A Study of the Impact of Methotrexate (MTX) Discontinuation on the Efficacy of Subcutaneous (SC) Tocilizumab (TCZ) With MTX: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of the Impact of Methotrexate (MTX) Discontinuation on the Efficacy of Subcutaneous (SC) Tocilizumab (TCZ) With MTX (NCT01855789), a randomized phase 3 study. The primary outcome is Change From Week 24 in Disease Activity Score Based on 28 Joints (DAS28) Score at Week 40.

2. Clinical Question

The key design feature is that the randomized question applies to a selected group: participants who completed the initial 24-week treatment with open-label tocilizumab and methotrexate and achieved a DAS28 score at or below 3.2. It is not a randomized comparison of starting tocilizumab with versus without methotrexate. Nor does the randomized comparison cover participants who did not meet the Week 24 threshold; they continued open-label treatment without random assignment.

The registry describes a double-blind, parallel-group randomized trial with 718 participants enrolled overall. That enrollment figure is not the number randomized at Week 24. The number randomized, the allocation ratio, and the group sizes for the primary analysis are not reported in the registry record. Thus, the clinical question is specific both to the treatment decision after Week 24 and to the participants eligible for randomization at that point.

3. Trial Design

This was a Phase 3, randomized, double-blind, parallel-group study in participants with rheumatoid arthritis. The registry lists Hoffmann-La Roche as the sponsor and an overall enrollment of 718. The study started on 2013-11-07, reached primary completion on 2016-10-14, and is recorded as completed.

The study had an initial open-label treatment period. Participants received subcutaneous tocilizumab weekly or every 2 weeks together with oral methotrexate at a stable dose for 24 weeks. At Week 24, disease activity was assessed using DAS28. Participants whose score was at or below 3.2 were randomized either to continue methotrexate with tocilizumab or to switch from methotrexate to matched placebo while continuing tocilizumab. The blinded randomized treatment continued up to Week 52.

Participants whose Week 24 DAS28 score was above the eligibility threshold did not enter the randomized comparison. They continued tocilizumab with methotrexate in an open-label, non-randomized manner up to Week 52. This group therefore followed a different treatment and analysis pathway from the randomized participants.

The primary endpoint was change in DAS28 from Week 24 to Week 40 among the randomized group. The registry describes the analysis population as participants in the ITT population who were randomized and received blinded treatment at Week 24. It also states that missing Week 40 values were imputed using the last available assessment obtained after Week 24. The number analyzed is described as the number of participants.

Specific details of the randomization procedure, allocation ratio, and sample-size justification are not reported in the registry record. The record also does not describe interim analyses or give detailed masking procedures beyond identifying the study as double-blind.

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: Randomized participants constituted the primary analysis population for the efficacy assessments, while all other analyses including safety assessments were performed on overall study population. An MRI sub-study was planned to evaluate joint inflammation in conjunction with other clinical signs and symptoms.

1) Baseline up to Week 24

Reasons for not completing this period:

2) Week 24 up to End of Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Analysis was performed on randomized group which included participants in ITT population who were randomized and received blinded treatment at Week 24. Missing Week 40 values were imputed using last available assessment obtained after Week 24. Here, 'Number Analyzed' signifies number of participants with assessment at specified time point.

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 (ITT) population included all participants who received at least 1 dose of study medication and had at least 1 post-baseline efficacy assessment.

6. Primary Endpoint

The registry lists the primary outcome as Change From Week 24 in Disease Activity Score Based on 28 Joints (DAS28) Score at Week 40, measured over Week 24, Week 40.

The primary endpoint was mean change in DAS28 score from Week 24 to Week 40, expressed in units on a scale. DAS28 combines measures of inflammation and joint involvement: erythrocyte sedimentation rate, tender joint count across 28 joints, swollen joint count across 28 joints, and the participant’s global assessment of disease activity. The registry describes a score ranging from 0 to 10, with higher values representing higher disease activity.

For this endpoint, negative change from Week 24 indicates improvement in disease activity, while a positive change indicates an increase in the score. Comparing mean changes between the randomized groups is intended to characterize the average difference in disease-activity trajectory after one group discontinued methotrexate and the other continued it, with both groups receiving tocilizumab.

