A Study of RoActemra/Actemra and, if Initially Inadequately Responded to RoActemra/Actemra, Followed by MabThera/Rituxan in Patients With Rheumatoid Arthritis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of RoActemra/Actemra and, if Initially Inadequately Responded to RoActemra/Actemra, Followed by MabThera/Rituxan in Patients With Rheumatoid Arthritis (NCT01332994), a non-randomized phase 3 study. The primary outcome is Percentage of Participants Achieving Remission at Week 16 According to DAS28.

2. Clinical Question

The study concerns a sequential treatment strategy rather than a randomized comparison of tocilizumab against rituximab. Participants initially received tocilizumab; the planned subsequent treatment depended on their clinical response after the initial treatment period. Consequently, the primary question is about the remission percentage at Week 16 within the study population, not about whether one treatment is superior to another.

The registry lists rheumatoid arthritis as the condition and describes the participants as DMARD-IR. The meaning of the abbreviation and more detailed eligibility requirements are not reported in the registry record. Nor does the record specify how the response categories were determined beyond describing an inadequate clinical response after the first 12 weeks.

The endpoint is a proportion, so it describes the share of participants meeting the specified DAS28 threshold at the assessment. It is not a measure of the size of an average change in disease activity, nor does it by itself quantify a causal treatment effect. Without a reported numerical estimate, the primary remission percentage cannot be assessed from the registry record.

3. Trial Design

The registry identifies this as a Phase 3, non-randomized, open-label, parallel study of a sequential treatment approach in participants with rheumatoid arthritis. It is also described in the study summary as multi-center and uncontrolled. The listed enrollment is 519. The study status is completed; its recorded start is 2011-03 and its primary completion is 2014-02.

Participants were to start tocilizumab at 8 mg/kg intravenously every 4 weeks for 12 weeks. Those with an adequate response could receive tocilizumab for a further 12 weeks. Those with an inadequate clinical response after the initial 12 weeks were to receive rituximab at 1 g intravenously at Week 16 and Week 18. The anticipated time of study treatment is 32 weeks. These descriptions define the intended treatment sequence, but the registry record does not report how many participants followed each pathway or how treatment exposure was handled in the outcome analyses.

Non-randomized allocation means that treatment pathways were not assigned by random allocation as described in the registry. The response-dependent sequence also means that participants receiving rituximab were selected on the basis of their initial response to tocilizumab. Outcomes in that subgroup therefore cannot be interpreted as a direct comparison with outcomes among participants who continued tocilizumab. Differences between the groups could reflect the response-based selection as well as subsequent treatment.

The registry labels the model parallel but also describes the study as uncontrolled and non-randomized. The arm descriptions are blank. The record therefore does not clarify the operational definition of each arm, how participants were assigned to them, or whether the analyses consistently followed the initially planned pathway. No randomization method, masking procedure, or control-treatment regimen is reported; masking is listed as none.

The primary outcome is remission at Week 16 according to DAS28, among the Main ITT Population. The registry also lists multiple secondary outcomes, including DAS28 remission and low disease activity at several assessment times and in treatment-defined subgroups. Outcome-specific numerical results are not included in the registry record for this explanation, so no response rate or between-group estimate can be described.

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: After a screening period of up to 4 weeks, eligible participants were treated according to a predefined treatment algorithm based on disease response. In certain cases a re-screening was allowed. Participants were enrolled into the study with the administration of their first dose.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Main ITT Population

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: Main Intent-to-Treat (ITT) Population: All participants who received at least one dose of TCZ.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants Achieving Remission at Week 16 According to DAS28, measured over Week 16.

The primary endpoint is the percentage of participants achieving remission at Week 16 according to DAS28, in the Main ITT Population. DAS28 combines information about swollen and tender joints, erythrocyte sedimentation rate (ESR), and the participant’s global assessment of disease activity. The registry describes the score as ranging from 0 to 10, with higher values indicating greater disease activity. Remission is defined as a score <2.6 at the assessment visit.

A threshold-based endpoint converts a continuous score into a classification: each participant either meets the remission definition at the assessment or does not. The percentage expresses how common that classification was in the analysis population. Such endpoints can be clinically interpretable because they describe the proportion reaching a defined state, although they do not show how far participants were from the threshold or how their scores changed over time.

