This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for A Study of Ocrelizumab Compared to Placebo in Patients With Active Rheumatoid Arthritis Continuing Methotrexate Treatment (STAGE) (NCT00406419), a randomized phase 3 study. The primary outcome is Percentage of Participants With American College of Rheumatology (ACR) 20 Response at Week 24.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00406419 |
| Phase | Phase 3 |
| Condition | Rheumatoid Arthritis |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 1015 |
| Arms in the primary outcome | 3 |
| Lead sponsor | Genentech, Inc. |
| Results posted | Yes |
2. Clinical Question
Population
Participants had active rheumatoid arthritis and an inadequate response to methotrexate. The primary analysis population was the ITT population: randomized participants who received any part of an infusion of study medication. The registry record reports enrollment of 1015 participants.
Intervention
Participants received ocrelizumab with continuing methotrexate. The two regimens were 200 mg or 500 mg by intravenous infusion on Day 1 and Day 15, followed by a repeat course at Weeks 24 and 26. Methotrexate was administered weekly at a stable dose.
Comparator
The comparator was matching intravenous placebo on Day 1 and Day 15, with a repeat course at Weeks 24 and 26. Participants in this group also continued weekly methotrexate at a stable dose.
Primary question
At Week 24, what percentage of participants in each group had an ACR20 response? ACR20 requires the specified improvement in both swollen and tender joint counts and in three of five additional assessments. The posted results compare each ocrelizumab group with the placebo-plus-methotrexate group.
This question concerns a defined clinical response at a specified follow-up time, rather than a measure of change in just one symptom or examination finding. The comparison is between treatment groups that all continued methotrexate, so the posted contrast is for adding the assigned infusion regimen in that study context. The registry record identifies the trial as randomized, double-blind, parallel-group, and Phase 3. The primary results are reported as percentages with 95% confidence intervals, with sponsor analyses also reported for the two comparisons.
3. Trial Design
STAGE was a randomized, double-blind, parallel-group, international Phase 3 study of ocrelizumab compared with placebo in participants with active rheumatoid arthritis who had an inadequate response to methotrexate. The registry record lists enrollment as 1015 and the study status as terminated. The recorded study start was 2006-12-27, and primary completion was 2009-10-06. The reason for termination is not reported in the registry record.
Participants were assigned to one of three groups: matching placebo plus methotrexate, ocrelizumab 200 mg plus methotrexate, or ocrelizumab 500 mg plus methotrexate. Infusions were scheduled for Day 1 and Day 15, with repeat courses at Weeks 24 and 26. Weekly methotrexate was continued at a stable dose. The brief summary describes the anticipated time on study treatment as 1-2 years and gives a target sample size of 1000; the recorded enrollment was 1015.
The primary endpoint was the percentage of participants with an ACR20 response at Week 24. Its defined analysis population included all randomized participants who received any part of an infusion of study medication. Arm-level primary results list 320 participants for placebo plus methotrexate, 343 for ocrelizumab 200 mg plus methotrexate, and 343 for ocrelizumab 500 mg plus methotrexate. These analysis counts need not be identical to enrollment because the endpoint population has its own stated inclusion condition.
The record lists 26 outcome measures and 4 reported analyses. It also lists secondary measures covering clinical response, disease activity, joint counts, and radiographic change. The original statistical plan, including the trial's own sample-size justification and handling of interim analyses, is not reported in the registry record.
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
| Stage | Placebo × 2 IV + MTX | Ocrelizumab 200 mg × 2 IV + MTX | Ocrelizumab 500 mg × 2 IV + MTX |
|---|---|---|---|
| STARTED | 324 | 344 | 347 |
| COMPLETED | 0 | 0 | 0 |
| NOT COMPLETED | 324 | 344 | 347 |
Reasons for not completing this period:
| Reason | Placebo × 2 IV + MTX | Ocrelizumab 200 mg × 2 IV + MTX | Ocrelizumab 500 mg × 2 IV + MTX |
|---|---|---|---|
| Death | 3 | 1 | 4 |
| Protocol Violation | 6 | 2 | 5 |
| Study Terminated by Sponsor | 260 | 288 | 282 |
| Withdrawal by Subject | 17 | 24 | 26 |
| Lack of Efficacy | 12 | 1 | 6 |
| Non-Compliance with Study Drug | 5 | 6 | 3 |
| Adverse Event | 10 | 12 | 16 |
| Lost to Follow-up | 11 | 10 | 5 |
Analysis population for the primary outcome, as defined in the registry: ITT population included all randomised participants who had received any part of an infusion of study medication.
