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 Study of Subcutaneous (sc) Mircera in Dialysis Patients With Chronic Renal Anemia. (NCT00364832), a randomized phase 2 study. The primary outcome is Median Change From Baseline in Hemoglobin Levels to End of Initial Treatment Under Constant Dosing Regimen.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00364832 |
| Phase | Phase 2 |
| Condition | Anemia |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 137 |
| Arms in the primary outcome | 9 |
| Lead sponsor | Hoffmann-La Roche |
| Results posted | Yes |
2. Clinical Question
Population
Dialysis patients with chronic renal anemia who had been receiving subcutaneous epoetin alfa or beta. The primary analysis population was the intention-to-treat population, defined in the registry record as all randomized participants. Eligibility criteria beyond this description are not reported in the registry record.
Intervention
Maintenance treatment with subcutaneous RO0503821, identified as methoxy polyethylene glycol-epoetin beta (Mircera). Nine cohorts tested conversion factors of 0.4/150, 0.8/150, or 1.2/150 microgram per kilogram of the previous weekly ESA dose, with administration once weekly, once every three weeks, or once every four weeks.
Comparator
The cohorts represent different Mircera dose-conversion and frequency regimens. The registry record does not describe a placebo or untreated control group, and it does not identify a single cohort as the comparator for the primary endpoint. The sponsor’s reported conversion-factor analyses group cohorts by administration frequency.
Primary question
What median change in hemoglobin occurred from baseline to the end of initial treatment under a constant dosing regimen in each Mircera cohort? The study also reported estimated conversion factors by administration frequency. The registry record does not state a single formal hypothesis comparing the nine regimens.
3. Trial Design
This was a randomized, open-label, multicenter, parallel phase 2 study sponsored by Hoffmann-La Roche. The registry lists 137 participants enrolled and the study status as completed. The recorded study start was 2001-10, and primary completion was 2005-07. The brief summary describes an anticipated time on study treatment of 3-12 months and a target sample size of 100-500 individuals; the actual enrollment was 137.
The study concerned maintenance treatment after switching from subcutaneous epoetin alfa or beta to subcutaneous RO0503821, also called Mircera. Participants were assigned to nine cohorts crossing three conversion factors with three dosing frequencies. The conversion factors were 0.4/150, 0.8/150, and 1.2/150 microgram per kilogram of the previous weekly ESA dose. The cohort descriptions characterize these as 50%, 100%, and 150% of the assumed equi-effective dose, respectively. Each conversion factor was used with weekly administration, administration once every three weeks, or administration once every four weeks.
The initial treatment endpoint was assessed at Week 19 or Week 21, depending on the regimen. The cohort descriptions specify core treatment up to 19 weeks for several regimens and up to 21 weeks for the four-week schedules; follow-up is also mentioned for several cohorts. The description for Cohort A is incomplete, and the full follow-up arrangements are not consistently available in the registry record. Detailed allocation procedures, eligibility criteria, planned enrollment by cohort, and the rationale for the selected regimens are not reported in the registry record. Masking was none, so the study was open label.
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.
Only the first 6 of 12 groups are shown.
