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 Romiplostim Treatment of Thrombocytopenia in Subjects With Low or Intermediate-1 Risk Myelodysplastic Syndrome (MDS) (NCT00614523), a randomized phase 2 study. The primary outcome is Number of Clinically Significant Bleeding Events.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00614523 |
| Phase | Phase 2 |
| Condition | MDS, Myelodysplastic Syndromes, Thrombocytopenia |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 250 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Amgen |
| Results posted | Yes |
2. Clinical Question
Population
Participants with thrombocytopenia and low or intermediate-1 risk myelodysplastic syndrome were eligible. The primary analysis population is described as the full analysis set, including all randomized patients. The registry reports enrollment of 250; the primary outcome results list 83 participants in the placebo arm and 167 in the romiplostim arm.
Intervention
Romiplostim was given by weekly subcutaneous dosing based on platelet count. The Test Treatment Period was 26 weeks. The arm description also includes an Extended Treatment Period of 24 weeks, separated from the test period by a 4-week interim washout period. The starting dose was 750 μg, with a maximum of 1000 μg and a minimum of 250 μg.
Comparator
The control group received weekly subcutaneous blinded matching placebo. The placebo schedule was described for the 26-week Test Treatment Period and the 24-week Extended Treatment Period, with a 4-week interim washout period between them.
Primary question
The primary outcome was the number of clinically significant bleeding events during the Test Treatment Period, defined as Weeks 1-26. The registry reports 116 events in the placebo arm and 178 in the romiplostim arm. Because the arm sizes differ, those raw totals alone do not give a like-for-like comparison. The reported comparative analysis used an Anderson-Gill model and estimated a hazard ratio of 0.83, with a 95% confidence interval of 0.66 to 1.05 and p = 0.13. The result therefore needs to be read with its model-based uncertainty, rather than as a comparison of event totals alone.
3. Trial Design
This was a randomized, double-blind, parallel-group, placebo-controlled Phase 2 study sponsored by Amgen. The registry lists 250 participants as enrolled. The primary outcome population is the full analysis set, defined in the record as all randomized patients. The arm-level primary results list 83 participants for placebo and 167 for romiplostim. The registry record does not report the reason for this difference in group size or further details about the randomization procedure.
Romiplostim and placebo were administered weekly by subcutaneous dosing. The Test Treatment Period lasted 26 weeks; the primary outcome was measured during Weeks 1-26. The intervention descriptions also specify a 24-week Extended Treatment Period, separated from the test period by a 4-week interim washout period. For romiplostim, dosing was based on platelet count, with a starting dose of 750 μg, a maximum of 1000 μg, and a minimum of 250 μg. The placebo was described as a blinded matching dose level.
The registry lists the study as completed, with a start date of 2008-07-21 and a primary completion date of 2011-03-31. Its brief summary states that the Data Monitoring Committee recommended that all subjects discontinue study drug and continue long-term follow-up, and that Amgen adopted the recommendation. The record does not report the timing or rationale for that recommendation, how it affected follow-up or endpoint observation, or whether any additional protocol changes occurred. Further details about the randomization process, allocation concealment, and adherence are 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.
Overall Study
| Stage | Placebo | Romiplostim |
|---|---|---|
| STARTED | 83 | 167 |
| COMPLETED | 20 | 36 |
| NOT COMPLETED | 63 | 131 |
Reasons for not completing this period:
| Reason | Placebo | Romiplostim |
|---|---|---|
| Adverse Event | 4 | 20 |
| Death | 5 | 8 |
| Lack of Efficacy | 9 | 11 |
| Lost to Follow-up | 0 | 1 |
| Protocol Violation | 3 | 2 |
| Withdrawal by Subject | 12 | 22 |
| Other | 6 | 11 |
| Ineligibility Determined | 1 | 2 |
| Disease Progression | 0 | 8 |
| Administrative Decision | 23 | 46 |
Analysis population for the primary outcome, as defined in the registry: Full analysis set includes all randomized patients.