A continuous endpoint retains information about the amount of change in the score rather than reducing each participant to a yes-or-no category. The mean summarizes the group’s average change, while the standard deviation describes variation in individual changes around that mean. General statistical interpretation: the mean alone does not show how consistently individual participants changed, and the standard deviation is not a measure of uncertainty in the estimated group mean.

The registry record does not report the endpoint’s group means, standard deviations, treatment contrast, confidence interval, or p-value. Therefore, the direction and precision of the posted primary comparison cannot be described from the registry information available.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The methods described for the primary endpoint center on a continuous change score, a randomized analysis population, and an imputation rule for missing Week 40 assessments. These elements define what the reported estimate would represent and what assumptions affect its interpretation.

What does a mean change summarize?

The endpoint is the average change in DAS28 from Week 24 to Week 40. Each participant’s change is measured relative to their Week 24 score; the group mean then summarizes the average change among participants included in the analysis. The registry states that negative changes indicate improvement in disease activity. A mean is useful for summarizing a continuous measure, but it can conceal variation: participants may have changes quite different from the average.

The registry identifies standard deviation as the measure of dispersion. General statistical explanation: standard deviation describes how spread out individual observations are around a group mean. It is not itself the standard error or a confidence interval, and it does not directly quantify the precision of the difference between randomized groups. Precision also depends on the amount of information in the analysis, including the number of analyzed participants and variability. The registry record does not report the primary endpoint’s group standard deviations or analyzed group sizes.

What does randomization contribute?

Randomization assigns eligible participants to continue methotrexate or switch to matched placebo, with tocilizumab continuing in both groups. In general, random assignment helps make treatment groups comparable at the point of assignment, on average, and supports causal interpretation of a between-group contrast for the randomized population. Here, that population is restricted to participants who completed the initial treatment and had a Week 24 DAS28 score at or below 3.2. The randomized contrast therefore does not directly represent all enrolled participants.

The study is described as double-blind. Blinding can reduce the risk that knowledge of assigned treatment influences participant reports, clinical assessments, or other study conduct. The registry record does not provide further details about who was blinded or how the blinding was maintained.

How should change-score results be read?

A group mean change describes the average movement in the score from the Week 24 reference. A treatment contrast would compare the mean change in the continued-methotrexate group with the mean change in the placebo-for-methotrexate group. Its sign depends on the subtraction order used, which is not specified in the registry description. Without the estimates, the contrast, and its uncertainty, no direction or magnitude of the comparative result can be stated.

General statistical explanation: a confidence interval expresses uncertainty around an estimate under the analysis method. A narrow interval indicates greater precision than a wide interval, all else equal. A p-value addresses compatibility between the observed data and a specified null hypothesis under a statistical model; it does not measure the size or clinical importance of an effect. The registry record does not report a confidence interval or p-value for the primary comparison.

What does the missing-data rule do?

For missing Week 40 values, the primary analysis used the last available assessment obtained after Week 24. This is a last-observation-carried-forward approach: an available post-Week 24 measurement is used in place of the missing Week 40 assessment. It retains participants with a missing endpoint under the stated procedure, but it treats the carried-forward observation as a stand-in for the later value.

General statistical explanation: this approach implicitly relies on assumptions about the unobserved trajectory after the last available assessment. If disease activity would have changed after that assessment, carrying the earlier value forward may not represent the missing Week 40 value. The direction and size of any resulting effect cannot be determined without information on missingness and observed trajectories. The registry record does not report missing-data counts, reasons for missing assessments, or sensitivity analyses comparing alternative assumptions.

How do planning illustrations differ from trial results?

Any sample-size illustrations associated with the posted results are planning calculations, not observed power. The stated assumptions for those illustrations are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect defined as a stated fraction of the observed effect. Such calculations illustrate participant numbers under assumptions; they do not establish the trial’s original design rationale or demonstrate the probability that the completed trial detected an effect. The sample-size justification and the original protocol assumptions are not reported in the registry record.

