A Study of RoActemra/Actemra (Tocilizumab) Versus Adalimumab in Combination With Methotrexate (MTX) in Patients With Moderate to Severe Active Rheumatoid Arthritis And an Inadequate Response to Treatment With Only One Tumor Necrosis Factor (TNF)-Inhibitor: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study of RoActemra/Actemra (Tocilizumab) Versus Adalimumab in Combination With Methotrexate (MTX) in Patients With Moderate to Severe Active Rheumatoid Arthritis And an Inadequate Response to Treatment With Only One Tumor Necrosis Factor (TNF)-Inhibitor (NCT01283971), a randomized phase 4 study. The primary outcome is Percentage of Participants With Disease Activity Score 28 Joints (DAS28) Remission at Week 24.

2. Clinical Question

3. Trial Design

This was a randomized, open-label, parallel-group phase 4 study of tocilizumab versus adalimumab, with methotrexate in both groups. The registry record gives enrollment as 96 and lists the study status as terminated. The start and primary completion are recorded as 2011-05 and 2012-08. The reason for termination and its timing relative to recruitment or follow-up are not reported in the registry record.

Participants were assigned to tocilizumab plus methotrexate or adalimumab plus methotrexate. The intervention descriptions also list matching placebo tocilizumab or placebo to adalimumab, consistent with the stated dosing schedules, although masking is recorded as none. Tocilizumab was given intravenously at 8 mg/kg every 4 weeks; adalimumab was given subcutaneously at 40 mg every 2 weeks. Both groups received methotrexate 10-25 mg weekly and folate at least 5 mg weekly. The anticipated time on study treatment was 24 weeks.

The primary outcome was the percentage of participants with DAS28 remission at Week 24. The reported outcome denominators were 33 for adalimumab plus methotrexate and 36 for tocilizumab plus methotrexate. These are smaller than the registry enrollment figure, but the reason for the difference is not reported in the registry record. The primary endpoint population description requires receipt of study drug, at least one post-baseline efficacy assessment, and data available at baseline and Week 24.

Random allocation can reduce systematic differences between groups, while an open-label design means treatment assignment was not masked. The randomization method, allocation concealment, sites, eligibility criteria in full, and original sample-size justification are not reported in the registry record. The posted result therefore needs to be read as a comparison in the described study population, rather than as a complete account of design or conduct.

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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants from the Intent-to-treat population, all randomized participants who received study drug and had at least 1 post-baseline efficacy assessment, with data available at Baseline and Week 24. Last observation carried forward was used to impute missing tender and swollen joint counts. All other ACR components were used as observed.

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 Percentage of Participants With Disease Activity Score 28 Joints (DAS28) Remission at Week 24, measured over Week 24.

The primary endpoint was the percentage of participants with DAS28 remission at Week 24. DAS28 combines tender joint count and swollen joint count across 28 joints, the participant’s global assessment of disease activity on a visual analog scale, and the erythrocyte sedimentation rate. The score can range from 2 to 10. The registry defines remission as a score <2.6.

A percentage endpoint answers a direct question about how many participants in each group met a stated threshold at the specified assessment. It differs from a mean change score: it classifies each participant according to whether the remission criterion was met, then reports the group percentage. The result depends on the threshold and on the availability and handling of the measurements used to construct DAS28.

DAS28 is used as a composite measure because it brings several aspects of disease activity into one score. A threshold makes it possible to report a remission category, but it also means participants with scores on opposite sides of the cutoff are classified differently even when their scores are close. The endpoint description says last observation carried forward was used to impute missing tender and swollen joint counts. Other details about missing component values and the handling of the remaining score components are not reported in the registry record.

7. Results Posted in the Registry

• Tocilizumab + Methotrexate: 36.1% (13/36) versus 30.3% (10/33) on control; risk difference 5.8 percentage points (95% CI -16.4 to 28.0), risk ratio 1.19 (0.61–2.34); unadjusted two-proportion z-test p = 0.609.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Percentage of Participants With American College of Rheumatology (ACR20) Response at Week 24

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Tocilizumab + Methotrexate: 62.9% (22/35) versus 66.7% (22/33) on control; risk difference -3.8 percentage points (95% CI -26.5 to 18.9), risk ratio 0.94 (0.66–1.34); unadjusted two-proportion z-test p = 0.743.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Tocilizumab + Methotrexate and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants With ACR50 Response at Week 24