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.
| Characteristic | Unit | Placebo × 2 IV + MTX | Ocrelizumab 200 mg × 2 IV + MTX | Ocrelizumab 500 mg × 2 IV + MTX | Total |
|---|---|---|---|---|---|
| Participants analysed | 320 | 343 | 343 | 1006 | |
| Age, Continuous | Years | 50.5 (SD 11.54) | 51.8 (SD 11.97) | 50.9 (SD 12.23) | 51.1 (SD 11.91) |
| Sex: Female, Male: Female | Participants | 248 | 283 | 282 | 813 |
| Sex: Female, Male: Male | Participants | 72 | 60 | 61 | 193 |
6. Primary Endpoint
The registry lists the primary outcome as Percentage of Participants With American College of Rheumatology (ACR) 20 Response at Week 24, measured over Week 24.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | 95% Confidence Interval |
| Unit | Percentage of participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint is the percentage of participants with an American College of Rheumatology 20 response at Week 24. ACR20 is a composite response definition: it requires the specified improvement in both swollen joint count and tender joint count, plus the specified improvement in three of five additional assessments. Those assessments cover participant-rated pain, participant-rated disease activity, investigator or physician assessment of disease activity, disability measured by HAQ-DI, and an acute-phase reactant, either ESR or CRP.
A composite endpoint is used to represent response across several aspects of a condition, rather than relying on a single measure. In this case, the definition combines joint examination findings with assessments of symptoms, disease activity, disability, and inflammation. Meeting the definition therefore depends on the required pattern across these components, not merely on change in one component.
The registry record reports the endpoint as a percentage and gives a 95% confidence interval for each group. The primary population is the ITT population as defined in the record: all randomized participants who received any part of an infusion of study medication. The endpoint is assessed at Week 24, so it describes response at that assessment time; it does not by itself describe response at every later time or the duration of response. Other listed outcomes include ACR response measures at different thresholds and times, disease activity measures, and joint-count changes, but those do not replace the stated primary endpoint.
7. Results Posted in the Registry
- Ocrelizumab 200 mg × 2 IV + MTX: 56.9% (195/343) versus 35.6% (114/320) on control; risk difference 21.2 percentage points (95% CI 13.8 to 28.6), risk ratio 1.60 (1.34–1.90); unadjusted two-proportion z-test p = <0.001.
- Ocrelizumab 500 mg × 2 IV + MTX: 54.5% (187/343) versus 35.6% (114/320) on control; risk difference 18.9 percentage points (95% CI 11.5 to 26.3), risk ratio 1.53 (1.28–1.83); unadjusted two-proportion z-test p = <0.001.
Risk difference: Ocrelizumab 200 mg × 2 IV + MTX
95% CI: 13.8 to 28.6 percentage points
Recalculated from the posted summary statistics
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.
| Secondary outcome | Type | Effect measure | Estimate (95% CI) | Planning size, 80% power |
|---|---|---|---|---|
| Percentage of Participants With a Major Clinical Response (ACR70 for ≥ 6 Months) at Week 48 | Binary Endpoint | Risk difference | +5.2 pts (2.4 to 7.9) | 198 per arm |
| Percentage of Participants With a Reduction in Modified Total Sharp Score (mTSS) From Baseline at Week 48 | Binary Endpoint | Risk difference | +10.9 pts (5.0 to 16.7) | 189 per arm |
Percentage of Participants With a Major Clinical Response (ACR70 for ≥ 6 Months) at Week 48
- 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:
- Ocrelizumab 200 mg × 2 IV + MTX: 6.1% (21/343) versus 0.9% (3/320) on control; risk difference 5.2 percentage points (95% CI 2.4 to 7.9), risk ratio 6.53 (1.97–21.68); unadjusted two-proportion z-test p = <0.001.