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.
| Characteristic | Unit | Placebo | Romiplostim | Total |
|---|---|---|---|---|
| Participants analysed | 83 | 167 | 250 | |
| Age, Continuous | years | 67 (SD 11.5) | 68.4 (SD 12) | 67.9 (SD 11.8) |
| Sex: Female, Male: Female | Participants | 30 | 72 | 102 |
| Sex: Female, Male: Male | Participants | 53 | 95 | 148 |
| Race/Ethnicity, Customized: White or Caucasian | Participants | 79 | 156 | 235 |
| Race/Ethnicity, Customized: Black or African American | Participants | 0 | 1 | 1 |
| Race/Ethnicity, Customized: Hispanic or Latino | Participants | 1 | 4 | 5 |
| Race/Ethnicity, Customized: Asian | Participants | 1 | 1 | 2 |
| Race/Ethnicity, Customized: Other | Participants | 2 | 5 | 7 |
| Myelodysplastic Syndromes World Health Organization Classification: RA | Participants | 5 | 6 | 11 |
| Myelodysplastic Syndromes World Health Organization Classification: RARS | Participants | 0 | 2 | 2 |
| Myelodysplastic Syndromes World Health Organization Classification: RAEB-1 | Participants | 9 | 24 | 33 |
| Myelodysplastic Syndromes World Health Organization Classification: RAEB-2 | Participants | 0 | 1 | 1 |
| Myelodysplastic Syndromes World Health Organization Classification: RCMD | Participants | 55 | 114 | 169 |
| Myelodysplastic Syndromes World Health Organization Classification: RCMD-RS | Participants | 2 | 4 | 6 |
| Myelodysplastic Syndromes World Health Organization Classification: MDS-U | Participants | 12 | 16 | 28 |
| Myelodysplastic Syndromes World Health Organization Classification: MDS associated with isolated del 5Q | Participants | 0 | 0 | 0 |
| Prior MDS Therapy: No | Participants | 70 | 133 | 203 |
| Prior MDS Therapy: Yes | Participants | 13 | 34 | 47 |
| Baseline Platelet Counts | 10^9/L | 21.5 (SD 13) | 22.3 (SD 11.5) | 22.0 (SD 12.0) |
| International Prognostic Scoring System (IPSS) Total Score: 0 | Participants | 23 | 40 | 63 |
| International Prognostic Scoring System (IPSS) Total Score: 0.5 | Participants | 38 | 86 | 124 |
| International Prognostic Scoring System (IPSS) Total Score: 1 | Participants | 20 | 34 | 54 |
| International Prognostic Scoring System (IPSS) Total Score: 1.5 | Participants | 0 | 1 | 1 |
| International Prognostic Scoring System (IPSS) Total Score: Missing | Participants | 2 | 6 | 8 |
6. Primary Endpoint
The registry lists the primary outcome as Number of Clinically Significant Bleeding Events, measured over Test Treatment Period (Weeks 1-26).
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | events |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint was the number of clinically significant bleeding events during the Test Treatment Period, defined as Weeks 1-26. The registry defines a clinically significant event as bleeding of grade 2 or higher on the modified World Health Organization bleeding scale. The listed grades range from no bleeding through fatal bleeding, with grade 2 including melena, hematemesis, hematuria, and hemoptysis. The definition also includes more severe bleeding events, such as bleeding requiring red cell transfusion and cerebral bleeding.
The counting rules matter because the outcome is recurrent: a participant could contribute more than one event. Bleeding that continued for more than 7 days was counted as a separate event every eighth day. Multiple events arising from one organ system on one day were collapsed into one event. Events starting between the first dose date and the last dose date of the test period plus 7 days were included. These rules shape what the reported count represents.
Counting clinically significant bleeding events can describe the burden of recurrent bleeding during the defined treatment period. A count, however, is not automatically comparable across groups without considering exposure and the analysis method. The arm-level totals are 116 for placebo and 178 for romiplostim, but the group sizes differ. The reported comparative result therefore comes from the Anderson-Gill model, not from treating the two event totals as if they arose from equal-sized groups. Details of additional model specifications are not reported in the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
During the Test Treatment Period, the registry reports 116 clinically significant bleeding events in the placebo arm and 178 in the romiplostim arm. These are event totals, not participant counts. The arm sizes were 83 and 167, respectively, so the totals should not be interpreted as a direct comparison of individual participants’ risk or as a simple measure of how often an event occurred per person.