Core Period
| Stage | Cohort A (0.4/150, 1x/ Week) | Cohort B (0.4/150, 1x/ 3 Weeks) | Cohort C (0.4/150, 1x/ 4 Weeks) | Cohort D (0.8/150, 1x/ Week) | Cohort E (0.8/150, 1x/ 3 Weeks) | Cohort F (0.8/150, 1x/ 4 Weeks) |
|---|---|---|---|---|---|---|
| STARTED | 16 | 15 | 15 | 16 | 16 | 12 |
| COMPLETED | 14 | 13 | 14 | 15 | 16 | 12 |
| NOT COMPLETED | 2 | 2 | 1 | 1 | 0 | 0 |
Reasons for not completing this period:
| Reason | Cohort A (0.4/150, 1x/ Week) | Cohort B (0.4/150, 1x/ 3 Weeks) | Cohort C (0.4/150, 1x/ 4 Weeks) | Cohort D (0.8/150, 1x/ Week) | Cohort E (0.8/150, 1x/ 3 Weeks) | Cohort F (0.8/150, 1x/ 4 Weeks) |
|---|---|---|---|---|---|---|
| Early Withdrawals | 2 | 2 | 1 | 1 | 0 | 0 |
Extension Year 1
| Stage | Cohort A (0.4/150, 1x/ Week) | Cohort B (0.4/150, 1x/ 3 Weeks) | Cohort C (0.4/150, 1x/ 4 Weeks) | Cohort D (0.8/150, 1x/ Week) | Cohort E (0.8/150, 1x/ 3 Weeks) | Cohort F (0.8/150, 1x/ 4 Weeks) |
|---|---|---|---|---|---|---|
| STARTED | 0 | 0 | 0 | 0 | 0 | 0 |
| COMPLETED | 0 | 0 | 0 | 0 | 0 | 0 |
| NOT COMPLETED | 0 | 0 | 0 | 0 | 0 | 0 |
Reasons for not completing this period:
| Reason | Cohort A (0.4/150, 1x/ Week) | Cohort B (0.4/150, 1x/ 3 Weeks) | Cohort C (0.4/150, 1x/ 4 Weeks) | Cohort D (0.8/150, 1x/ Week) | Cohort E (0.8/150, 1x/ 3 Weeks) | Cohort F (0.8/150, 1x/ 4 Weeks) |
|---|---|---|---|---|---|---|
| Early Withdrawals | 0 | 0 | 0 | 0 | 0 | 0 |
Extension Year 2
| Stage | Cohort A (0.4/150, 1x/ Week) | Cohort B (0.4/150, 1x/ 3 Weeks) | Cohort C (0.4/150, 1x/ 4 Weeks) | Cohort D (0.8/150, 1x/ Week) | Cohort E (0.8/150, 1x/ 3 Weeks) | Cohort F (0.8/150, 1x/ 4 Weeks) |
|---|---|---|---|---|---|---|
| STARTED | 0 | 0 | 0 | 0 | 0 | 0 |
| COMPLETED | 0 | 0 | 0 | 0 | 0 | 0 |
| NOT COMPLETED | 0 | 0 | 0 | 0 | 0 | 0 |
Reasons for not completing this period:
| Reason | Cohort A (0.4/150, 1x/ Week) | Cohort B (0.4/150, 1x/ 3 Weeks) | Cohort C (0.4/150, 1x/ 4 Weeks) | Cohort D (0.8/150, 1x/ Week) | Cohort E (0.8/150, 1x/ 3 Weeks) | Cohort F (0.8/150, 1x/ 4 Weeks) |
|---|---|---|---|---|---|---|
| Early Withdrawals | 0 | 0 | 0 | 0 | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: The ITT population was considered for analysis which included all randomized participants.
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: The Intent-to-Treat (ITT) population was considered for analysis which included all randomized participants.
Only the first 6 of 10 groups are shown.
| Characteristic | Unit | Cohort A (0.4/150, 1x/ Week) | Cohort B (0.4/150, 1x/ 3 Weeks) | Cohort C (0.4/150, 1x/ 4 Weeks) | Cohort D (0.8/150, 1x/ Week) | Cohort E (0.8/150, 1x/ 3 Weeks) | Cohort F (0.8/150, 1x/ 4 Weeks) |
|---|---|---|---|---|---|---|---|
| Participants analysed | 16 | 15 | 15 | 16 | 16 | 12 | |
| Age, Continuous | years | 67.0 (SD 13.03) | 65.6 (SD 11.29) | 64.8 (SD 10.28) | 61.8 (SD 13.51) | 64.1 (SD 12.83) | 67.3 (SD 13.30) |
| Gender: Female | Participants | 5 | 8 | 5 | 6 | 5 | 7 |
| Gender: Male | Participants | 11 | 7 | 10 | 10 | 11 | 5 |
6. Primary Endpoint
The registry lists the primary outcome as Median Change From Baseline in Hemoglobin Levels to End of Initial Treatment Under Constant Dosing Regimen, measured over From Baseline (Day -28 to Day 1) to EOIT (Week 19 or Week 21).