- Ocrelizumab 500 mg × 2 IV + MTX: 7.3% (25/343) versus 0.9% (3/320) on control; risk difference 6.4 percentage points (95% CI 3.4 to 9.3), risk ratio 7.77 (2.37–25.50); unadjusted two-proportion z-test p = <0.001.
- Interpretation:
- The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ocrelizumab 200 mg × 2 IV + MTX than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
- The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ocrelizumab 500 mg × 2 IV + MTX than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
Percentage of Participants With a Reduction in Modified Total Sharp Score (mTSS) From Baseline at Week 48
- 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:
- Ocrelizumab 200 mg × 2 IV + MTX: 22.7% (73/322) versus 11.8% (36/305) on control; risk difference 10.9 percentage points (95% CI 5.0 to 16.7), risk ratio 1.92 (1.33–2.77); unadjusted two-proportion z-test p = <0.001.
- Ocrelizumab 500 mg × 2 IV + MTX: 29.8% (98/329) versus 11.8% (36/305) on control; risk difference 18.0 percentage points (95% CI 11.9 to 24.1), risk ratio 2.52 (1.78–3.58); unadjusted two-proportion z-test p = <0.001.
- Interpretation:
- The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ocrelizumab 200 mg × 2 IV + MTX than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
- The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ocrelizumab 500 mg × 2 IV + MTX than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
The posted secondary outcomes span several related dimensions: sustained major clinical response, disease activity and remission, categorical disease-activity response, ACR response thresholds, joint counts, and radiographic change. Across the secondary results reported in detail, the group summaries generally follow a pattern in which the ocrelizumab groups have higher percentages of participants meeting the specified binary outcomes than the placebo-plus-methotrexate group. This broad pattern is related to the primary result because several measures concern clinical response or disease activity, while others capture different aspects of the condition.
Secondary endpoints are not substitutes for the primary endpoint. They use different definitions, time points, and, in some cases, analysis populations or denominators. A consistent direction across related measures can add context, but it does not make every outcome equally confirmatory. The secondary findings should be interpreted in light of the primary question and the endpoint-specific uncertainty, rather than treated as a single combined result.
Multiplicity matters because the record lists many outcome measures and reports several comparisons. When many outcomes are examined, the chance of finding apparently notable results by chance can increase if the analysis does not account for that multiplicity. The registry record does not report a multiplicity adjustment or a hierarchical testing plan for the secondary outcomes. Unadjusted secondary results therefore carry less confirmatory weight than an outcome supported by a prespecified testing strategy with a reported multiplicity plan. The full set of secondary results should be regarded as supporting context, not as independent proof for each separate claim.
Secondary outcomes registered for this trial, for reference (only those above were recalculated):
- Percentage of Participants With a Major Clinical Response (ACR70 for ≥ 6 Months) at Week 48
- Percentage of Participants Achieving Disease Activity Score 28 (DAS28) Remission (DAS28 < 2.6) at Weeks 24 and 48
- Change From Baseline in DAS28 at Weeks 24 and 48
- European League Against Rheumatism (EULAR) Response Rates (Categorical DAS Responders) at Weeks 24 and 48
- Percentage of Participants Achieving an ACR50 Response at Weeks 24 and 48
- Percentage of Participants Achieving an ACR70 Response at Weeks 24 and 48
- Percent Change From Baseline in Swollen Joint Count (SJC) at Weeks 24 and 48
- Change From Baseline in Tender Joint Count (TJC) at Weeks 24 and 48
9. Statistical Interpretation
At Week 24, the recorded ACR20 response percentage was 35.7% in the placebo-plus-methotrexate group, with a 95% confidence interval of 30.5% to 41.0%. The corresponding percentages were 56.9% for ocrelizumab 200 mg plus methotrexate, with a 95% confidence interval of 51.6% to 62.1%, and 54.5% for ocrelizumab 500 mg plus methotrexate, with a 95% confidence interval of 49.2% to 59.8%. These are observed group summaries at the specified endpoint and time.