DAS28 incorporates several aspects of disease activity rather than relying on a single joint count or laboratory measure. In general, combining measures can give a broader picture, while also making interpretation dependent on the component measurements and the stated scoring rule. The registry gives the DAS28 formula and identifies the assessment as occurring at Week 16.

The registry lists a 95% confidence interval as the endpoint’s dispersion measure, but its numerical bounds and the remission percentage are not reported in the registry record. It is therefore possible to describe what the endpoint was intended to measure, but not the observed remission frequency or the precision of its estimate.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a percentage endpoint and lists a 95% confidence interval as its dispersion measure. The endpoint is remission at Week 16 according to the stated DAS28 threshold. The percentage is a proportion: it represents the participants classified as having remission relative to the analysis population used for that outcome. The numerical proportion and interval are not reported in the registry record.

What does a confidence interval convey?

In general, a confidence interval communicates the statistical uncertainty around an estimate under the assumptions of the analysis. For a percentage, a narrower interval indicates greater precision in the estimated proportion, while a wider interval indicates less precision. The interval does not describe the range of outcomes for individual participants, and it does not account automatically for bias arising from study design, measurement, or missing observations. The registry record gives the intended confidence level but not the interval bounds, so precision cannot be evaluated for this endpoint.

Confidence intervals should be read alongside the design and the analysis population. They quantify sampling uncertainty under a statistical model; they do not turn a non-randomized, uncontrolled study into a randomized comparison. Even a precise estimate of the Week 16 remission percentage would remain a description of the study population unless supported by an appropriate comparative design and analysis.

Why does the treatment sequence matter?

In general, when later treatment depends on earlier response, participants who receive different later treatments are not exchangeable simply because they appear in separate groups. The reason for receiving a treatment is linked to an earlier outcome. This can create selection differences between groups and complicate comparisons of later outcomes. In this study, rituximab was planned for participants with an inadequate clinical response after the first 12 weeks of tocilizumab, while adequate responders could continue tocilizumab. The registry record does not describe a method that would remove the resulting selection problem from any subgroup comparison.

The listed parallel model does not, by itself, establish that the treatment groups were comparable. The allocation is explicitly listed as non-randomized, and the study is described as uncontrolled. Random allocation, when used, can help balance measured and unmeasured characteristics on average; without it, differences in patient characteristics, response history, or clinical decisions may influence observed outcomes. The registry does not report the allocation procedure or adjusted comparative estimates.

What can the endpoint definition tell the reader?

DAS28 combines joint counts, ESR, and a patient global assessment on a visual analog scale. The endpoint then classifies participants by whether the score was below the remission threshold at the Week 16 assessment. In general, converting a continuous score into a threshold-based outcome can make a target state easier to interpret, but it discards information about the amount of change and the distance from the cutoff. A participant just on either side of a threshold may have similar scores but be assigned to different categories.

The registry says that the endpoint is a percentage and identifies the Main ITT Population. It does not provide a detailed definition of that analysis population, the denominator used for the Week 16 percentage, or the handling of participants without an assessment. Those details matter because percentages depend on how missing observations and analysis eligibility are treated. The record also does not specify whether the analysis used any adjustment or which interval procedure was used.

How should planning illustrations be interpreted?

In general, sample-size calculations are planning tools: they describe the amount of information a hypothetical design might need under stated assumptions. The planning illustrations associated with this analysis use 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. These are illustrations based on the observed effect, not the trial’s own design. They are not observed power and do not show that the study had a particular probability of detecting an effect.

The registry record does not report the design assumptions used to set enrollment, a sample-size justification, or a statistical analysis plan. The planning assumptions for an illustration should therefore not be treated as evidence of the original design rationale. Nor can any planning estimate be interpreted as an outcome: it does not add information about the observed remission percentage or its uncertainty.

What statistical results are available?