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Tocilizumab + Methotrexate: 42.9% (15/35) versus 24.2% (8/33) on control; risk difference 18.6 percentage points (95% CI -3.4 to 40.6), risk ratio 1.77 (0.87–3.61); unadjusted two-proportion z-test p = 0.105.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Tocilizumab + Methotrexate and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants With ACR70 Response at Week 24

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Tocilizumab + Methotrexate: 22.9% (8/35) versus 18.2% (6/33) on control; risk difference 4.7 percentage points (95% CI -14.5 to 23.8), risk ratio 1.26 (0.49–3.24); unadjusted two-proportion z-test p = 0.634.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Tocilizumab + Methotrexate and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists multiple secondary measures, including ACR20, ACR50, ACR70, EULAR DAS28 response, low disease activity, and changes in DAS28, swollen joint count, and tender joint count. Among the posted secondary results described in detail, the ACR response percentages do not form a uniform pattern across response thresholds: the group contrast differs in direction and size across the thresholds. That variation makes it difficult to treat any one secondary result as a consistent summary of the overall treatment comparison.

These outcomes are related to the primary remission result because they also describe clinical response or disease activity, but they are not interchangeable. ACR response thresholds use a different response definition from DAS28 remission, while joint counts and score changes measure outcomes on their own scales. Agreement across such measures can add context; disagreement can reflect different definitions, thresholds, or measurement features. The secondary outcomes do not replace the primary endpoint, which remains the designated Week 24 DAS28 remission measure.

The registry record reports 19 secondary outcomes, while only some have detailed results in the information summarized for interpretation. Numerical results for the other listed secondary measures are not reported in the registry record. The secondary outcome pattern should therefore not be generalized beyond the posted results that are available.

Multiple outcomes increase the chance that at least one result will appear notable by chance. In general, multiplicity adjustments or a prespecified hierarchy can control how results are interpreted across a family of tests. The registry record does not report whether such a strategy was used. The posted secondary comparisons are unadjusted, so their p-values, where reported, carry less confirmatory weight than a result supported by a specified multiplicity plan. They are best regarded as supportive context alongside the imprecise primary comparison, not as stand-alone proof of a treatment difference.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Percentage of Participants With American College of Rheumatology (ACR20) Response at Week 24
• Percentage of Participants With ACR50 Response at Week 24
• Percentage of Participants With ACR70 Response at Week 24
• Percentage of Participants With Good or Moderate European League Against Rheumatism (EULAR) DAS28 Responses at Week 24
• Percentage of Participants With DAS28 Low Disease Activity (LDAS) at Week 24
• Change From Baseline in DAS28 Score at Week 24
• Change From Baseline in Swollen Joint Count (SJC) at Week 24
• Change From Baseline in Tender Joint Count (TJC) at Week 24

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary at the participant level: each participant either met the DAS28 remission definition at Week 24 or did not. The group percentages describe the proportion classified as in remission. For a binary outcome, a risk difference expresses the difference between the two group proportions on the probability scale. A positive value in the posted comparison indicates a higher observed proportion in the tocilizumab group, while a negative value would indicate a higher proportion in the adalimumab group. The reported risk difference is 0.058.

What does the confidence interval represent?

In general, a confidence interval describes the range of values reasonably compatible with the observed data under a particular statistical method and its assumptions. It communicates precision as well as direction: a narrow interval indicates a more precise estimate, while a wide interval leaves more uncertainty about the effect’s size. For this comparison, the interval runs from -0.164 to 0.280 and crosses zero. That means the data do not pin down a clear direction of the difference. The interval method is not reported in the registry record, so the precise assumptions used to construct it are not available.

How should the p-value be read?

The unadjusted p-value is 0.609. In general, a p-value assesses the compatibility of the data with a null hypothesis, given the statistical model used. A larger p-value is not evidence that the null is true, and it is not the probability that the observed effect arose by chance. Nor does it measure clinical importance. The test used to obtain this p-value is not reported in the registry record. Because it is unadjusted, it also does not account for the possibility of chance findings across a family of multiple outcomes or comparisons.

What assumptions matter for a percentage comparison?

Methods for comparing proportions rely on the outcome classification and the participants included in each analysis. In general, uncertainty estimates also depend on assumptions about independent observations and the sampling distribution used by the method. The registry record does not identify the specific primary comparison procedure, its model, or any covariate adjustment. Accordingly, the posted unadjusted risk difference and p-value should be treated as the reported comparison rather than as evidence of a particular model specification.

The randomized parallel-group structure is relevant because random assignment is intended to support comparable groups at the start of a trial. The record identifies randomization but does not describe the sequence-generation process or allocation concealment. No masking was used. Awareness of treatment can potentially affect participant reporting, clinical assessments, or decisions about care; whether and how such effects were addressed is not reported in the registry record.

