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 Study to Assess the Efficacy, Safety, Tolerability, and Pharmacokinetics of BIIB033 in Participants With Relapsing Forms of Multiple Sclerosis When Used Concurrently With Avonex (NCT01864148), a randomized phase 2 study. The primary outcome is Proportion of Participants Confirmed as Improvement Responders for Primary Multicomponent Endpoint.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01864148 |
| Phase | Phase 2 |
| Condition | Multiple Sclerosis |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 419 |
| Arms in the primary outcome | 5 |
| Lead sponsor | Biogen |
| Results posted | Yes |
2. Clinical Question
Population
Participants with active relapsing multiple sclerosis were studied. The efficacy population included randomized participants who received at least 1 dose of study treatment; 6 participants were excluded from that analysis because of a study-site Good Clinical Practice deviation.
Intervention
BIIB033 was evaluated at 3 mg/kg, 10 mg/kg, 30 mg/kg, or 100 mg/kg. Each dose was given by intravenous infusion once every 4 weeks up to Week 72, concurrently with once-weekly intramuscular Avonex up to Week 84.
Comparator
The comparator was placebo infusion once every 4 weeks up to Week 72, with the same once-weekly Avonex schedule up to Week 84.
Primary question
Over 72 weeks, what proportion of participants in each group were estimated to have a confirmed improvement in at least one component of the multicomponent endpoint? The components assessed disability, hand function, walking, and cognitive performance. The analysis compared each BIIB033 dose group with placebo and also included a trend test across groups.
The trial’s primary objective was to evaluate BIIB033 efficacy when used concurrently with Avonex. Safety, tolerability, and population pharmacokinetics were secondary objectives. The registry record describes the endpoint as the estimated proportion of participants confirmed as improvement responders; it does not establish that any one component alone determined responder status.
3. Trial Design
This was a randomized, double-blind, parallel-group, placebo-controlled phase 2 study sponsored by Biogen. The registry record lists an enrollment of 419 participants and a completed study status. The study began in 2013-08, with primary completion in 2015-12.
Participants were assigned to placebo or one of four BIIB033 dose groups: 3 mg/kg, 10 mg/kg, 30 mg/kg, or 100 mg/kg. The assigned infusion was administered once every 4 weeks up to Week 72. All groups also received once-weekly intramuscular Avonex up to Week 84. The placebo group therefore provides a comparison of BIIB033 added to Avonex against placebo added to Avonex, rather than against no treatment.
Randomization and double masking are features intended to limit systematic differences between groups and influence from knowledge of assignment. The parallel-group design means the groups were followed under their assigned regimens rather than being described as a crossover comparison. The record does not report the randomization procedure, allocation concealment details, or how masking was maintained.
The primary efficacy analysis population was not simply everyone counted in enrollment: it consisted of randomized participants who received at least 1 dose of study treatment and were included in the efficacy analysis, with 6 participants excluded due to a study-site Good Clinical Practice deviation. The record does not report further detail on participant flow or the number contributing data at each assessment. The listed secondary outcomes address confirmed worsening responders, adverse events and discontinuations, and BIIB033 plasma concentrations.
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: A total of 419 participants were randomized; 1 participant was not dosed.
Overall Study
| Stage | Placebo | BIIB033, 3 mg/kg | BIIB033, 10 mg/kg | BIIB033, 30 mg/kg | BIIB033, 100 mg/kg |
|---|---|---|---|---|---|
| STARTED | 93 | 45 | 95 | 94 | 92 |
| Randomized and Dosed | 93 | 45 | 95 | 93 | 92 |
| COMPLETED | 73 | 40 | 84 | 68 | 69 |
| NOT COMPLETED | 20 | 5 | 11 | 26 | 23 |
Reasons for not completing this period:
| Reason | Placebo | BIIB033, 3 mg/kg | BIIB033, 10 mg/kg | BIIB033, 30 mg/kg | BIIB033, 100 mg/kg |
|---|---|---|---|---|---|
| Other | 2 | 1 | 1 | 3 | 1 |
| Death | 0 | 0 | 0 | 1 | 0 |
| Investigator Decision | 5 | 0 | 0 | 4 | 6 |
| Withdrawal by Subject | 9 | 2 | 6 | 8 | 8 |
| Lost to Follow-up | 0 | 0 | 0 | 2 | 1 |
| Adverse Event | 4 | 2 | 4 | 7 | 7 |
| Not Dosed | 0 | 0 | 0 | 1 | 0 |
Analysis population for the primary outcome, as defined in the registry: Intent-to-treat population: all randomized participants who received at least 1 dose of study treatment and included in the efficacy analysis (6 participants were excluded due to study site Good Clinical Practice deviation).