The record identifies the primary endpoint, its assessment time, population label, and intended confidence-interval level. It also lists a substantial number of outcome measures and analyses. The numerical primary estimate, confidence-interval bounds, p-value, and analysis details are not reported in the registry record. Without those results, the endpoint’s observed frequency and inferential precision cannot be assessed. General statistical principles can clarify how such results would be read, but they cannot substitute for the missing numerical findings.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("TCZ/TCZ or TCZ/RTX")
n      <- c(519)
events <- c(222)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  TCZ/TCZ or TCZ/RTX  519  222  42.8  38.5  47.2  38.6  47.1  38.6  47.1

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 53 outcome measures and 69 analyses, including secondary outcomes at different assessment times and among treatment-defined subgroups. When many outcomes or analyses are examined, the chance of obtaining an apparently notable result by chance can increase. In general, a statistical plan may address this through a prespecified hierarchy, adjustment, or a clearly defined distinction between confirmatory and exploratory analyses. The registry record does not report whether any such approach was used, nor does it describe how the primary outcome was prioritized relative to the other analyses.

The primary endpoint is identified for the Main ITT Population, but the registry record does not define that population in detail. It does not report the denominator used for the Week 16 percentage, the handling of missing DAS28 assessments, or whether any imputation or sensitivity analyses were performed. These omissions matter for a percentage endpoint because decisions about participants without observed assessments can affect the estimated proportion and its uncertainty.

The record also does not report the statistical analysis plan, the method used to calculate the 95% confidence interval, or whether analyses were adjusted for baseline characteristics. It does not describe interim analyses or stopping rules; these details are not reported in the registry record. No inference should be made about whether such methods were or were not used beyond what the entry states.

The presence of a Main ITT Population label does not resolve these questions. The record identifies the label but does not specify how participants were included in that population or how deviations from the planned treatment sequence were handled. Without these analysis details and the numerical endpoint results, the effect of missingness and multiplicity on interpretation cannot be assessed.

12. 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: bronchitis (0.6%), bursitis (0.6%), abdominal pain (0.4%), hypersensitivity (0.4%), diverticulitis (0.4%).

The registry lists a group labelled TCZ/TCZ or TCZ/RTX, with 519 at risk for the serious adverse-event measure and 54 affected. For the other adverse-event measure, it lists 519 at risk and 104 affected. These are registry counts for the measures as recorded; they do not provide a randomized comparison between treatment pathways.

The registry also lists serious-event terms with percentages: bronchitis and bursitis at 0.6% each; abdominal pain, hypersensitivity, diverticulitis, and pneumonia at 0.4% each. These entries indicate the listed terms and their recorded percentages. The record does not provide further detail about event definitions, severity grading, duration, recurrence, or exposure-adjusted rates. The terms should not be interpreted as a complete account of all safety outcomes.

The group label combines the tocilizumab-to-tocilizumab and tocilizumab-to-rituximab pathways. Since the study was non-randomized and the later treatment depended on initial response, these safety figures cannot establish a comparative safety effect of one pathway versus another. The registry record does not report a death count. The counts also should not be treated as counts of distinct participants unless the record specifies that interpretation.

Safety counts describe what was recorded in the study; they do not by themselves establish that an event was caused by treatment. The registry does not report the details needed to assess causal attribution or compare event rates under a common exposure period.

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

The record establishes that the study examined a sequential approach beginning with tocilizumab and planned rituximab for participants with an inadequate initial response. It identifies remission at Week 16 according to DAS28 as the primary endpoint and defines remission as a score <2.6. It also describes the intended treatment schedule and lists selected safety information.

Those details do not reveal the observed primary outcome. The remission percentage and its 95% confidence interval are not reported in the registry record. It is therefore not possible to say what proportion achieved remission, how precise that estimate was, or how it compared with another treatment. The absence of a numerical result in this record should not be mistaken for a result showing either a high or a low remission rate.

Even if a remission percentage were available, the study design would shape what it could mean. Non-randomized allocation and the lack of a control group limit causal inference. Participants receiving rituximab were selected following an inadequate response to tocilizumab, so outcomes in the later treatment pathway would not be directly comparable with outcomes among participants who continued tocilizumab. The registry record does not describe a method that would establish a causal contrast between these paths.

The endpoint is a classification at a specified assessment, not a complete description of disease activity over time. It does not by itself show the extent of score changes, how durable remission was, or whether participants who did not meet the threshold experienced meaningful improvement. The secondary outcomes listed in the record include other DAS28 assessments, but their numerical results are not reported in the registry record for this explanation.

