A Phase 2 Study of Cladribine Add-on to Interferon-beta (IFN-beta) Therapy in Multiple Sclerosis (MS) Subjects With Active Disease (ONWARD): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase 2 Study of Cladribine Add-on to Interferon-beta (IFN-beta) Therapy in Multiple Sclerosis (MS) Subjects With Active Disease (ONWARD) (NCT00436826), a randomized phase 2 study. The primary outcome is Double Blind Period: Percentage of Participants With Grade 3 or 4 (Common Terminology Criteria for Adverse Events [CTCAE] v 4.0) Hematological or Liver Toxicity.

2. Clinical Question

This question is primarily about safety and tolerability during add-on treatment, rather than a direct measure of relapse control or disability. The registry lists several secondary outcomes related to MRI lesions, brain volume, and qualifying relapse rate, but their results are not included in the record. Consequently, the available account establishes what the trial set out to assess, not whether either treatment group had better outcomes. The registry lists enrollment as 172; the brief summary describes the study as randomizing around 200 participants. These are differently worded enrollment statements, and the registry record does not explain the difference.

3. Trial Design

ONWARD is described as a Phase 2, multicenter, randomized, double-blind, placebo-controlled, parallel-group study. The registry also labels masking as triple. The study compared oral cladribine or matching placebo, each added to interferon-beta therapy. The brief summary states that randomization was in a 2:1 fashion. The registry lists enrollment as 172 and identifies the study as completed.

Eligibility focused on active multiple sclerosis while a participant was taking interferon-beta: the brief summary specifies at least one relapse within 48 weeks before screening. It also describes inclusion of participants with secondary progressive multiple sclerosis who continued to experience relapses, and those with prior disease-modifying drug use other than interferon-beta who were currently receiving interferon-beta and had recent active symptoms. The registry record does not give a more detailed account of eligibility criteria in the information available for this analysis.

In the double-blind period, the cladribine group received a cumulative dose of 3.5 mg/kg in cycles described at Week 1, 5, 48, and 52. The placebo group received matching tablets alongside interferon-beta. Interferon-beta could be one of the listed regimens: Rebif, Avonex, or Betaseron. The registry record does not report how many participants used each interferon-beta regimen.

Participants who completed the double-blind period were invited to an open-label extension. Eligible participants could receive open-label cladribine with interferon-beta; participants who did not meet extension eligibility criteria entered safety follow-up with interferon-beta only. These later groups reflect study phases after the randomized period and should not be treated as new randomized comparisons. The registry record describes approximately 50 sites worldwide in the brief summary, but does not report site-level results or the allocation process in greater detail.

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: Initially 42 participants randomized under original protocol but enrollment terminated because of hematological toxicities. Protocol amendment 1 and 2 were implemented and enrolled 172 participants. Protocol amendment 5 was generated, resulting in discontinuation of Cladribine and reduction of Extension period and Safety Follow up to 48 weeks.

Double Blind Period (DBP)

Reasons for not completing this period:

Ext. Period (With Cladribine Treatment)

Reasons for not completing this period:

Ext. Period (No Cladribine Treatment)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety population: all randomized participants who received at least 1 dose of study drug in DBP and had follow-up safety data.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Double Blind Period: Percentage of Participants With Grade 3 or 4 (Common Terminology Criteria for Adverse Events [CTCAE] v 4.0) Hematological or Liver Toxicity, measured over Baseline up to Week 96.

The primary endpoint was the percentage of participants with Grade 3 or 4 toxicity in hematological or liver-function parameters, measured from baseline up to Week 96. The listed parameters were lymphocytes, CD4 cells, neutrophils, white blood cells, hemoglobin, ALT, AST, platelets, and bilirubin. The registry defines Grade 3 as severe and Grade 4 as life-threatening or disabling; it also describes Grade 1 as mild, Grade 2 as moderate, and Grade 5 as death.

This type of endpoint is used to describe the frequency of substantial laboratory toxicity during a treatment period. It brings several safety measures into the assessment, because abnormalities may occur in different blood-cell or liver-function tests. However, the endpoint title and description do not establish that all parameters were combined into a single composite result. The registry’s skip note indicates that the primary outcome has several categories or time points and was not reduced to one comparison.

