reported in the registry record. Safety entries include on-treatment adverse-event counts and listed serious-event terms; they do not provide a controlled comparison or establish causality.