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: Intent-to-treat population: all randomized participants who received at least 1 dose of study treatment.
| Characteristic | Unit | Placebo | BIIB033, 3 mg/kg | BIIB033, 10 mg/kg | BIIB033, 30 mg/kg | BIIB033, 100 mg/kg | Total |
|---|---|---|---|---|---|---|---|
| Participants analysed | 93 | 45 | 95 | 93 | 92 | 418 | |
| Age, Continuous | years | 39.5 (SD 9.29) | 36.5 (SD 9.47) | 40.5 (SD 9.78) | 40.9 (SD 9.70) | 39.8 (SD 9.10) | 39.8 (SD 9.51) |
| Sex: Female, Male: Female | Participants | 67 | 24 | 59 | 61 | 66 | 277 |
| Sex: Female, Male: Male | Participants | 26 | 21 | 36 | 32 | 26 | 141 |
6. Primary Endpoint
The registry lists the primary outcome as Proportion of Participants Confirmed as Improvement Responders for Primary Multicomponent Endpoint, measured over 72 weeks.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | proportion of participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint was the estimated proportion of participants confirmed as improvement responders for a multicomponent measure, assessed over 72 weeks. A participant could qualify through improvement in at least one of four areas: disability measured by the Expanded Disability Status Scale (EDSS), hand function measured by the 9-Hole Peg Test (9HPT), walking measured by the Timed 25-Foot Walk (T25FW), or cognitive performance measured by the 3-Second Paced Auditory Serial Addition Test (PASAT-3).
The registry definition requires improvement to be sustained for at least 3 months. For EDSS, the decrease was assessed from a baseline score of no more than 6.0. For the 9HPT and T25FW, improvement was defined relative to baseline test time; the 9HPT assessment used the average of 2 trials per hand at the same visit, and the T25FW used the average of 2 trials at the same visit. The PASAT-3 criterion was based on improvement from baseline score sustained for 3 months or longer.
A composite responder endpoint can capture change across different domains rather than relying on a single measure. It also combines distinct ways of improving, so the overall responder proportion does not show which component drove the result. Component-specific contributions and their uncertainty are not detailed in the posted primary results. The estimated proportions were based on logistic regression adjusted for multiple sclerosis type, region, and baseline component assessments.
7. Results Posted in the Registry
- BIIB033, 3 mg/kg: 0.0% (0/45) versus 0.0% (0/91) on control; risk difference 0.0 percentage points (95% CI 0.0 to 0.0); unadjusted two-proportion z-test p = n/a.
- BIIB033, 10 mg/kg: 1.1% (1/94) versus 0.0% (0/91) on control; risk difference 1.1 percentage points (95% CI -1.0 to 3.1); unadjusted two-proportion z-test p = 0.324.
- BIIB033, 30 mg/kg: 1.1% (1/91) versus 0.0% (0/91) on control; risk difference 1.1 percentage points (95% CI -1.0 to 3.2); unadjusted two-proportion z-test p = 0.316.
- BIIB033, 100 mg/kg: 0.0% (0/91) versus 0.0% (0/91) on control; risk difference 0.0 percentage points (95% CI 0.0 to 0.0); unadjusted two-proportion z-test p = n/a.