The unadjusted recalculated risk differences versus placebo plus methotrexate are 0.212 for the 200 mg group, with a 95% confidence interval of 0.138 to 0.286, and 0.189 for the 500 mg group, with a 95% confidence interval of 0.115 to 0.263. The unadjusted p-values are <0.001 for both comparisons. A risk difference expresses the contrast in the proportion meeting the endpoint; it does not establish why individual participants did or did not meet the response definition.
The sponsor's reported analyses use a Cochran-Mantel-Haenszel method and describe weighted differences. For the comparison of placebo with ocrelizumab 200 mg, the reported weighted difference is 18.5, with a 95% confidence interval of 11 to 25.9 and p < 0.001. For placebo compared with ocrelizumab 500 mg, the reported weighted difference is 21.4, with a 95% confidence interval of 14.1 to 28.8 and p < 0.001. These are sponsor-reported analyses and should be read as distinct from the simple unadjusted recalculations.
The intervals convey uncertainty around the estimated group responses and contrasts. They do not describe the range of outcomes for individual participants. Both the endpoint percentages and the contrast estimates are specific to the recorded analysis population and Week 24 assessment. The results do not establish a comparison between the two ocrelizumab doses: the listed sponsor analyses compare each dose group with placebo, and no dose-versus-dose analysis is reported in the registry record.
These results describe the posted endpoint data and their statistical uncertainty. They do not, by themselves, resolve how missing outcome data were handled, whether additional analyses were prespecified, or how the primary endpoint was situated in a full multiplicity plan; those details are not reported in the registry record. Interpretation should remain tied to the defined response measure, study groups, and analysis population rather than being extended to unmeasured outcomes or other populations.
10. Statistical Methods Explained
The primary outcome is binary at the participant level: each participant either met the ACR20 response definition at Week 24 or did not. The registry reports the percentage meeting the endpoint in each group, together with a 95% confidence interval. A percentage is an estimate of the response proportion in the analyzed group; the interval indicates the statistical uncertainty around that estimate under the method used to construct it. It is not a range containing a stated proportion of individual participants.
What does the risk difference represent?
The recalculated comparison uses an unadjusted risk difference. This is the difference between the observed response proportions in the treatment group and the placebo group, expressed in the calculation as a proportion rather than as a percentage. A positive value means the observed proportion was higher in the named treatment group in this dataset. The estimate is accompanied by a confidence interval, which conveys uncertainty about the contrast. These simple calculations use the posted event counts and denominators and do not apply covariate or stratification adjustments.
The unadjusted recalculations are not the only analyses reported. The registry record gives sponsor analyses using the Cochran-Mantel-Haenszel method, with a weighted difference, confidence interval, and p-value for each comparison. This method can combine information across strata to form a weighted comparison. The specific stratification variables and details of implementation are not reported in the registry record, so the reported weighted differences should not be reverse-engineered into an assumed analysis plan. A weighted estimate and a simple unadjusted estimate can differ because they are not necessarily estimating the contrast in the same way.
How should the p-values be read?
A p-value describes how compatible the observed data are with a specified null hypothesis, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. The posted unadjusted p-values and sponsor-reported p-values are both small relative to conventional thresholds, but statistical evidence should be read alongside the estimated contrast, its confidence interval, the analysis method, and the wider plan for evaluating multiple outcomes. That wider multiplicity plan is not reported in the registry record.
What is the role of randomization and blinding?
Random allocation is intended to make treatment groups comparable on average, including with respect to factors that are measured and factors that are not measured. Double masking is intended to limit the influence of treatment knowledge on participant experience, care, and outcome assessment. These design features support interpretation of between-group comparisons, but they do not remove uncertainty or replace details about the analysis population, missing data, or protocol implementation. The registry identifies the trial as randomized and double-blind but does not report further allocation or masking procedures.
What does the endpoint definition contribute?
ACR20 is a composite rule requiring improvement in both swollen and tender joint counts and in three of five additional assessments. Converting a set of clinical measures into a binary response makes a group-level response percentage possible, but it also means that participants near the response threshold may be classified differently from participants who do not meet the full rule. The endpoint summarizes the specified components; it does not display the separate component results or explain how each component changed. The registry's primary endpoint definition and ITT population are reported, but details of the missing-data and imputation methods are not reported in the registry record.