The primary endpoint population is described as participants from the Intent-to-treat population who were randomized, received study drug, had at least one post-baseline efficacy assessment, and had data at baseline and Week 24. This differs from an analysis that includes all randomized participants regardless of treatment receipt or available endpoint data. The record also says last observation carried forward was used for missing tender and swollen joint counts. In general, carrying a prior value forward fills a missing later measurement with an earlier one; it can make analysis possible but relies on an assumption about the unobserved later value. The number and pattern of missing measurements, and the handling of missing values for other DAS28 components, are not reported in the registry record.

The record gives an enrollment of 96, while the primary endpoint results are reported for 33 participants in the adalimumab group and 36 in the tocilizumab group. It does not explain how those endpoint denominators arose from enrollment. The primary endpoint percentages should therefore be interpreted using the reported analysis population rather than assumed to represent every enrolled participant.

The study includes ACR20, ACR50, ACR70, and other disease-activity outcomes as secondary measures. Each additional outcome creates another opportunity for a chance finding when many results are considered. The registry record does not report a multiplicity strategy, hierarchy among outcomes, or an overall statistical analysis plan. The secondary comparisons are therefore described as unadjusted where that is how they are posted, and they should be treated as supportive and exploratory rather than as independent confirmatory evidence.

Sample-size calculations associated with the posted primary result are planning illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. These assumptions describe the illustrations, not the trial’s original design. The original design’s sample-size justification and statistical plan are not reported in the registry record. A planning calculation does not reveal the probability that this completed study detected a true effect; it describes participant counts under specified hypothetical assumptions.

11. Confidence Intervals and Precision

The observed remission percentages are 30.3% for adalimumab plus methotrexate and 36.1% for tocilizumab plus methotrexate. Their unadjusted risk-difference confidence interval extends from -0.164 to 0.280. The interval crosses zero, meaning that no difference remains compatible with the interval’s range. Its negative and positive values also allow for uncertainty in either direction. The estimate is therefore not precise enough to distinguish a clear advantage for either group.

A confidence interval is not a range that contains a fixed proportion of individual participant outcomes, and it is not a prediction interval for a future trial. In general, it reflects uncertainty in an estimated group contrast under a chosen statistical method. The method used to calculate this interval is not reported in the registry record. The interval should be interpreted alongside the group percentages, analysis sizes, outcome definition, and stated population, rather than in isolation.

The unadjusted p-value of 0.609 is consistent with the interval’s inclusion of zero, but the p-value and confidence interval answer different questions. The interval communicates a range of effect estimates compatible with the data under its method; the p-value summarizes compatibility with a null comparison under a test. Neither establishes equivalence. An equivalence conclusion would require a prespecified margin and an analysis designed to evaluate that margin, neither of which is reported in the registry record.

The analysis sizes of 33 and 36 are modest in relation to the enrollment figure of 96, and the record does not explain the difference. Small analysis groups generally yield less precise estimates for a binary endpoint than larger groups, although the actual precision also depends on the observed proportions and analysis method. Here, the interval is broad enough that the observed difference should not be read as a stable estimate of the true difference in a wider population.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size figures associated with the primary endpoint are planning illustrations, not the trial’s observed sample sizes. They are stated as participants per arm, rather than total events. Under the illustration for 100% of the observed effect, the calculation gives 1,031 participants per arm for 80% power and 1,380 participants per arm for 90% power. For 75% of the observed effect, the corresponding figures are 1,814 and 2,427 participants per arm. For 50% of the observed effect, they are 4,033 and 5,399 participants per arm.

In general, power is the probability that a planned study will detect an effect of a specified size under the assumptions used for the calculation. The 80% power column represents a planning target with an 80% probability of detecting the stated effect if that effect and the other assumptions hold; the 90% column uses a higher target. These are not guarantees about a future study’s result.

The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The assumed effect is the stated fraction of the observed effect. The registry record does not report the original sample-size justification or design assumptions, so these calculations should not be treated as the basis on which this trial was planned. This is not observed power.