The sponsor-reported primary analysis used an Anderson-Gill model and gave a hazard ratio of 0.83, with a 95% confidence interval from 0.66 to 1.05 and p = 0.13. The point estimate is below 1, but the confidence interval includes 1. In general, a confidence interval describes a range of values compatible with the data under the model and its assumptions; it communicates uncertainty rather than guaranteeing that the true value lies within the interval. The reported p-value also indicates that the model-based result is not especially precise evidence against the null value. The registry record does not report the prespecified significance threshold or further details of the primary analysis plan.
The point estimate and the uncertainty need to be read together. The data do not establish a clear between-group difference in the primary outcome under the reported analysis. That is not the same as demonstrating that the groups were equivalent, nor does the estimate by itself settle the size or direction of any underlying difference. The confidence interval allows values on either side of 1, including values compatible with a lower event hazard in the romiplostim arm and values compatible with little or no difference.
The primary analysis is distinct from the posted secondary outcomes. Several secondary rate comparisons have reported Poisson regression results, but those are separate outcomes and do not replace the primary analysis. Their p-values are unadjusted in the recalculations, and the registry record does not report a multiplicity strategy. The secondary findings should therefore be read as supportive descriptions of those outcomes, not as a substitute for the primary endpoint result or as a single confirmatory conclusion.
Finally, the registry’s brief summary records the Data Monitoring Committee recommendation to stop study drug and continue long-term follow-up. The record does not explain the timing or reason for this recommendation, so it cannot be used to infer how the decision affected the primary outcome or to attribute the event totals to a particular cause. The primary result pertains to the defined test-period endpoint and the reported model, with the limits of those details in mind.
9. Statistical Methods Explained
The primary endpoint is a recurrent event count: participants could experience multiple qualifying bleeding events during the defined period. The registry specifies rules for events lasting more than 7 days, multiple events from one organ system on the same day, and the window in which event start dates were counted. Such rules are important because they determine whether a prolonged episode contributes once or more than once, and how concurrent events are grouped. They also make clear that the outcome is not simply the number of participants who ever bled.
What does the primary model estimate?
The reported primary analysis used an Anderson-Gill model and reported a hazard ratio. In general, this model is used for recurrent-event data and relates event occurrence over follow-up to group assignment while allowing repeated events to contribute. A hazard ratio compares modeled event rates over time under the model’s assumptions. It is not the same as a ratio of the raw event totals, and it does not directly report the number of participants who experienced an event. The registry record does not report further model details, such as covariates, handling of within-participant dependence, or diagnostics.
The reported hazard ratio is 0.83, its 95% confidence interval is 0.66 to 1.05, and p = 0.13. In general, a confidence interval gives a range of estimates compatible with the observed data and the statistical model, subject to the model assumptions. A wider interval indicates greater uncertainty about the estimate. The interval includes 1, the null value for a ratio, so the result is compatible with no difference as well as with values in either direction. A p-value describes how unusual the observed data, or more extreme data, would be under a specified null model; it does not give the probability that the null hypothesis is true.
How are the secondary rates analyzed?
The posted secondary outcomes include annualized rates of platelet transfusion events, overall bleeding events, and total platelet transfusion units. Rates per 100 patient-years express events or units relative to time at risk, rather than only presenting a total count. The registry reports Poisson regression models for these outcomes. In general, Poisson regression is used for count outcomes and can incorporate exposure time, often through a model offset. The registry record does not report the precise model specification or the assumptions and checks used.
Poisson models commonly assume that the conditional variance is tied to the conditional mean. In general, over-dispersion occurs when the observed variability is greater than that assumption allows; if it is present and not appropriately handled, uncertainty estimates and tests may be unreliable. The record does not report an over-dispersion assessment or a correction. This is a general consideration for count-rate analyses, not a finding that over-dispersion occurred in this study.
The posted secondary results include both sponsor-reported Poisson regression risk ratios and unadjusted recalculations from posted summary rates. These are not necessarily identical estimands or analyses. A regression result may reflect the model specification, while a simple rate-based recalculation relies on the posted summary statistics and does not reproduce unreported adjustments. The recalculations are therefore descriptive checks, not replacements for the sponsor’s analysis.