The stated safety population included randomized participants who received at least one dose of study drug in the double-blind period and had follow-up safety data. The registry record does not report the primary endpoint percentages, a treatment contrast, or uncertainty intervals for the endpoint. It therefore supports description of the endpoint’s purpose and definition, but not a numerical comparison of the randomized groups on this measure.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies a binary percentage endpoint: whether a participant had Grade 3 or 4 toxicity on the listed hematological or liver-function measures. In general, a binary endpoint classifies participants into event and no-event categories over a defined period. A percentage describes the share in the event category, but its interpretation depends on a clear event definition, the population included, and adequate handling of participants without complete follow-up. The registry defines the event grades and names the safety population, but does not report the observed primary-endpoint percentages or the analysis details needed to reconstruct them.

What does the endpoint represent?

The endpoint title refers to participants with severe or life-threatening or disabling toxicity. The description lists separate laboratory parameters, which matter because toxicity can be identified in more than one kind of measurement. A reader needs to know whether a participant is counted once if any parameter meets the threshold, whether results are reported separately by parameter, and how repeated measurements are handled. Those details are not reported in the registry record. The record’s note that the outcome has several categories or time points and is not reduced to one comparison reinforces that a single summary cannot be assumed.

How should group percentages be compared?

In general, a treatment comparison for a binary endpoint may report an absolute difference in percentages, a relative measure, or both. Each answers a different question: an absolute difference describes the separation between group event frequencies, while a relative measure compares their scale. These comparisons require clearly specified denominators and an analysis population. No primary-endpoint group percentages or contrast are reported, so none of these measures can be interpreted for this endpoint. No new comparison should be calculated from the adverse-event counts, because they describe different categories and study periods and are not a substitute for the primary analysis.

A confidence interval, in general, communicates the range of effect estimates compatible with the analysis and its assumptions. A narrow interval indicates greater precision than a wide interval, while an interval spanning values that would matter in different directions can leave the interpretation uncertain. The registry record does not report a confidence interval for the primary endpoint. It also does not report a p-value or the statistical test used for the primary endpoint. Without those results, the strength and precision of a treatment contrast cannot be assessed.

Why does the analysis population matter?

The stated safety population consists of randomized participants who received at least one dose of study drug in the double-blind period and had follow-up safety data. This definition focuses the endpoint on participants with exposure and available safety follow-up. In general, analysis populations can differ in whether they preserve randomized assignment, require treatment exposure, or require outcome data. Such differences can affect how a result is interpreted. The registry record does not report a separate intention-to-treat analysis population, the number excluded from the stated safety population, or reasons for missing safety follow-up.

For a safety endpoint, exposure and observation time are relevant because participants may have different opportunities for an event to be recorded. The primary timeframe is baseline up to Week 96, but the registry record does not provide participant-level follow-up duration or methods for accounting for incomplete observation. It is therefore not possible to tell from the available description how missing or partial measurements entered the primary analysis. In general, missing outcomes can bias a percentage if the probability of being observed is related to toxicity or treatment; the direction and extent depend on the missingness pattern and analysis assumptions.

What does over-dispersion mean in general?

Over-dispersion is a general concern for count outcomes when observed variation exceeds what a model assumes. It can make standard errors too small if the extra variability is ignored. This concept may be relevant to some count-based outcomes, such as lesion counts, but the primary endpoint is described as a percentage of participants with toxicity. The registry record does not identify an over-dispersion adjustment or a count model for the primary endpoint, so no such method should be attributed to this trial.

The record lists 39 outcome measures and 1 analysis. A count of outcomes does not, by itself, describe the statistical model, testing hierarchy, or control of false-positive findings across multiple outcomes. The registry record does not report the test, model, covariate adjustment, confidence-interval method, or multiplicity procedure for the primary endpoint. A detailed statistical analysis plan is not reported in the registry record.

How should planning illustrations be read?