General statistical explanation: over-dispersion means that observed variation is greater than a model assumes. If a model understates variability, its uncertainty estimates may be too small and its inferences too optimistic. The registry record does not identify an over-dispersion adjustment for this continuous primary endpoint, so none should be inferred.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary endpoint and multiple secondary outcomes. The secondary outcomes include ACR20, ACR50, and ACR70 response; worsening in DAS28 from Week 24; DAS28 remission and low disease activity categories; change in bone erosion score in an MRI substudy; and anti-therapeutic antibodies to tocilizumab. It also records 11 outcome measures and 17 analyses. The multiplicity strategy, including whether or how significance levels were adjusted across outcomes or analyses, is not reported in the registry record.

Multiplicity matters because assessing several outcomes or analyses can increase the chance of at least one apparently notable result arising by chance. General statistical explanation: adjustment methods can change how evidence is evaluated across a family of tests, while prespecified hierarchies can define which results are confirmatory. Without the analysis plan, it is not possible to identify the intended testing family or determine whether any reported secondary finding should be treated as confirmatory.

For the primary endpoint, the registry states that Week 40 values missing from the randomized analysis were imputed using the last available assessment obtained after Week 24. It does not report how many values were missing, why they were missing, or whether sensitivity analyses examined alternative missing-data assumptions. The handling of missing data for the secondary outcomes is not reported in the registry record.

The analysis population is described as randomized participants in the ITT population who received blinded treatment at Week 24. The record says that the number analyzed means the number of participants, but does not give primary analysis counts by group or explain how participants with no usable post-Week 24 assessment were handled. Further details of the analysis populations and any protocol deviations are not reported in the registry record.

Interim analyses and any associated stopping rules are not reported in the registry record. Without these details, the multiplicity and missing-data features of the analysis cannot be fully evaluated.

11. Safety Results

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

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: pneumonia (1.4%), myocardial infarction (0.7%), chest pain (0.7%), acute respiratory failure (0.6%), cellulitis (0.4%).

The registry reports a safety population of 713 participants at risk for both serious adverse events and other adverse events. Serious adverse events were reported for 72 participants, and other adverse events for 348 participants. These are counts in the overall study population, not treatment-group comparisons. The registry record does not provide the corresponding randomized-group breakdown in the information.

The listed serious adverse event terms and percentages are pneumonia, 1.4%; myocardial infarction, 0.7%; chest pain, 0.7%; acute respiratory failure, 0.6%; cellulitis, 0.4%; and arthritis infective, 0.3%. These entries describe the registry’s reported terms and percentages; they do not, on their own, show whether an event was caused by a treatment or whether rates differed between randomized groups.

General statistical explanation: adverse-event counts are descriptive safety summaries. Comparisons require attention to the group denominators, exposure time, event definitions, and whether participants can contribute more than one event. The listed information does not provide enough detail to make such comparisons. Deaths are not reported in the registry record.

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

The study’s randomized design addresses a focused question: among participants who received open-label tocilizumab with methotrexate and reached a DAS28 score at or below 3.2 at Week 24, how did mean DAS28 change through Week 40 compare between continuing methotrexate and switching to matched placebo? Both randomized groups continued tocilizumab. This is a question about methotrexate continuation within a selected group, not a test of tocilizumab versus no tocilizumab.

The registry record does not report the primary group estimates or their comparison. It therefore does not establish whether the mean change differed, which group had the more favorable observed change, or how precisely any difference was estimated. No conclusion about statistical significance or clinical importance can be drawn from the endpoint description alone.

The enrollment of 718 includes participants who followed both randomized and non-randomized pathways. The randomized findings, if reported, would apply most directly to the eligible participants who entered that comparison. They would not automatically generalize to participants who did not meet the Week 24 criterion, to people who did not complete the initial treatment period, or to other treatment settings.

General statistical explanation: a statistically detectable difference is not necessarily large enough to be important to patients, and a result that is not statistically detectable does not demonstrate that two treatments are equivalent. Assessing those questions requires the estimated treatment contrast, its uncertainty, an appropriate clinical context, and the study’s analysis plan.

The record’s safety counts are overall descriptive summaries. Without randomized-group safety results and relevant exposure information, they cannot establish comparative safety. The listed serious adverse event terms should likewise not be interpreted as evidence of causation. The registry information supports understanding the trial’s design and intended comparisons, but the missing comparative results prevent conclusions about the primary outcome.