The illustrations show the general relationship between effect size and required sample size: smaller effects need more participants to detect reliably under otherwise unchanged assumptions. Because the observed primary comparison is imprecise, planning based on a fraction of its observed effect should be viewed as a scenario, not as a definitive recruitment target. The original design’s intended power and target sample size are not reported in the registry record.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.303; treatment proportion p₁ = 0.361; average p̄ = 0.332.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6660 = 1.3054.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6648 = 0.5595.
4. n = (1.3054 + 0.5595)² / (0.0581)² = 1,030.9, rounded up to 1,031 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.303030303030303                          # control proportion
p1 <- 0.361111111111111                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(1283971)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 1,031 (80% power), 1,380 (90% power)
 75% of observed effect: n per arm = 1,814 (80% power), 2,427 (90% power)
 50% of observed effect: n per arm = 4,033 (80% power), 5,399 (90% power)
power.prop.test: 1030.9 
Simulated power at n = 1,031 per arm: 0.796 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

The study names a primary endpoint and several secondary outcomes, including three ACR response thresholds, EULAR DAS28 response, DAS28 low disease activity, and changes in DAS28, swollen joint count, and tender joint count. The registry record lists 19 secondary outcomes. Evaluating many outcomes can increase the chance of finding an apparently positive result even when no true effect exists. In general, a multiplicity plan may define a hierarchy, adjust significance thresholds, or otherwise specify how a family of results will be interpreted. Such a plan is not reported in the registry record.

The posted secondary comparisons are unadjusted. This does not make them uninformative, but it means their nominal statistical evidence should not be read as if only one hypothesis had been tested. The analysis plan, the number of planned comparisons, and any prespecified outcome hierarchy are not reported in the registry record. Detailed numerical results are not reported in the registry record for all listed secondary measures.

The primary endpoint population is described as randomized participants who received study drug and had at least one post-baseline efficacy assessment, with data available at baseline and Week 24. The description also says last observation carried forward was used to impute missing tender and swollen joint counts. The endpoint population wording is truncated in the registry record, and further details are not reported. The number of participants with missing data, reasons for missingness, and sensitivity analyses for missing outcomes are not reported in the registry record.

Last observation carried forward uses a participant’s earlier observed joint count in place of a missing later count. In general, this approach assumes that the carried-forward value is a reasonable stand-in for the missing measurement. If participants’ disease activity changes after their last observation, that assumption may not hold. The registry record does not give enough information to assess the extent or likely direction of any bias. It also does not specify how missing values in the other components of DAS28 were handled.

The reported analysis sizes differ by outcome: the primary result includes 33 and 36 participants, while the ACR20, ACR50, and ACR70 results include 33 and 35. Safety groups report 48 participants at risk in each group. The reasons for these denominator differences, and the exact relationship between the enrollment figure and each analysis set, are not reported in the registry record. This limits comparisons across endpoints and makes it important to read each result with its stated analysis size.

15. 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 (2.1%), appendicitis (2.1%), bursitis infective (2.1%), urinary tract infection (2.1%), hip fracture (2.1%).

The safety record reports serious adverse events for 2 participants in the adalimumab plus methotrexate group and 8 in the tocilizumab plus methotrexate group; each group lists 48 participants at risk. Other adverse events are reported for 21 and 11 participants, respectively, with 48 at risk in each group. These are counts, not percentages, and the record does not explain why the safety group sizes differ from the efficacy analysis sizes.

The serious-event terms listed include pneumonia, appendicitis, infective bursitis, urinary tract infection, hip fracture, and tibia fracture, each reported as 2.1%. The registry record does not provide enough detail to establish whether these terms were considered related to study treatment. A serious adverse event is a safety classification, not necessarily an event caused by a drug; the event definition and attribution methods are not reported in the registry record.

Deaths are not reported in the registry record. The counts should be read as descriptive safety information, not as proof that one regimen is safer. With no information on event timing, duration of exposure, severity beyond the serious classification, or treatment attribution, the figures cannot support a causal comparison by themselves.

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

The result describes the proportion of participants in each analysis group who met the registry’s DAS28 remission definition at Week 24. The observed percentages were 30.3% with adalimumab plus methotrexate and 36.1% with tocilizumab plus methotrexate. The estimated risk difference was positive for the tocilizumab group, but its confidence interval included zero and extended in both directions. This leaves substantial uncertainty about the size and direction of the underlying difference.

The posted data do not establish that the treatments are equivalent. Failure to show a clear difference is not the same as showing that any difference is small enough to be unimportant. The registry record does not report an equivalence or non-inferiority margin or a design intended to test such a claim. Nor does the result establish that either treatment works or does not work; it is a comparative endpoint result in this trial, not a clinical recommendation.

The result applies to the stated Week 24 endpoint and its described analysis population. It does not by itself establish what happened at other times, for participants excluded from the endpoint analysis, or for outcomes not represented by DAS28 remission. The endpoint uses a cutoff and combines several measurements, so it does not convey every aspect of a participant’s disease activity or response.