The reported primary analysis therefore combines a defined binary outcome, a stated analysis population, arm-level proportions, confidence intervals, and comparative analyses. Recalculated unadjusted results are a useful descriptive contrast from posted summary counts; the sponsor's own reported method remains a separate analysis with its own weighting approach. Neither should be treated as a complete account of all possible sensitivity analyses, since those analyses and the full statistical plan are not reported in the registry record.
11. Confidence Intervals and Precision
The primary group estimates have 95% confidence intervals. For placebo plus methotrexate, the response percentage is 35.7%, with an interval from 30.5% to 41.0%. For ocrelizumab 200 mg plus methotrexate, it is 56.9%, with an interval from 51.6% to 62.1%. For ocrelizumab 500 mg plus methotrexate, it is 54.5%, with an interval from 49.2% to 59.8%. The intervals show that each estimate has uncertainty; they do not imply that the true value is equally likely to be anywhere in the interval.
For the unadjusted risk differences, the interval is 0.138 to 0.286 for the 200 mg comparison and 0.115 to 0.263 for the 500 mg comparison. Each interval is reported with its corresponding point estimate. The sponsor's weighted-difference intervals are reported separately: 11 to 25.9 for the 200 mg comparison and 14.1 to 28.8 for the 500 mg comparison. These sponsor intervals correspond to the reported Cochran-Mantel-Haenszel weighted differences, not to the unadjusted recalculations.
Confidence intervals convey precision, not the probability that a treatment effect is true or the probability that a future study will reproduce the estimate. Their width reflects uncertainty under the analysis and data used. The intervals also do not address possible bias from missing data, selective reporting, or the choice among multiple outcomes. Those issues require information about the statistical plan and conduct that is not reported in the registry record. The posted intervals should therefore be interpreted together with the stated analysis population, endpoint definition, and analysis method.
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.
| Effect to detect | Participants per arm, 80% power | Participants per arm, 90% power |
|---|---|---|
| 100% of the observed effect | 86 | 114 |
| 75% of the observed effect | 152 | 202 |
| 50% of the observed effect | 336 | 449 |
The sample-size illustration gives the number of participants per arm for the binary outcomes described in the record. It is not a statement of the trial's original design justification. For the binary outcomes, the figures are participants per arm, not total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The assumed effect to detect is a stated fraction of the observed effect.
The 80% power column describes the approximate number of participants per arm under which a study planned around those assumptions would have an 80% chance of detecting the specified effect, if that effect and the assumptions were correct. The 90% power column uses a higher probability of detecting that same specified effect, and therefore generally calls for a larger planned sample. For each comparison, the illustration considers planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. The entries are planning illustrations rather than results from a new trial.
Smaller effects require more participants to distinguish them statistically under otherwise unchanged assumptions. This is why the planning figures increase when the effect being targeted is a smaller fraction of the observed effect. The observed effect is an uncertain estimate from the present trial, so planning around it does not guarantee that a future study will see the same effect.
These figures are not observed power. The trial's original sample-size assumptions and justification are not reported in the registry record. The planning illustration should be read as a conditional scenario based on its stated calculation assumptions, not as an explanation of why this trial enrolled its recorded 1015 participants or as evidence about the probability that the trial's result is correct.
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 z1−α/2 = 1.960) and the desired power (z1−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
- Control proportion p0 = 0.356; treatment proportion p1 = 0.569; average p̄ = 0.462.
- First term: z1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7051 = 1.3820.
- Second term: z1−β × √[p0(1−p0) + p1(1−p1)] = 0.842 × 0.6889 = 0.5798.
- n = (1.3820 + 0.5798)2 / (0.2123)2 = 85.4, rounded up to 86 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 <- 0.35625 # control proportion
p1 <- 0.568513119533528 # 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(406419)
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 = 86 (80% power), 114 (90% power)
75% of observed effect: n per arm = 152 (80% power), 202 (90% power)
50% of observed effect: n per arm = 336 (80% power), 449 (90% power)
power.prop.test: 85.4
Simulated power at n = 86 per arm: 0.818 (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. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Cochran-Mantel-Haenszel | Weighted difference 18.5 | 11 to 25.9 | < 0.001 |
| Cochran-Mantel-Haenszel | Weighted difference 21.4 | 14.1 to 28.8 | < 0.001 |
The sponsor-reported primary analyses use a Cochran-Mantel-Haenszel method and report weighted differences. The comparison of placebo plus methotrexate with ocrelizumab 200 mg plus methotrexate has a reported weighted difference of 18.5, a 95% confidence interval of 11 to 25.9, and p < 0.001. The corresponding comparison with ocrelizumab 500 mg plus methotrexate has a reported weighted difference of 21.4, a 95% confidence interval of 14.1 to 28.8, and p < 0.001.