Risk difference: BIIB033, 3 mg/kg
95% CI: 0.0 to 0.0 percentage points
Recalculated from the posted summary statistics
8. Statistical Interpretation
The registry reports estimated improvement-responder proportions of 0.516 for placebo, 0.511 for BIIB033 3 mg/kg, 0.656 for 10 mg/kg, 0.688 for 30 mg/kg, and 0.412 for 100 mg/kg. These are model-based estimates for the composite responder outcome, not raw counts of participants who improved. The estimates do not rise steadily across the dose groups: the 3 mg/kg estimate is close to the placebo estimate, the estimates at 10 mg/kg and 30 mg/kg are higher, and the estimate at 100 mg/kg is lower than the placebo estimate.
The sponsor-reported logistic regression compared each dose group with placebo. The odds ratio was 0.98 for 3 mg/kg, with a 95% confidence interval from 0.46 to 2.07; 1.79 for 10 mg/kg, with a 95% confidence interval from 0.97 to 3.31; 2.06 for 30 mg/kg, with a 95% confidence interval from 1.11 to 3.84; and 0.66 for 100 mg/kg, with a 95% confidence interval from 0.36 to 1.21. These odds ratios describe relative odds of being classified as an improvement responder under the reported model. The intervals for the 3 mg/kg, 10 mg/kg, and 100 mg/kg comparisons include the null odds ratio of 1; the interval for the 30 mg/kg comparison does not.
The registry also reports a trend test across all groups, but the posted record does not establish a clear dose-response pattern from the arm estimates. A single comparison with an interval that excludes the null should not be read in isolation from the other dose comparisons, the trend analysis, or the absence of reported multiplicity procedures. The registry’s primary result is an estimate of a composite response, not a direct statement about each component or about every participant’s course.
Simple unadjusted recalculations from posted summary statistics do not reproduce the sponsor’s reported analysis. In view of this discrepancy, the sponsor’s own analysis takes precedence for interpreting the primary endpoint. The recalculations should not be substituted for the adjusted logistic-regression estimates. The record does not report enough detail to resolve the discrepancy or to establish whether the calculations use identical endpoint definitions and analysis populations.
These data describe the results recorded for this trial and the precision of the reported estimates. They do not, by themselves, establish a broader clinical conclusion. The interpretation is limited by the missing multiplicity and missing-data details, the composite nature of the endpoint, and the differences between the reported analysis and the simple recalculations.
9. Statistical Methods Explained
The primary outcome is binary at the participant level: each participant is classified as an improvement responder or not, according to the multicomponent definition. The reported group result is an estimated proportion. In general, binary outcomes are often analysed with logistic regression because the model relates participant characteristics and group assignment to the probability of an event while keeping predicted probabilities within their possible range.
The registry specifies logistic regression adjusted for multiple sclerosis type, region, and baseline component assessments. Adjustment means that the reported dose-versus-placebo comparisons account for these listed factors in the model. The record does not give the model equation, the coding of these factors, any further covariates, or details of how missing component assessments were handled. The displayed proportions are described as estimated, so they should not automatically be treated as simple fractions calculated directly from the listed group sizes.
What does an odds ratio describe?
An odds ratio compares the odds of the endpoint between two groups. An odds ratio of 1 is the null value: it represents equal odds under the comparison. For an improvement endpoint, an odds ratio above 1 indicates higher estimated odds in the BIIB033 group than in placebo under the fitted model; an odds ratio below 1 indicates lower estimated odds. The odds ratio is not itself a difference in proportions, and it should not be read as a percentage increase or decrease in the chance of response.
The reported comparisons use placebo as the reference group. For 3 mg/kg, the odds ratio is 0.98; for 10 mg/kg, it is 1.79; for 30 mg/kg, it is 2.06; and for 100 mg/kg, it is 0.66. The interpretation of each estimate depends on its confidence interval as well as its point estimate. Estimates can be compatible with a range of effects, particularly when their intervals are wide.
Why use a confidence interval?
A confidence interval describes the uncertainty around an estimate under the statistical model and its assumptions. In general, a 95% confidence interval is generated by a procedure that would cover the target parameter in 95% of repeated samples under those assumptions. It is not a guarantee that the true effect lies within the interval, and it does not measure the probability that the treatment has a particular effect. Here, the 95% intervals for the dose comparisons indicate substantial uncertainty for some estimates.
What does the trend test add?