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | Inter-Quartile Range |
| Unit | gram per deciliter |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint was the median change in hemoglobin from baseline to the end of initial treatment under a constant dosing regimen, measured in gram per deciliter. Hemoglobin is the laboratory measure used for this outcome; the endpoint summarizes how its measured level changed during the specified treatment period. A median describes the middle of the observed changes rather than their arithmetic average. The registry also identifies the inter-quartile range as the dispersion measure, although the posted cohort results list no spread value.
Baseline was defined using the mean of the screening assessment and the run-in period at Week -2 and Week -1, within the stated baseline interval of Day -28 to Day 1. The end-of-initial-treatment value was the last observed hemoglobin value before a dose change or blood transfusion. For participants without dose adjustments or transfusion, this value was the Week 19 or Week 21 value. The registry says that individual slope values were multiplied by 42 to estimate change over six weeks, a transformation described as being for ease of interpretation.
This endpoint is useful for describing hemoglobin changes during maintenance dosing, while the median can be less influenced by unusually large or small individual changes than the mean. It does not by itself establish why a change occurred or whether one regimen is superior to another. The registry does not report a formal comparison test for the cohort medians.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted primary results are medians for each of the nine dose-and-frequency cohorts. The median change was -0.8 gram per deciliter in Cohort A, -0.6 in Cohort B, and -0.29 in Cohort C, all using the 0.4/150 conversion factor. At 0.8/150, the medians were 0.02 in Cohort D, -0.14 in Cohort E, and -0.28 in Cohort F. At 1.2/150, they were 0.65 in Cohort G, 0.33 in Cohort H, and 0.22 in Cohort I. Cohort sizes ranged from 12.0 to 16.0 in these primary results.
These are within-cohort summaries, not direct estimates of differences between regimens. The reported interval bounds vary by cohort: for example, Cohort A has bounds of -1.6 and -0.36, while Cohort G has bounds of 0.12 and 1.57. The registry does not label the interval type for these cohort bounds in the primary outcome record. They should therefore be read as reported bounds, not assigned an unreported confidence level. The bounds also do not replace a direct randomized comparison or a test of a prespecified hypothesis.
The sponsor reported estimated conversion factors with 90% confidence intervals, grouped across cohorts by administration frequency. For weekly administration, the estimate was 0.81 with a 90% confidence interval from 0.73 to 0.90. For administration once every three weeks, it was 0.72 with a 90% confidence interval from 0.55 to 0.86. For administration once every four weeks, it was 0.93 with a 90% confidence interval from 0.81 to 1.09. These estimates are distinct from the individual cohort median hemoglobin changes. The registry does not report p-values for these conversion-factor analyses.
The broad pattern in the cohort medians is descriptive: the lowest conversion-factor cohorts have negative median changes, the intermediate conversion-factor medians are near zero or negative, and the highest conversion-factor medians are positive. That pattern alone cannot show a dose-response relationship or establish that frequency has no influence. Cohort sizes are modest, and the record does not report a formal model or an adjusted comparison for the primary endpoint. Interpretation should remain centered on the estimates, their uncertainty, and the distinction between cohort descriptions and the sponsor’s grouped conversion-factor analyses.
9. Statistical Methods Explained
What does a median describe?
The median is the central value when observed participant changes are ordered. It gives a location summary that is less sensitive than a mean to extreme observations, but it does not describe every participant’s response or the full shape of the data. The registry identifies the inter-quartile range as the dispersion measure for the primary endpoint. That range would describe the middle portion of the observed changes, but no dispersion values are posted for the primary cohort results. The reported cohort medians should therefore be read as summaries, not as complete descriptions of variation.
How should the interval bounds be read?