The interpretation is also bounded by the study’s open-label design, the stated missing-data handling, the difference between enrollment and reported analysis denominators, and the absence of a reported multiplicity plan. These issues do not automatically invalidate the result, but they limit how strongly it can support a broad conclusion. The registry record also marks the study terminated without reporting the reason.

Overall, the result is a small-group comparison with an uncertain effect estimate. It contributes information about the observed Week 24 remission percentages, but it does not provide a definitive basis for choosing between treatments, predicting an individual outcome, or making a claim about safety or efficacy beyond the reported measures.

17. Why This Trial Matters Statistically

This trial is statistically informative because it makes the interpretive difference between an observed contrast and a precise estimate visible. The tocilizumab group had a higher posted remission percentage, but the unadjusted interval includes zero and accommodates results in either direction. A point estimate alone can suggest a direction; the interval shows why that direction should not be treated as settled.

The record also illustrates how analysis populations affect interpretation. Enrollment is recorded as 96, while the primary endpoint result uses 33 and 36 participants. The registry does not explain the difference. The primary population description includes receipt of study drug, a post-baseline assessment, and baseline and Week 24 data. Readers need to distinguish enrollment from the denominator relevant to a particular endpoint.

Multiple secondary outcomes add breadth but also complicate inference. ACR response thresholds and DAS28-based measures capture related yet different aspects of response. Without a reported multiplicity strategy, secondary results are more appropriately treated as supporting evidence than as a collection of independent confirmatory tests.

The sample-size illustrations also show why planning depends on the effect one wants to detect. Under the listed assumptions, smaller effects require more participants per arm. Those illustrations are distinct from the original trial design, whose sample-size rationale is not reported. Together, the estimate, its uncertainty, endpoint population, and planning assumptions make this a useful example of why statistical interpretation requires more than comparing percentages or focusing on a single p-value.

18. 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.

Several features specific to this record limit interpretation. The study is described as open-label, so participants and study personnel were not masked to treatment assignment. The registry does not report whether outcome assessment was masked or what safeguards were used to limit potential effects of treatment awareness.

The record gives enrollment as 96, but the primary efficacy results use smaller denominators, and the reason is not reported in the registry record. The safety groups list 48 participants at risk each, which also differs from the efficacy analysis sizes. Without an account of participant flow, exclusions, discontinuations, and missing assessments, it is not possible to reconcile these populations or judge whether missingness could have affected the endpoint comparison.

The primary endpoint description specifies last observation carried forward for missing tender and swollen joint counts. It does not report how many values were imputed, the reasons for missing data, whether the other DAS28 components had missing values, or whether sensitivity analyses were performed. The endpoint population description is truncated, leaving some analysis details unavailable.

The registry does not report the specific statistical test or interval method for the primary comparison, an adjustment strategy for multiple outcomes, the randomization procedure, allocation concealment, or the original sample-size justification. It also marks the study terminated without giving the reason. These missing design and conduct details make it harder to assess the intended inference and the possible impact of deviations or incomplete follow-up.

Finally, the primary result is tied to a single Week 24 threshold outcome, and the reported confidence interval is broad. The registry lists many secondary measures, but detailed results are not reported for all of them in the record used for this interpretation. These limitations call for restraint in generalizing the posted comparison beyond the population, endpoint, and follow-up described.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Study of RoActemra/Actemra (Tocilizumab) Versus Adalimumab in Combination With Methotrexate (MTX) in Patients With Moderate to Severe Active Rheumatoid Arthritis And an Inadequate Response to Treatment With Only One Tumor Necrosis Factor (TNF)-Inhibitor — NCT01283971.
• Registry record last updated 2014-02-10; status last verified 2014-01.
• Results first posted 2014-02-10.
• 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.

22. Record Summary

This randomized, open-label, parallel-group phase 4 study compared tocilizumab plus methotrexate with adalimumab plus methotrexate in moderate to severe active rheumatoid arthritis after inadequate response to one TNF inhibitor. Enrollment was 96, and the registry status is terminated. Both groups were treated for an anticipated 24 weeks.

The primary endpoint was Week 24 DAS28 remission, defined as a score <2.6. Reported percentages were 30.3% for adalimumab plus methotrexate and 36.1% for tocilizumab plus methotrexate. The unadjusted risk-difference confidence interval included zero, leaving the comparison imprecise. The registry record does not report the original sample-size justification, full analysis plan, multiplicity strategy, or reason for termination.