The registry reports a trend test across placebo and all BIIB033 dose groups. A trend test evaluates a pattern across ordered groups rather than treating every dose-versus-placebo contrast as an isolated question. The test’s result is not a substitute for the individual comparisons, and a trend test does not by itself establish that effects increase consistently with dose. The reported arm estimates are not monotonic across dose levels. The record does not describe the trend test’s exact model, its dose coding, or whether it was prespecified.
How should the recalculations be treated?
The recalculated comparisons are simple, unadjusted calculations based on posted summary statistics. They are not the sponsor’s adjusted logistic-regression model. The two approaches can differ because they use different methods and may target different estimands. Here the record marks a discrepancy, so the sponsor’s own analysis takes precedence. The recalculated results are useful as a check on the posted summaries, not as a replacement for the registry’s reported primary analysis.
The sample-size illustrations use the observed effect as a planning reference, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is expressed as a stated fraction of the observed effect. Such calculations illustrate planning under assumptions; they are not observed power, do not explain the trial’s original enrollment decision, and should not be mistaken for the trial’s own design justification. The registry record does not report the trial’s sample-size justification.
10. Confidence Intervals and Precision
The sponsor-reported 95% confidence intervals show how imprecisely the odds ratios are estimated. For 3 mg/kg versus placebo, the odds ratio is 0.98 with an interval from 0.46 to 2.07. For 10 mg/kg, the estimate is 1.79 with an interval from 0.97 to 3.31. For 30 mg/kg, it is 2.06 with an interval from 1.11 to 3.84. For 100 mg/kg, it is 0.66 with an interval from 0.36 to 1.21.
In general, a wider confidence interval indicates less precision about the size and direction of an effect. These intervals span materially different odds ratios. The interval for 10 mg/kg comes close to the null value of 1, while the intervals for 3 mg/kg and 100 mg/kg include it. The interval for 30 mg/kg lies above 1, but it still covers a range of possible relative odds. A point estimate alone would conceal this uncertainty.
The registry reports no confidence intervals alongside the estimated responder proportions. The recalculated risk-difference comparisons are not aligned with the sponsor’s adjusted logistic analysis, and the record identifies a discrepancy. Consequently, the sponsor-reported odds-ratio intervals are the appropriate intervals to use when describing the primary analysis. The posted group sizes should not be used to infer precision for those adjusted model estimates without additional information about the analysis.
Confidence intervals address sampling uncertainty under a model; they do not account for every possible source of uncertainty. They do not resolve concerns about missing data, endpoint definition, selection into the efficacy population, or multiple comparisons. The registry record does not report enough information to assess those issues fully.
11. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Regression, Logistic | Odds Ratio (OR) 0.98 | 0.46 to 2.07 | 0.958 |
| Regression, Logistic | Odds Ratio (OR) 1.79 | 0.97 to 3.31 | 0.064 |
| Regression, Logistic | Odds Ratio (OR) 2.06 | 1.11 to 3.84 | 0.022 |
| Regression, Logistic | Odds Ratio (OR) 0.66 | 0.36 to 1.21 | 0.177 |
| Trend test | not reported | 0.893 |
The sponsor’s primary analysis is a logistic regression adjusted for multiple sclerosis type, region, and baseline component assessments. It reports odds ratios and 95% confidence intervals for each BIIB033 dose group against placebo, plus a trend test across groups. Those are the registry’s stated primary-analysis results.
The simple recalculations use posted summary statistics and are unadjusted. They do not reproduce the sponsor’s reported comparisons, and the record identifies a discrepancy. Differences between an unadjusted recalculation and an adjusted regression analysis are not necessarily a like-for-like comparison: the model, adjustment, and possibly the analysis inputs can differ. The registry record does not provide enough information to determine the precise reason for the mismatch.
Because the discrepancy is flagged, the sponsor’s own analysis takes precedence. The recalculations should be read as checks on the posted summaries rather than as corrected versions of the sponsor’s results. The distinction matters especially because the recalculations are expressed as risk differences, whereas the sponsor reports odds ratios. These measures have different scales and cannot be substituted for one another without a clearly defined analysis.