The primary cohort results include lower and upper bounds, but the registry record does not identify their confidence level or provide a method for their calculation. A confidence interval, in general, expresses uncertainty around an estimate under a stated statistical procedure; it is not a guarantee that the true value lies inside the interval. Because the interval type is not specified for these primary results, the bounds should not be described as a particular confidence interval. By contrast, the conversion-factor analyses explicitly report 90% confidence intervals. These provide a range of values compatible with the analysis under its assumptions, and the interval width conveys precision. The method used to calculate those intervals is not reported in the registry record.
What is the distinction between cohort results and conversion-factor analyses?
The primary outcome lists median hemoglobin changes separately in each of the nine cohorts. The sponsor’s reported conversion-factor analyses instead group cohorts by frequency: weekly, once every three weeks, and once every four weeks. Each reported estimate is an estimated conversion factor with a 90% confidence interval. These are different summaries with different meanings. They should not be treated as though the conversion-factor estimate were a direct comparison of the cohort medians, or as though the individual cohort results directly reproduce the sponsor’s grouped analysis. The analysis method and the precise estimand for the conversion-factor estimates are not reported in the registry record.
What do the recalculated pulse-rate comparisons add?
The secondary pulse-rate results include simple unadjusted recalculations comparing Cohort A with Cohorts B and C. A mean difference is the difference between the two posted group means, and its interval communicates uncertainty under the recalculation’s assumptions. The comparison of Cohort A with Cohort B has a recalculated mean difference of 4.0 beats per minute, with bounds from -2.309 to 10.31 and an unadjusted p-value of 0.214. The comparison with Cohort C has a recalculated mean difference of 13.0 beats per minute, with bounds from 6.691 to 19.31 and an unadjusted p-value of <0.001. These recalculations are not the sponsor’s reported primary analysis, and they are not adjusted for multiple testing.
The associated planning illustrations translate an observed mean difference into hypothetical sample sizes under stated assumptions. They use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the target effect set to a stated fraction of the observed effect. For the Cohort A versus Cohort B comparison, planning for 100% of the observed effect gives 79 participants per arm for 80% power and 106 participants per arm for 90% power. Planning for 75% gives 140 and 187 participants per arm, respectively; planning for 50% gives 315 and 421 participants per arm. For the Cohort A versus Cohort C comparison, the corresponding values are 8 and 10 for 100%, 14 and 18 for 75%, and 30 and 40 for 50%. These are planning illustrations, not observed power and not evidence about the original trial’s sample-size justification.
Why does over-dispersion matter in general?
In general, over-dispersion means that observed variability is greater than a statistical model expects. If a model assumes less variability than the data contain, standard errors can be too small and intervals too narrow, making apparent precision misleading. This concern is relevant to count or event outcomes when a model is used, but the registry record does not describe a model for the safety summaries or report an over-dispersion assessment. It is therefore a general statistical consideration, not a finding about this trial.
What statistical plan is available?
The registry record reports the endpoint definitions, analysis population, cohort summaries, conversion-factor estimates, and selected secondary outcomes. It does not report the sample-size justification, a complete statistical analysis plan, interim analyses, or the methods used for the sponsor’s conversion-factor estimates. The record also does not identify a multiplicity-control strategy. These omissions limit how far the results can be interpreted as confirmatory evidence.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| not stated | Estimated Conversion Factor 90% CI 0.81 | 0.73 to 0.90 | not reported |
| not stated | Estimated Conversion Factor 90% CI 0.72 | 0.55 to 0.86 | not reported |
| not stated | Estimated Conversion Factor 90% CI 0.93 | 0.81 to 1.09 | not reported |
The sponsor’s reported analysis concerns estimated conversion factors, with separate estimates for the weekly, once-every-three-weeks, and once-every-four-weeks administration groups. The estimates were 0.81, 0.72, and 0.93, respectively, with 90% confidence intervals of 0.73 to 0.90, 0.55 to 0.86, and 0.81 to 1.09. No p-values or analysis method are reported for these estimates. The registry does not explain how the estimates were derived or how the grouped results relate to the individual cohort summaries.