The recalculated comparisons are simple unadjusted risk differences based on the posted summary counts. They are 0.212, with a 95% confidence interval of 0.138 to 0.286, for the 200 mg comparison and 0.189, with a 95% confidence interval of 0.115 to 0.263, for the 500 mg comparison. The sponsor's weighted differences and the unadjusted risk differences arise from different analysis approaches and should not be treated as interchangeable estimates.
The registry record does not report the details needed to reconstruct the sponsor's weighting or stratification. The sponsor analysis is therefore presented as reported, while the recalculation serves as a descriptive, unadjusted check using posted summary statistics. The record does not flag a discrepancy between these results.
15. Multiplicity, Missing Data and Analysis Populations
The registry record lists 26 outcome measures and reports 4 analyses. It also includes secondary outcomes covering clinical response, disease activity, joint counts, and radiographic change. Evaluating several outcomes creates a multiplicity issue: without an appropriate plan, the chance of at least one apparently notable result arising by chance can increase as more comparisons are considered. The registry record does not report whether the study used a family-wise error procedure, a hierarchical testing sequence, or another multiplicity strategy. It is therefore not possible to determine from the record how the secondary p-values should be interpreted as a family of tests.
The primary analysis population is described as all randomized participants who received any part of an infusion of study medication. That definition is informative, but it does not provide all details about how participants with missing Week 24 measurements or other post-randomization events were handled. Missing-data methods, imputation rules, sensitivity analyses, and the handling of participants who discontinued are not reported in the registry record. The brief summary says that the study was terminated, but the reason is not reported in the registry record.
These omissions do not establish that a particular method was or was not used. They limit what can be concluded from the registry account about the robustness of the analysis and the control of false-positive findings across outcomes. The primary endpoint result should be interpreted using its stated population, endpoint definition, and reported analyses, while recognizing that the complete analysis plan and missing-data details are not reported in the registry record.
16. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Placebo × 2 IV + MTX | 73 | 318 | 23.0% |
| Ocrelizumab 200 mg × 2 IV + MTX | 61 | 343 | 17.8% |
| Ocrelizumab 500 mg × 2 IV + MTX | 78 | 345 | 22.6% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Placebo × 2 IV + MTX | 5 of 318 |
| Ocrelizumab 200 mg × 2 IV + MTX | 2 of 343 |
| Ocrelizumab 500 mg × 2 IV + MTX | 8 of 345 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0.05% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Placebo × 2 IV + MTX | 228 of 318 |
| Ocrelizumab 200 mg × 2 IV + MTX | 272 of 343 |
| Ocrelizumab 500 mg × 2 IV + MTX | 273 of 345 |
Most frequent serious adverse event terms
| Event term | Organ system | Placebo × 2 IV + MTX | Ocrelizumab 200 mg × 2 IV + MTX | Ocrelizumab 500 mg × 2 IV + MTX |
|---|---|---|---|---|
| Pneumonia | Infections and infestations | 3 of 318 | 4 of 343 | 10 of 345 |
| Urinary tract infection | Infections and infestations | 4 of 318 | 1 of 343 | 3 of 345 |
| Rheumatoid arthritis | Musculoskeletal and connective tissue disorders | 4 of 318 | 1 of 343 | 2 of 345 |
| Cellulitis | Infections and infestations | 3 of 318 | 1 of 343 | 3 of 345 |
| Myocardial infarction | Cardiac disorders | 3 of 318 | 1 of 343 | 2 of 345 |
| Vomiting | Gastrointestinal disorders | 2 of 318 | 3 of 343 | 1 of 345 |
| Renal failure acute | Renal and urinary disorders | 2 of 318 | 3 of 343 | 1 of 345 |
| Histoplasmosis | Infections and infestations | 1 of 318 | 2 of 343 | 3 of 345 |
| Sepsis | Infections and infestations | 4 of 318 | 1 of 343 | 0 of 345 |