In general, sample-size planning calculations describe the number of participants that might be needed under stated assumptions to detect a chosen effect with a specified error rate and power. They are planning illustrations, not observed power, and they do not show what the trial actually found. The planning assumptions stated for the illustration are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. The record does not give the numerical sample-size outputs or the observed primary-endpoint effect in the details. These assumptions should not be mistaken for the trial’s own design justification; that justification is not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 39 outcome measures and 1 analysis, including a primary safety outcome and secondary MRI, brain-volume, and relapse outcomes. Testing many outcomes can increase the chance of at least one apparently unusual result when each is assessed separately. In general, a multiplicity plan may specify a hierarchy, adjust significance thresholds, or distinguish confirmatory from exploratory outcomes. The registry record does not report such a plan, nor does it state whether or how the listed outcomes were prioritized for statistical testing. The number of outcomes alone cannot establish that multiplicity was or was not controlled.

The primary endpoint includes multiple laboratory parameters and the record notes multiple categories or time points. The registry does not report whether these were analyzed as separate endpoints, combined under a prespecified rule, or handled through another approach. It also does not report the statistical model, testing procedure, or adjustment for repeated measurements. The primary outcome is therefore not reducible to one reported comparison from the information available.

The stated safety population includes randomized participants who received at least one dose during the double-blind period and had follow-up safety data. The registry record does not report how many participants lacked follow-up safety data, how missing laboratory measurements were handled, or whether sensitivity analyses were conducted. It also does not describe imputation or other missing-data methods. These omissions limit assessment of whether the reported safety population and its results could be affected by incomplete observation.

In general, missingness can be consequential when the reason a measurement is unavailable is related to the outcome or treatment. An analysis that excludes participants with missing measurements may differ from one that uses a prespecified model or imputation approach. No particular method should be assumed for this trial: missing-data handling and sensitivity analyses are not reported in the registry record.

11. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: traumatic haematoma (3.6%), ulna fracture (3.6%), arthralgia (3.6%), hepatic cyst (2.1%), non-cardiac chest pain (2.1%).

The registry separates serious adverse events from a second category of other adverse events, and reports the number at risk and the number affected for each study group. During the double-blind period, serious adverse events affected 12 participants among 124 at risk in the cladribine group and 5 among 48 at risk in the placebo group. In the separately listed other-adverse-event category, the reported affected counts were 111 among 124 and 33 among 48, respectively. The registry reports 0 deaths in both double-blind groups.

Later study phases are listed separately. In the open-label extension, serious adverse events affected 0 among 47 in the group that initially received cladribine and 1 among 28 in the group that initially received placebo. The corresponding other-adverse-event counts were 24 among 47 and 12 among 28. In safety follow-up, serious adverse events affected 1 among 52 in the initial cladribine group and 0 among 7 in the initial placebo group; the listed other-adverse-event counts were 0 in both groups. The registry reports 0 deaths in each of these groups.

These counts describe participants affected, not event rates adjusted for time at risk. The extension and safety-follow-up groups are not the original randomized comparison: entry and treatment depended on completion and eligibility. The record does not provide uncertainty intervals or comparative tests for these figures. Listed top serious-event terms include traumatic haematoma, ulna fracture, and arthralgia, each at 3.6%, as well as hepatic cyst, non-cardiac chest pain, and abortion spontaneous, each at 2.1%. Their denominators are not reported in the registry record, so the percentages should not be used to infer group-specific risks.

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

The registry record establishes that ONWARD was designed to assess cladribine added to interferon-beta in participants with active multiple sclerosis, with a primary safety endpoint covering Grade 3 or 4 hematological or liver toxicity through Week 96. It also records serious and other adverse-event counts across the double-blind period, open-label extension, and safety follow-up. Those descriptions are informative about the trial’s questions, treatment periods, and reported safety categories.

The record does not provide a single primary-endpoint comparison, a confidence interval, or a p-value. It also does not report the results of the listed secondary outcomes in the information available for this analysis. Thus, the record does not support a conclusion about the size, precision, or statistical significance of a treatment difference in the primary endpoint or about comparative efficacy. Nor can adverse-event counts alone establish that one treatment caused more or fewer events; the counts are descriptive and do not include an adjusted analysis.

In general, a trial result applies most directly to the population, interventions, outcome definitions, and follow-up that were studied. ONWARD’s described population was selected for active disease while receiving interferon-beta, and the brief summary specifies a recent relapse as part of that description. The treatment comparison was add-on cladribine versus matching placebo, with interferon-beta continued. Extension participants entered later study phases under different eligibility and treatment conditions. Results from those phases should not be read as if they preserved the original randomized comparison.