The recalculated results concern a different secondary outcome: mean change in pulse rate. They compare Cohort A with Cohort B and Cohort C using posted means and standard deviations. They do not reproduce or validate the sponsor’s conversion-factor estimates, and they should not be interpreted as a reanalysis of the primary endpoint. The primary cohort medians are also not a direct recalculation of the sponsor’s grouped conversion-factor analysis. Because the estimands and outcomes differ, the analyses answer different descriptive questions rather than competing versions of the same test. The registry record does not report a discrepancy between sponsor results and recalculated results.
11. Multiplicity, Missing Data and Analysis Populations
The study has multiple cohorts, a primary endpoint, several secondary outcomes, and sponsor-reported conversion-factor analyses. Considering many endpoints or comparisons increases the chance that some results may appear noteworthy by chance, even when there is no underlying difference. In general, multiplicity can be handled by prespecifying a primary contrast, adjusting significance thresholds, controlling an error rate, or treating secondary analyses as exploratory. The registry record does not report a multiplicity strategy, a hierarchy among secondary outcomes, or which cohort comparisons were prespecified.
The pulse-rate recalculations are explicitly unadjusted. Their p-values should therefore be interpreted as unadjusted evidence for those particular comparisons, not as accounting for all comparisons made across cohorts and outcomes. The registry’s conversion-factor analyses have 90% confidence intervals but no reported p-values or method. Whether those analyses were prespecified and how they fit into the overall testing plan are not reported in the registry record.
The primary endpoint population is described as the intention-to-treat population, including all randomized participants. The endpoint value is defined using the last observed hemoglobin before a dose change or blood transfusion; when neither occurred, the Week 19 or Week 21 value was used. Baseline was calculated from the mean of the screening assessment and run-in measurements. These rules explain which measurements contributed to the endpoint, but they do not describe how missing measurements were handled. The record does not report imputation, sensitivity analyses, reasons for missing data, or the number of participants with unavailable primary endpoint measurements.
The registry lists 137 enrolled participants, while the primary cohort results contain group-specific sample sizes. It does not give a participant flow account explaining exclusions, withdrawals, or deviations from randomized assignment. Nor does it report whether all randomized participants contributed data to every secondary outcome. A named intention-to-treat population does not by itself establish how missing outcomes were managed. Without fuller information on missingness and the analysis plan, the precision and potential bias of estimates cannot be fully assessed.
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 |
|---|---|---|---|
| RO0503821 (1x/ Week) | 15 | 48 | 31.2% |
| RO0503821 (1x/ 3 Weeks) | 13 | 46 | 28.3% |
| RO0503821 (1x/ 4 Weeks) | 20 | 43 | 46.5% |
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 |
|---|---|
| RO0503821 (1x/ Week) | 36 of 48 |
| RO0503821 (1x/ 3 Weeks) | 30 of 46 |
| RO0503821 (1x/ 4 Weeks) | 39 of 43 |
Most frequent serious adverse event terms
| Event term | Organ system | RO0503821 (1x/ Week) | RO0503821 (1x/ 3 Weeks) | RO0503821 (1x/ 4 Weeks) |
|---|---|---|---|---|
| Cardio-Respiratory Arrest | Cardiac disorders | 0 of 48 | 1 of 46 | 3 of 43 |
| Myocardial Infarction | Cardiac disorders | 1 of 48 | 0 of 46 | 2 of 43 |
| Sepsis | Infections and infestations | 1 of 48 | 2 of 46 | 0 of 43 |
| Cerebral haemorrhage | Nervous system disorders | 3 of 48 | 0 of 46 | 0 of 43 |
| Pyrexia | General disorders | 1 of 48 | 0 of 46 | 2 of 43 |
| Fluid overload | Metabolism and nutrition disorders | 1 of 48 | 0 of 46 | 2 of 43 |
| Cardiac failure | Cardiac disorders | 1 of 48 | 1 of 46 | 0 of 43 |
| Bronchopneumonia | Infections and infestations | 0 of 48 | 1 of 46 | 1 of 43 |
| Central line infection | Infections and infestations | 1 of 48 | 1 of 46 | 0 of 43 |