What do planning illustrations represent?
The planning illustrations use the observed effects as hypothetical effects to detect, with two-sided significance level 0.05, equal allocation, and a normal approximation. They describe participants per arm for 80% or 90% power under the listed effect assumptions. In general, sample-size calculations depend on assumptions about the target effect, variability, allocation, and analysis. The illustrations are not the trial’s original design justification, and they do not describe the probability that the completed study detected an effect. The registry record does not report the original sample-size justification.
Some registry outcomes and analyses are not available as complete two-arm numeric comparisons. The primary endpoint is reported as a count, with no arm-level dispersion or confidence limits in the registry results. The primary comparison is nevertheless reported separately through the Anderson-Gill model. The distinction between group-level descriptive counts and a model-based comparison is central: the former describes observed event totals, while the latter estimates a comparative association using a stated analysis method.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Anderson-Gill model | Hazard Ratio (HR) 0.83 | 0.66 to 1.05 | 0.130 |
For the primary endpoint, the registry reports the sponsor’s Anderson-Gill model result: hazard ratio 0.83, 95% confidence interval 0.66 to 1.05, and p = 0.13. The posted primary arm values are event counts without dispersion measures, and a simple unadjusted recalculation is not available for this count outcome. Accordingly, there is no recalculated primary estimate to set against the sponsor’s model-based result. The primary inference should be based on the reported analysis and its uncertainty, while recognizing that model details beyond the method name are not reported in the registry record.
For the secondary rate outcomes, the registry reports Poisson regression analyses, while the recalculations use posted summary rates to obtain unadjusted comparisons. For example, the platelet transfusion events comparison has a recalculated rate ratio of 0.739, while the sponsor-reported risk ratio is 0.766. The two estimates are close in direction but are not identical. The posted results identify different analysis approaches and label the sponsor’s result as a risk ratio, while the recalculation is a rate ratio. Further model details that could explain the difference are not reported in the registry record. The same caution applies to the other secondary comparisons: a simple recalculation is not a reproduction of the sponsor’s regression analysis. Where both are available, the sponsor’s reported analysis remains the primary source for the trial’s own inferential result.
11. Multiplicity, Missing Data and Analysis Populations
The primary outcome population is described as the full analysis set, including all randomized patients. The registry does not provide additional detail about the handling of participants with missing outcome data, incomplete follow-up, treatment discontinuation, or deviations from assigned treatment. The number of participants included in each secondary analysis and the reasons for any differences are not reported in the registry record. It is therefore not possible to assess from the record whether missingness could have affected estimates or uncertainty.
The registry lists multiple secondary outcomes and reports results for several rate measures. It does not report a multiplicity adjustment, an outcome hierarchy, or a distinction between confirmatory and exploratory secondary analyses. When many outcomes are tested, the chance of at least one nominally notable result can increase. In general, adjustment procedures or prespecified testing strategies can address that issue, but neither is described in the registry record. The secondary recalculated p-values are unadjusted and should not be read as controlling error across the family of outcomes.
The primary analysis is identified as an Anderson-Gill model, but the record does not report whether there were interim efficacy analyses, a prespecified stopping boundary, or other sequential monitoring rules. The brief summary does state that the Data Monitoring Committee recommended discontinuing study drug and continuing long-term follow-up, and that Amgen adopted the recommendation. The timing, rationale, and statistical basis of the recommendation are not reported. It should not be assumed to reflect a particular efficacy, safety, or futility finding.
The registry reports that the primary analysis includes randomized patients, but does not describe imputation, censoring rules beyond the endpoint’s event window, or sensitivity analyses. Further statistical-plan details—including the intended primary hypothesis, significance threshold, sample-size justification, and treatment of multiplicity—are not reported in the registry record. These omissions limit assessment of how closely the posted analyses reflect the original protocol and how robust the estimates are to alternative assumptions.