12. Multiplicity, Missing Data and Analysis Populations
The registry lists 4 outcome measures, 3 secondary outcomes, and 10 analyses. It reports separate dose-versus-placebo comparisons for the primary endpoint and a trend test. When multiple comparisons are examined, the chance of at least one apparently unusual result can increase. Statistical procedures can account for multiplicity, but the registry record does not report whether any adjustment was made, which comparisons formed a testing family, or which analysis was designated as confirmatory.
This limits interpretation of the individual dose comparisons, particularly when one interval excludes the null while the others include it. The record does not describe interim analyses, stopping rules, or a complete statistical analysis plan. These details are not reported in the registry record and should not be inferred from the posted comparisons.
The efficacy population is described as randomized participants who received at least 1 dose of study treatment and were included in the efficacy analysis; 6 participants were excluded due to a study-site Good Clinical Practice deviation. The record does not provide a fuller account of exclusions, missing endpoint components, or the number assessed at each time point. It also does not report whether missing data were imputed, how withdrawals were handled, or what sensitivity analyses were used. These omissions matter because the endpoint requires confirmation of sustained improvement and combines several component assessments.
The enrollment figure and efficacy-analysis population description are not interchangeable. The registry record does not explain the relationship between the enrollment total and each outcome-specific analysis population. It also does not report a missing-data assumption or a detailed participant-flow account. As a result, readers cannot assess from this record alone how incomplete follow-up may have affected the estimated responder proportions.
13. 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 | 13 | 93 | 14.0% |
| BIIB033 3 mg/kg | 4 | 45 | 8.9% |
| BIIB033 10 mg/kg | 11 | 95 | 11.6% |
| BIIB033 30 mg/kg | 20 | 93 | 21.5% |
| BIIB033 100 mg/kg | 16 | 92 | 17.4% |
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 93 |
| BIIB033 3 mg/kg | 38 of 45 |
| BIIB033 10 mg/kg | 79 of 95 |
| BIIB033 30 mg/kg | 70 of 93 |
| BIIB033 100 mg/kg | 67 of 92 |
Most frequent serious adverse event terms
| Event term | Organ system | Placebo | BIIB033 3 mg/kg | BIIB033 10 mg/kg | BIIB033 30 mg/kg | BIIB033 100 mg/kg |
|---|---|---|---|---|---|---|
| Multiple sclerosis relapse | Nervous system disorders | 7 of 93 | 2 of 45 | 6 of 95 | 10 of 93 | 6 of 92 |
| Hypersensitivity | Immune system disorders | 0 of 93 | 0 of 45 | 0 of 95 | 0 of 93 | 4 of 92 |
| Urinary tract infection | Infections and infestations | 2 of 93 | 0 of 45 | 0 of 95 | 0 of 93 | 1 of 92 |
| Hypochromic anaemia | Blood and lymphatic system disorders | 1 of 93 | 0 of 45 | 0 of 95 | 0 of 93 | 0 of 92 |
| Acute myocardial infarction | Cardiac disorders | 0 of 93 | 0 of 45 | 0 of 95 | 1 of 93 | 0 of 92 |
| Myocardial infarction | Cardiac disorders | 0 of 93 | 0 of 45 | 1 of 95 | 0 of 93 | 0 of 92 |
| Myocardial ischaemia | Cardiac disorders | 0 of 93 | 1 of 45 | 0 of 95 | 0 of 93 | 0 of 92 |
| Ileus | Gastrointestinal disorders | 0 of 93 | 0 of 45 | 1 of 95 | 0 of 93 | 0 of 92 |
| Pancreatitis | Gastrointestinal disorders | 0 of 93 | 0 of 45 | 0 of 95 | 0 of 93 | 1 of 92 |
| Small intestinal obstruction | Gastrointestinal disorders | 1 of 93 | 0 of 45 | 0 of 95 | 0 of 93 | 0 of 92 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Placebo | BIIB033 3 mg/kg | BIIB033 10 mg/kg | BIIB033 30 mg/kg | BIIB033 100 mg/kg |
|---|---|---|---|---|---|---|
| Influenza like illness | General disorders | 37 of 93 | 17 of 45 | 51 of 95 | 34 of 93 | 38 of 92 |
| Multiple sclerosis relapse | Nervous system disorders | 30 of 93 | 17 of 45 | 35 of 95 | 36 of 93 | 28 of 92 |
| Headache | Nervous system disorders | 23 of 93 | 8 of 45 | 19 of 95 | 13 of 93 | 11 of 92 |