| Pneumonia | Infections and infestations | 1 of 48 | 0 of 46 | 1 of 43 |
Most frequent non-serious adverse event terms
| Event term | Organ system | RO0503821 (1x/ Week) | RO0503821 (1x/ 3 Weeks) | RO0503821 (1x/ 4 Weeks) |
|---|---|---|---|---|
| Muscle spasms | Musculoskeletal and connective tissue disorders | 5 of 48 | 6 of 46 | 11 of 43 |
| Hypotension | Vascular disorders | 8 of 48 | 5 of 46 | 8 of 43 |
| Nasopharyngitis | Infections and infestations | 8 of 48 | 4 of 46 | 8 of 43 |
| Cough | Respiratory, thoracic and mediastinal disorders | 6 of 48 | 6 of 46 | 6 of 43 |
| Hypertension | Vascular disorders | 6 of 48 | 4 of 46 | 6 of 43 |
| Procedural hypotension | Injury, poisoning and procedural complications | 6 of 48 | 0 of 46 | 9 of 43 |
| Headache | Nervous system disorders | 4 of 48 | 6 of 46 | 4 of 43 |
| Vomiting | Gastrointestinal disorders | 5 of 48 | 4 of 46 | 4 of 43 |
| Back pain | Musculoskeletal and connective tissue disorders | 5 of 48 | 2 of 46 | 5 of 43 |
| Pyrexia | General disorders | 3 of 48 | 3 of 46 | 5 of 43 |
Serious events with the highest arm-level rate: cardio-respiratory arrest (7.0%), cerebral haemorrhage (6.2%), myocardial infarction (4.7%), pyrexia (4.7%), fluid overload (4.7%).
The posted safety summaries are organized by administration frequency rather than by the nine dose-and-frequency cohorts. For weekly administration, 15 of 48 participants were reported with a serious adverse event and 36 of 48 with any other adverse event. For administration once every three weeks, the corresponding counts were 13 of 46 and 30 of 46. For administration once every four weeks, they were 20 of 43 and 39 of 43. These are participant counts in the registry’s safety groups; they do not show event timing, duration, severity beyond the serious classification, or whether events were judged related to treatment.
The listed serious-event terms include Cardio-Respiratory Arrest at 7.0%, Cerebral haemorrhage at 6.2%, and Myocardial Infarction, Pyrexia, Fluid overload, and Pulmonary oedema at 4.7% each. The registry record does not provide enough context to interpret those term percentages as comparative risks between regimens. Death counts are not reported in the registry record. These summaries describe reported events and should not be used alone to infer causation or to compare safety across schedules.
13. What the Result Does, and Does Not, Mean
The findings describe hemoglobin changes observed during initial maintenance treatment after switching from subcutaneous epoetin alfa or beta to subcutaneous Mircera, across the conversion-factor and frequency regimens studied. The cohort medians provide a compact account of the center of the observed changes, while the interval bounds and group sizes indicate that the estimates are not equally precise. The sponsor’s grouped conversion-factor estimates provide a separate summary by dosing frequency.
These data do not establish that any regimen is superior, equivalent, or non-inferior to another. The registry record does not report a formal primary comparison among the nine cohort medians, and the sponsor’s conversion-factor analysis method is not reported. The results also do not establish that a particular dose conversion is appropriate for an individual patient or that the observed changes were caused solely by the study treatment. Such conclusions would require evidence and analyses beyond these registry summaries.
The study was open label, and the posted record does not include a placebo or untreated group. The outcomes are therefore interpreted as observations within active-treatment cohorts, not as a comparison against no treatment. The modest cohort sizes and incomplete information about missing data and analysis planning further limit certainty. The secondary pulse-rate recalculations are unadjusted and concern a different outcome; they should not be treated as confirmation of the primary hemoglobin findings.
In general, an estimate is most informative when considered with its uncertainty, the analysis population, the outcome definition, and the planned comparison. Here, the available record supports a descriptive account of the posted results and the sponsor’s conversion-factor estimates. It does not support a definitive ranking of the dosing regimens or a broad conclusion about treatment effects beyond the measured outcomes and timeframes.