12. 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 |
|---|---|---|---|
| Placebo | 22 | 82 | 26.8% |
| Romiplostim | 67 | 168 | 39.9% |
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 |
|---|---|
| Placebo | 74 of 82 |
| Romiplostim | 146 of 168 |
Most frequent serious adverse event terms
| Event term | Organ system | Placebo | Romiplostim |
|---|---|---|---|
| Anaemia | Blood and lymphatic system disorders | 2 of 82 | 7 of 168 |
| Pyrexia | General disorders | 1 of 82 | 7 of 168 |
| Pneumonia | Infections and infestations | 0 of 82 | 8 of 168 |
| Thrombocytopenia | Blood and lymphatic system disorders | 0 of 82 | 6 of 168 |
| Atrial fibrillation | Cardiac disorders | 1 of 82 | 5 of 168 |
| Epistaxis | Respiratory, thoracic and mediastinal disorders | 1 of 82 | 4 of 168 |
| Febrile neutropenia | Blood and lymphatic system disorders | 2 of 82 | 2 of 168 |
| Gastrointestinal haemorrhage | Gastrointestinal disorders | 1 of 82 | 3 of 168 |
| Head injury | Injury, poisoning and procedural complications | 0 of 82 | 4 of 168 |
| Myocardial infarction | Cardiac disorders | 1 of 82 | 2 of 168 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Placebo | Romiplostim |
|---|---|---|---|
| Epistaxis | Respiratory, thoracic and mediastinal disorders | 32 of 82 | 66 of 168 |
| Haematoma | Vascular disorders | 33 of 82 | 58 of 168 |
| Petechiae | Skin and subcutaneous tissue disorders | 20 of 82 | 42 of 168 |
| Contusion | Injury, poisoning and procedural complications | 10 of 82 | 41 of 168 |
| Gingival bleeding | Gastrointestinal disorders | 14 of 82 | 33 of 168 |
| Headache | Nervous system disorders | 10 of 82 | 29 of 168 |
| Haemorrhage | Vascular disorders | 15 of 82 | 24 of 168 |
| Blood blister | Skin and subcutaneous tissue disorders | 11 of 82 | 25 of 168 |
| Cough | Respiratory, thoracic and mediastinal disorders | 10 of 82 | 25 of 168 |
| Nausea | Gastrointestinal disorders | 7 of 82 | 27 of 168 |
Serious events with the highest arm-level rate: pneumonia (4.8%), anaemia (4.2%), pyrexia (4.2%), thrombocytopenia (3.6%), atrial fibrillation (3.0%).
The registry reports serious adverse events in 22 of 82 participants at risk in the placebo group and 67 of 168 participants at risk in the romiplostim group. Other adverse events are reported in 74 of 82 participants at risk in the placebo group and 146 of 168 participants at risk in the romiplostim group. These are participant counts, not event rates, and the group sizes differ. The counts alone do not establish whether a difference is due to treatment, exposure, underlying illness, or other factors. The record does not provide the adverse-event observation period or further details needed to assess comparability.
The registry lists pneumonia at 4.8%, anaemia at 4.2%, pyrexia at 4.2%, thrombocytopenia at 3.6%, atrial fibrillation at 3.0%, and febrile neutropenia at 2.4% among the top serious terms. The posted information does not identify an arm-specific percentage for each term, so those values should not be assigned to either treatment group. Death counts are not reported in the adverse-event results. These figures describe what the registry lists; they do not by themselves establish causation or comparative safety.
13. What the Result Does, and Does Not, Mean
The main statistical message is that the primary analysis estimated a hazard ratio of 0.83 for clinically significant bleeding events, but uncertainty remains substantial: the 95% confidence interval runs from 0.66 to 1.05, and p = 0.13. The point estimate is below 1, yet the interval includes 1. The posted primary result therefore does not provide a clear, precise estimate of the difference between groups. It is not evidence that the groups are equivalent, and it does not establish the size of any true effect.
The raw event totals—116 in the placebo arm and 178 in the romiplostim arm—should not be read as a simple comparison because the arm sizes were 83 and 167. The primary model was intended to analyze recurrent events, but its full specification is not reported. The model-based result and its uncertainty are more relevant to the stated comparative question than the totals alone.
The secondary rate outcomes add context about transfusion and bleeding measures. Their estimates should be interpreted outcome by outcome, because they are not identical to the primary endpoint. In addition, the registry does not report a multiplicity plan, and recalculated secondary p-values are unadjusted. The secondary pattern should not be treated as independent confirmation of a primary finding.