| Upper respiratory tract infection | Infections and infestations | 13 of 93 | 4 of 45 | 21 of 95 | 11 of 93 | 9 of 92 |
| Urinary tract infection | Infections and infestations | 12 of 93 | 7 of 45 | 14 of 95 | 9 of 93 | 13 of 92 |
| Nasopharyngitis | Infections and infestations | 16 of 93 | 3 of 45 | 12 of 95 | 8 of 93 | 10 of 92 |
| Pyrexia | General disorders | 7 of 93 | 9 of 45 | 8 of 95 | 12 of 93 | 11 of 92 |
| Fatigue | General disorders | 8 of 93 | 6 of 45 | 5 of 95 | 7 of 93 | 7 of 92 |
| Back pain | Musculoskeletal and connective tissue disorders | 9 of 93 | 3 of 45 | 9 of 95 | 6 of 93 | 6 of 92 |
| Fall | Injury, poisoning and procedural complications | 10 of 93 | 0 of 45 | 4 of 95 | 8 of 93 | 5 of 92 |
Serious events with the highest arm-level rate: multiple sclerosis relapse (10.8%), hypersensitivity (4.3%), myocardial ischaemia (2.2%), fall (2.2%), femur fracture (2.2%).
The registry reports serious adverse events and other adverse events by group as participant counts at risk and counts affected. For serious adverse events, the counts are 13 of 93 for placebo, 4 of 45 for BIIB033 3 mg/kg, 11 of 95 for 10 mg/kg, 20 of 93 for 30 mg/kg, and 16 of 92 for 100 mg/kg. For other adverse events, the corresponding counts are 74 of 93, 38 of 45, 79 of 95, 70 of 93, and 67 of 92.
These are descriptive group counts, not adjusted comparisons. The registry also lists serious adverse-event terms, including multiple sclerosis relapse at 10.8%, hypersensitivity at 4.3%, and myocardial ischaemia, fall, femur fracture, and trigeminal neuralgia at 2.2% each. The record does not specify in this summary whether those term percentages apply across all groups or provide group-specific rates, so they should not be assigned to a particular dose group.
Deaths are not reported in the registry record. The listed counts do not establish comparative safety by themselves; interpretation would require details such as event definitions, exposure, follow-up, and statistical comparisons, which are not fully reported in the registry record.
14. What the Result Does, and Does Not, Mean
The primary endpoint measures whether participants met a sustained improvement criterion in at least one of several functional or cognitive components during the 72-week assessment period. It does not measure improvement in every component, and a composite responder estimate does not reveal which component was most influential. The registry reports adjusted responder proportions and dose-versus-placebo odds ratios; these are statistical summaries of the trial data, not a complete account of an individual participant’s experience.
The arm estimates vary across BIIB033 doses rather than following a steady increase. The sponsor-reported 30 mg/kg comparison has an odds-ratio confidence interval above the null, while the other dose-comparison intervals include the null. That pattern should be interpreted alongside the trend test, the absence of reported multiplicity details, and the discrepancy between sponsor analysis and unadjusted recalculations. No single comparison answers every question about the endpoint or the set of doses.
The record also includes adverse-event counts, but it does not provide enough information to treat those counts as a full comparative safety analysis. Nor does it report the detailed missing-data methods or complete analysis plan needed to assess all sources of bias and uncertainty. The trial’s findings therefore describe the posted results within this study and under the reported analysis, with important limits on precision and interpretation.
These results alone do not establish that BIIB033 does or does not produce a clinically meaningful benefit, and they do not support advice about treatment. They address a particular regimen: BIIB033 or placebo given concurrently with Avonex in participants with relapsing forms of multiple sclerosis. Generalization beyond the studied participants, regimens, endpoint, and follow-up requires evidence not contained in the registry record.