14. Why This Trial Matters Statistically
This trial is statistically instructive because it separates dose conversion from administration frequency in a multi-cohort design. The same three conversion factors were paired with weekly, once-every-three-weeks, and once-every-four-weeks schedules. That structure creates several related groups and makes clear why an endpoint summary for each cohort is not automatically the same as an analysis of conversion factors grouped by schedule. Readers need to identify what each estimate represents before comparing values.
The primary outcome also illustrates how endpoint construction affects interpretation. Baseline used screening and run-in measurements; the end-of-treatment measurement could be the last observed value before dose change or transfusion, or the Week 19 or Week 21 value when those events did not occur. The registry also describes multiplying individual slope values by 42 for ease of interpretation. These choices define the quantity being summarized, and a clear account of such rules is essential to understand what a reported median means.
The results show why precision matters alongside the point estimate. Cohort sizes are limited, and interval bounds differ across cohorts. The sponsor’s grouped estimates have explicitly reported 90% confidence intervals, while the interval type for the primary cohort bounds is not stated. That difference in reporting limits direct comparison of apparent precision across the two types of result.
Finally, the record highlights the importance of complete statistical reporting. The analysis method for conversion factors, missing-data procedures, multiplicity strategy, and sample-size justification are not reported. These omissions do not erase the descriptive information, but they constrain confirmatory interpretation. Distinguishing posted estimates from planning illustrations and unadjusted recalculations helps prevent analyses with different purposes from being conflated.
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.
Several limitations are specific to the available record. The trial was open label, with no masking. The registry describes nine active-treatment cohorts but does not identify a placebo or untreated comparison. The primary endpoint is reported as a median for each cohort, without a stated formal test comparing those medians. The primary bounds are not identified as a particular type of interval, and the inter-quartile range is named as the dispersion measure but not reported with the cohort values.
The sponsor’s conversion-factor analyses are reported by administration frequency, but the statistical method, estimand details, and any adjustment for covariates are not reported in the registry record. The relationship between those grouped estimates and the individual cohort endpoint values is not fully described. No p-values are reported for the conversion-factor estimates.
Information about participant flow and missing measurements is limited. The primary population is described as all randomized participants, and the endpoint rule specifies how values before dose change or transfusion were selected. The record does not explain imputation, missing-data sensitivity analyses, withdrawals, or the number of missing primary measurements. It also does not report full results for all listed secondary outcomes.
The safety summaries are grouped by administration frequency, not by all nine regimens, and do not provide treatment attribution or event timing. Death counts are not reported in the registry record. Finally, the sample-size justification and full statistical analysis plan are not reported. These limitations make the posted results most useful as descriptive evidence rather than as a complete basis for formal comparisons or a definitive ranking of regimens.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: A Study of Subcutaneous (sc) Mircera in Dialysis Patients With Chronic Renal Anemia. — NCT00364832.
- Registry record last updated 2016-12-20; status last verified 2016-10.
- Results first posted 2016-12-20.
- 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
NCT00364832 was a completed, randomized, open-label, parallel phase 2 study sponsored by Hoffmann-La Roche. It enrolled 137 dialysis patients with chronic renal anemia who were switching from subcutaneous epoetin alfa or beta to subcutaneous RO0503821 (Mircera). Nine cohorts combined three dose-conversion factors with weekly, once-every-three-weeks, or once-every-four-weeks administration.
The primary outcome was median hemoglobin change from baseline to the end of initial treatment under a constant dosing regimen. The registry also reported estimated conversion factors with 90% confidence intervals grouped by administration frequency. Secondary reporting includes pulse-rate summaries and safety counts, while results for several listed secondary outcomes are not reported in the registry record.
The record does not report the complete statistical plan, conversion-factor analysis method, multiplicity approach, or missing-data procedures. The available estimates therefore support descriptive interpretation, with caution about precision and the absence of a reported formal primary comparison between regimens.