13. Why This Trial Matters Statistically

Statistically, this trial is notable for separating an initial treatment period from a later randomized comparison. All participants began open-label tocilizumab with methotrexate; randomization occurred only after the Week 24 disease-activity assessment. This design makes the randomized question specific to participants who reached the stated threshold and allows the comparison to focus on whether methotrexate was continued alongside tocilizumab after that point.

That distinction is important when defining the estimand, the quantity a study aims to estimate. The primary endpoint is not the overall change from study entry, nor a comparison among every enrolled participant. It is mean DAS28 change from Week 24 to Week 40 in the randomized population. The non-randomized participants who did not meet the Week 24 criterion followed a different treatment pathway and should not be combined conceptually with the randomized comparison.

The endpoint also demonstrates why analysis details matter. A continuous change score, its standard deviation, and the handling of missing Week 40 measurements all shape how an estimate is understood. The stated last-available-assessment imputation rule is relevant to interpretation, while the absence of missing-data counts and sensitivity analyses limits evaluation of its impact.

The registry record does not report the primary results, so the statistical importance lies in clarifying the question and the boundaries of any interpretation rather than claiming a treatment effect. The record also lists multiple secondary outcomes and analyses without describing a multiplicity strategy. This makes it especially important to distinguish the designated primary endpoint from secondary findings in any complete results report.

Finally, the overall enrollment should not be mistaken for the randomized analysis size. Recognizing the two-stage structure prevents an overbroad interpretation of either the target population or the evidence the randomized comparison could provide.

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 principal limitation for interpreting the primary endpoint is that its numerical results are not reported in the registry record. Group means, standard deviations, analyzed sample sizes, a treatment contrast, confidence interval, and p-value are absent. Consequently, the size, direction, and precision of the primary comparison cannot be assessed.

The trial’s randomized population was selected after 24 weeks of open-label treatment. Only participants with a Week 24 DAS28 score at or below 3.2 were randomized; those above the threshold continued open-label treatment without randomization. This design limits the direct scope of the randomized comparison to the participants who met the criterion and completed the initial treatment pathway. The number randomized is not reported in the registry record.

The primary analysis used last available assessment after Week 24 to impute missing Week 40 values. The registry record does not report the number of missing assessments, reasons for missingness, or sensitivity analyses. The influence of this imputation on the estimated mean change therefore cannot be evaluated from the record.

Details of the randomization procedure, allocation ratio, sample-size justification, and full statistical analysis plan are not reported in the registry record. Neither interim analyses nor a multiplicity strategy is described. These omissions limit assessment of the original design rationale and the interpretation of multiple outcome analyses.

Safety information is reported for the overall study population, without the randomized-group breakdown or exposure details needed for comparative interpretation. The record lists serious adverse event terms but does not report deaths. These limitations mean the available safety summary cannot establish treatment-specific event rates or causation.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study of the Impact of Methotrexate (MTX) Discontinuation on the Efficacy of Subcutaneous (SC) Tocilizumab (TCZ) With MTX — NCT01855789.
• Registry record last updated 2017-12-26; status last verified 2017-11.
• Results first posted 2017-12-26.
• 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

NCT01855789 is a completed Phase 3 randomized, double-blind, parallel-group study sponsored by Hoffmann-La Roche. It enrolled 718 participants with rheumatoid arthritis and began on 2013-11-07; primary completion was 2016-10-14.

Participants initially received open-label subcutaneous tocilizumab weekly or every 2 weeks with stable-dose oral methotrexate for 24 weeks. Participants with a Week 24 DAS28 score at or below 3.2 were randomized to continue methotrexate or switch to matched placebo while continuing tocilizumab. Those above the threshold continued open-label tocilizumab with methotrexate without randomization.

The primary outcome was mean change in DAS28 from Week 24 to Week 40 in the randomized population. Negative changes indicated improvement. Missing Week 40 values were imputed using the last available assessment after Week 24. Primary outcome estimates and the treatment comparison are not reported in the registry record.

The overall safety population was 713 participants. Serious adverse events were reported for 72 participants and other adverse events for 348. The registry lists 11 outcome measures and 17 analyses.