These results describe the posted registry record, not all possible analyses or information that may exist outside it. Details about missing data, the complete statistical analysis plan, interim analyses, and the rationale for the Data Monitoring Committee recommendation are not reported. The conclusions that can be drawn are therefore limited to the reported population, outcomes, timeframes, estimates, and models. No medical advice or individual treatment decision follows from these statistics alone.
14. Why This Trial Matters Statistically
This trial illustrates why recurrent clinical events need more than a comparison of total counts. Participants can contribute repeated events, and the registry defines rules for prolonged bleeding and events occurring on the same day. Those rules affect the endpoint itself. The arm sizes also differ, so raw totals cannot answer the comparative question on their own. The sponsor’s Anderson-Gill model supplies a model-based comparison, but the estimate must be considered alongside its confidence interval and the limits of the methodological detail in the record.
The primary result is also a useful example of separating a point estimate from precision. A hazard ratio of 0.83 may look directional when viewed alone, but its 95% confidence interval of 0.66 to 1.05 includes the null value. The reported p-value is 0.13. Statistical interpretation depends on the complete pattern of estimate, uncertainty, and analysis assumptions, not on one number in isolation.
The secondary outcomes show another important distinction: sponsor-reported regression estimates and simple unadjusted recalculations need not be identical. Their analysis methods and estimands can differ, and the registry does not include enough detail to reconstruct every model. Planning illustrations based on observed effects are also not the same as the original trial design justification or an assessment of observed power. Together, these features show why clinical trial results require attention to endpoint definitions, analysis populations, model choice, uncertainty, and the registry record’s limits.
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 primary outcome is a recurrent count, but the posted arm-level results do not include dispersion measures or confidence limits. The comparative result is reported through an Anderson-Gill model, yet details of that model beyond its name are not reported in the registry record. In particular, model covariates, handling of within-participant dependence, and diagnostic checks are not described. This limits independent assessment of the primary estimate and its assumptions.
The arm sizes differ: the primary results list 83 participants for placebo and 167 for romiplostim. The registry record does not explain the allocation imbalance. Although the primary analysis population is described as all randomized patients, the record does not report missing-data methods, analysis-specific participant counts, or sensitivity analyses. These omissions make it difficult to evaluate the potential effect of incomplete follow-up or missing outcome information.
The registry’s brief summary says that the Data Monitoring Committee recommended stopping study drug and continuing long-term follow-up, and that Amgen adopted the recommendation. It does not report when or why this occurred or how it affected the duration of treatment and outcome ascertainment. The primary endpoint is limited to the Test Treatment Period (Weeks 1-26), so the posted result does not describe all potential longer-term outcomes.
Finally, the record does not report a multiplicity strategy, the original sample-size justification, interim analysis plans, or the full statistical analysis plan. Several secondary outcomes are listed, while numerical results are not posted for every listed outcome. The safety summary gives participant counts and top serious terms, but the adverse-event observation period and arm-specific percentages for those terms are not reported. These are registry-record limitations, not evidence that the missing methods or results were absent from the full study documentation.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Romiplostim Treatment of Thrombocytopenia in Subjects With Low or Intermediate-1 Risk Myelodysplastic Syndrome (MDS) — NCT00614523.
- Registry record last updated 2022-11-07; status last verified 2022-11.
- Results first posted 2013-08-13.
- 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.
19. Record Summary
NCT00614523 is a completed randomized, double-blind, parallel-group, placebo-controlled Phase 2 study of romiplostim in subjects with low or intermediate-1 risk myelodysplastic syndrome and thrombocytopenia. The registry lists enrollment of 250 and identifies Amgen as the sponsor. The primary outcome was the number of clinically significant bleeding events during Weeks 1-26. The reported Anderson-Gill model estimate was a hazard ratio of 0.83, with a 95% confidence interval of 0.66 to 1.05 and p = 0.13.
The record also reports secondary outcomes relating to bleeding and platelet transfusions, along with adverse-event counts. Its brief summary records a Data Monitoring Committee recommendation to discontinue study drug and continue long-term follow-up, which Amgen adopted. Further details about the recommendation and several statistical methods are not reported in the registry record.