The safety counts likewise describe recorded events rather than a comparative causal effect. Together, the registry information supports a careful account of the question and design, but not a conclusion that either treatment works or does not work. The numerical efficacy results and important analysis details are not reported in the registry record.

14. Why This Trial Matters Statistically

This study is statistically notable because its treatment pathway is response-dependent. Participants did not simply receive alternative treatments under randomized allocation: initial response to tocilizumab determined whether the planned subsequent path involved continued tocilizumab or rituximab. That structure makes the treatment groups clinically and statistically different by design. It highlights why subgroup outcomes should not automatically be interpreted as treatment comparisons.

The study also illustrates the distinction between a single-arm or uncontrolled outcome estimate and a causal treatment effect. A remission percentage at Week 16 would describe the observed population under the study’s care pathway. Without an appropriate comparator, it cannot isolate the effect of the treatment from other influences. The non-randomized allocation and open-label design reinforce the need to describe results in relation to their design rather than as if they came from a blinded randomized comparison.

Its primary outcome is a threshold-based remission definition using a composite disease-activity score. This makes the outcome understandable as the share reaching a stated condition at an assessment, while also emphasizing the information lost when a continuous score is reduced to a yes-or-no classification. The definition, assessment time, and analysis population all matter for interpreting a percentage.

The registry lists many outcomes and analyses, making transparent multiplicity procedures and clear prioritization important for interpretation. Yet the statistical plan and numerical primary result are not reported in the registry record. This gap shows why design descriptions and endpoint definitions, while useful, cannot replace reported estimates and their uncertainty. The record offers a basis for understanding what was studied, but leaves the main numerical efficacy question unanswered.

15. Limitations

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

The most important limitation for interpreting the posted registry information is that the numerical primary result is not reported in the registry record. The primary remission percentage and its 95% confidence interval are absent, so the observed outcome and its precision cannot be evaluated. The record also does not give numerical results for the secondary outcomes listed.

The study is non-randomized, open-label, and uncontrolled. There is no randomized comparison to support a like-for-like estimate of relative efficacy. Because the treatment pathway after the initial tocilizumab period depended on response, later-treatment groups are selected according to an earlier outcome. Any simple contrast between those groups would be vulnerable to selection differences.

The arm descriptions are blank, and the registry does not explain the allocation process or provide enough detail to reconstruct how participants were assigned to the stated parallel arms. It does not define the Main ITT Population beyond naming it for the primary endpoint. Details about the Week 16 denominator, missing outcome data, protocol deviations, and treatment adherence are not reported in the registry record.

The statistical analysis plan is also not reported. The record does not specify the method used for the confidence interval, any adjustment for baseline characteristics, the handling of multiple outcomes, or sensitivity analyses for missing data. It does not provide a sample-size justification or the original design assumptions. These omissions limit assessment of how analyses were planned and whether the reported endpoint, if available, would be robust to alternative assumptions.

Safety information is presented as counts and selected percentages for a combined treatment-pathway group. The record does not provide enough detail to assess exposure-adjusted incidence, attribution, or comparative risk. These limitations do not establish that a particular result is biased; they describe what cannot be determined from the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study of RoActemra/Actemra and, if Initially Inadequately Responded to RoActemra/Actemra, Followed by MabThera/Rituxan in Patients With Rheumatoid Arthritis — NCT01332994.
• Registry record last updated 2015-09-07; status last verified 2015-08.
• Results first posted 2015-09-07.
• 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.

19. Record Summary

The registry record describes NCT01332994, a completed Phase 3, open-label, non-randomized, uncontrolled study of a sequential treatment approach in rheumatoid arthritis. Listed enrollment is 519. Participants began tocilizumab, with continued treatment planned for adequate responders and rituximab planned for participants with an inadequate clinical response after the first 12 weeks.

The primary endpoint was the percentage achieving DAS28 remission at Week 16 in the Main ITT Population, with remission defined as a DAS28 score <2.6. A 95% confidence interval is identified as the dispersion measure, but the numerical primary result and interval are not reported in the registry record.

The entry lists 53 outcome measures and 69 analyses, along with safety counts and selected serious-event terms. The statistical analysis plan, missing-data methods, multiplicity approach, and detailed definition of the analysis population are not reported in the registry record. The non-randomized, response-dependent design does not provide a like-for-like treatment comparison.