15. Why This Trial Matters Statistically
This trial is statistically informative because it compares several BIIB033 dose levels with a shared placebo-plus-Avonex control and reports both a composite responder endpoint and model-adjusted comparisons. The range of estimates illustrates why dose-ranging results should not be reduced to a single point estimate or read as an automatically increasing response with dose. The 100 mg/kg estimate is lower than the placebo estimate, while the 10 mg/kg and 30 mg/kg estimates are higher; the overall pattern is not monotonic.
The endpoint also demonstrates a common trade-off in composite outcomes. Combining disability, hand function, walking, and cognitive measures can capture improvement across different domains, but the overall responder proportion does not show which domain contributed to the result. Component-level reporting and analysis details would be needed to understand that pattern fully.
The trial is also a useful example of why the statistical estimand and analysis method matter. The sponsor reports adjusted odds ratios from logistic regression, while the simple recalculations use unadjusted summary statistics and risk differences. Those outputs answer related but not identical questions. When the registry identifies a discrepancy, the sponsor analysis takes precedence rather than treating the recalculation as interchangeable.
Finally, confidence intervals, multiplicity, missing data, and population definitions are central to reading efficacy results. The reported intervals vary in relation to the null, but the record does not fully describe multiplicity control or missing-data handling. Those gaps limit how confidently individual comparisons can be isolated from the overall analysis.
16. 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.
- Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.
The registry record does not report the detailed randomization procedure, allocation concealment, or how masking was maintained. It describes a double-blind randomized parallel-group study, but does not include the full protocol or statistical analysis plan. The trial’s sample-size justification is not reported in the registry record, and planning illustrations based on observed effects should not be mistaken for the original design rationale.
The primary outcome is a composite defined by improvement in at least one component. The posted primary results do not give component-specific responder estimates or explain how each component contributed to the overall result. The record also does not provide confidence intervals for the estimated responder proportions.
The efficacy population excluded 6 participants because of a study-site Good Clinical Practice deviation, but the registry does not describe the deviation, its timing, or its potential effect on the analysis. Further participant flow, missing-data handling, withdrawals, and sensitivity analyses are not reported in the registry record. These omissions limit assessment of whether incomplete observations could have affected the endpoint estimates.
The record reports multiple dose comparisons and a trend test but does not state whether multiplicity was controlled or identify a complete testing hierarchy. It also flags a discrepancy between the sponsor’s adjusted results and simple unadjusted recalculations; the record does not explain the source. For safety, group counts are available, but the summary does not provide complete exposure-adjusted comparisons, and deaths are not reported in the registry record. These trial-specific reporting limits constrain interpretation of both efficacy and safety.
17. Related Tutorials
18. Related Statistical Calculators
19. Sources
- ClinicalTrials.gov: Study to Assess the Efficacy, Safety, Tolerability, and Pharmacokinetics of BIIB033 in Participants With Relapsing Forms of Multiple Sclerosis When Used Concurrently With Avonex — NCT01864148.
- Registry record last updated 2017-05-03; status last verified 2017-03.
- Results first posted 2017-05-03.
- 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.
20. Record Summary
NCT01864148 is a completed, randomized, double-blind, parallel-group, placebo-controlled phase 2 study sponsored by Biogen. It enrolled 419 participants and began in 2013-08, with primary completion in 2015-12. Participants received placebo or BIIB033 at 3 mg/kg, 10 mg/kg, 30 mg/kg, or 100 mg/kg, concurrently with Avonex.
The primary outcome was the estimated proportion confirmed as improvement responders for a multicomponent endpoint over 72 weeks. The sponsor’s reported analysis used adjusted logistic regression and included dose-versus-placebo odds ratios with 95% confidence intervals, as well as a trend test. The posted results also include adverse-event counts and secondary outcome listings.
The record identifies a discrepancy between the sponsor’s analysis and simple unadjusted recalculations, so the sponsor’s analysis takes precedence. Multiplicity procedures, detailed missing-data methods, and several other analysis details are not reported in the registry record.