The registry record also does not report how missing safety measurements were handled, how the multiple outcome measures were prioritized, or the statistical methods used for the primary endpoint. These gaps limit reproducibility and make it impossible to judge whether the observed data, if any, were analyzed with suitable uncertainty and multiplicity procedures. The appropriate reading is therefore narrow: the study design and certain adverse-event counts are reported, but the central comparative primary result and its precision are not reported in the registry record.

13. Why This Trial Matters Statistically

This trial is statistically notable because its primary question is framed as a safety percentage across several laboratory measures, while the registry does not present one reduced primary comparison. That distinction matters: knowing the endpoint definition is not the same as knowing the frequency of the endpoint in each group or the uncertainty around a treatment contrast. For researchers, it illustrates why an outcome’s operational definition and its reported analysis both need to be available before a comparative interpretation is possible.

The design also includes a randomized double-blind period followed by an open-label extension and safety follow-up. These phases answer different descriptive questions. The original randomized groups offer the basis for the intended blinded comparison, whereas extension groups are shaped by completion and eligibility and involve later treatment histories. Treating their adverse-event counts as a continuous randomized comparison would ignore those design differences.

The registry lists many outcomes spanning safety, MRI lesion activity, brain volume, and relapse. This breadth makes the statistical plan important: readers need to know which analyses were primary, how secondary outcomes were interpreted, and whether multiplicity was addressed. The record does not report those methods. This is a useful example of the limits of outcome inventories without analysis details or estimates.

Finally, the reported adverse-event counts are interpretable as descriptive counts, but they do not replace a treatment-effect estimate with uncertainty. A careful statistical reading separates what the record documents—the study design, outcome definitions, and counts—from what it does not document: a single primary-endpoint result, its precision, and the methods behind that comparison. That separation helps prevent a protocol description or an unadjusted count from being mistaken for a complete statistical conclusion.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The most consequential limitation of the registry record is that it does not report a single primary-endpoint comparison. The endpoint covers Grade 3 or 4 toxicity across several hematological and liver-function parameters, but group-specific primary percentages and uncertainty estimates are not reported in the record. The registry note that the outcome has several categories or time points and is not reduced to one comparison means that the central safety question cannot be summarized as one numerical contrast from this information.

The registry lists 39 outcome measures but does not report the statistical analysis plan, model, test, confidence-interval method, or multiplicity procedure. Missing-data handling and sensitivity analyses are also not reported. These omissions prevent assessment of how the primary endpoint was analyzed and how robust any result would be to incomplete follow-up or multiple testing.

The adverse-event data are presented across distinct study periods. The extension and safety-follow-up groups were formed after the double-blind period, with open-label treatment eligibility affecting subsequent exposure. Their counts are therefore not interchangeable with randomized double-blind comparisons. The registry also does not give uncertainty estimates or tests for the listed adverse-event counts, and denominators for the listed top serious-event percentages are not reported in the registry record.

The brief summary describes randomization of around 200 participants, while the registry enrollment field lists 172. The record does not explain that difference. It also does not report the number using each interferon-beta regimen, group-specific follow-up duration, or detailed site-level information. These limits should be kept in view when interpreting the scope and precision of the recorded safety information.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Phase 2 Study of Cladribine Add-on to Interferon-beta (IFN-beta) Therapy in Multiple Sclerosis (MS) Subjects With Active Disease (ONWARD) — NCT00436826.
• Registry record last updated 2020-10-12; status last verified 2020-10.
• Results first posted 2013-07-01.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

ONWARD was a completed Phase 2, randomized, double-blind, placebo-controlled, parallel-group study of oral cladribine added to interferon-beta for participants with active multiple sclerosis. The registry lists enrollment as 172 and describes randomization in a 2:1 fashion. The primary endpoint was the percentage of participants with Grade 3 or 4 hematological or liver toxicity from baseline up to Week 96.

The registry defines a safety population and reports adverse-event counts for the double-blind period, open-label extension, and safety follow-up. It lists 39 outcome measures and 1 analysis. A single primary-endpoint comparison, its uncertainty, and the results of the secondary outcomes are not reported in the registry record. Statistical methods for multiplicity and missing data are also not reported. The adverse-event counts should be read descriptively, with the later study phases kept separate from the randomized comparison.