| Deep vein thrombosis | Vascular disorders | 2 of 318 | 1 of 343 | 1 of 345 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Placebo × 2 IV + MTX | Ocrelizumab 200 mg × 2 IV + MTX | Ocrelizumab 500 mg × 2 IV + MTX |
|---|---|---|---|---|
| Infusion related reaction | Injury, poisoning and procedural complications | 62 of 318 | 89 of 343 | 96 of 345 |
| Upper respiratory tract infection | Infections and infestations | 64 of 318 | 79 of 343 | 85 of 345 |
| Nasopharyngitis | Infections and infestations | 45 of 318 | 43 of 343 | 51 of 345 |
| Urinary tract infection | Infections and infestations | 40 of 318 | 38 of 343 | 48 of 345 |
| Bronchitis | Infections and infestations | 39 of 318 | 43 of 343 | 40 of 345 |
| Hypertension | Vascular disorders | 26 of 318 | 42 of 343 | 37 of 345 |
| Headache | Nervous system disorders | 24 of 318 | 29 of 343 | 32 of 345 |
| Influenza | Infections and infestations | 20 of 318 | 31 of 343 | 29 of 345 |
| Sinusitis | Infections and infestations | 18 of 318 | 33 of 343 | 28 of 345 |
| Diarrhoea | Gastrointestinal disorders | 23 of 318 | 21 of 343 | 23 of 345 |
Serious events with the highest arm-level rate: pneumonia (2.9%), rheumatoid arthritis (1.3%), urinary tract infection (1.3%), sepsis (1.3%), myocardial infarction (0.9%).
The serious adverse event figures are counts of participants affected among those at risk in each group, as recorded for the safety results. In the placebo-plus-methotrexate group, 73 of 318 participants were affected by a serious adverse event, and 5 deaths are listed. In the ocrelizumab 200 mg-plus-methotrexate group, 61 of 343 participants were affected, and 2 deaths are listed. In the ocrelizumab 500 mg-plus-methotrexate group, 78 of 345 participants were affected, and 8 deaths are listed.
The record also lists the most frequent serious adverse event terms across the safety results as pneumonia at 2.9%, rheumatoid arthritis at 1.3%, urinary tract infection at 1.3%, sepsis at 1.3%, myocardial infarction at 0.9%, and cellulitis at 0.9%. These term frequencies are not accompanied in the record by arm-specific values in the listed summary, so they should not be assigned to a particular group.
Counts alone do not establish causation or explain clinical context, timing, duration, or severity. The denominators for safety results also differ from the primary endpoint arm counts in some groups. These figures are descriptive safety summaries, not a complete evaluation of comparative risk.
17. What the Result Does, and Does Not, Mean
The primary endpoint describes the proportion of participants in each group who met the registry's ACR20 definition at Week 24. The posted group percentages and comparative analyses show the observed results for participants in the stated analysis population. They provide a statistical summary of those data, with confidence intervals to convey uncertainty around the group estimates and contrasts.
ACR20 is a composite threshold, not a complete account of every aspect of rheumatoid arthritis or every participant's experience. A participant who met the definition may have had a different pattern of change across the component assessments from another participant who also met it. A participant who did not meet the full definition may still have changed in one or more components. The registry record does not provide the separate primary-endpoint component results in this summary.
The study compared ocrelizumab regimens with placebo while participants continued methotrexate. Its result therefore pertains to those assigned study groups and this defined endpoint, not to ocrelizumab without methotrexate, to other treatment settings, or to populations not described in the record. The posted results do not establish a dose-versus-dose comparison. They also do not by themselves answer questions about long-term outcomes beyond the specified assessments, individual benefit, or the causes of safety events.
Statistical significance is not the same as clinical importance, and a confidence interval is not a guarantee about future results. Interpretation also depends on the completeness of outcome data, analysis choices, and any multiplicity plan. Missing-data handling and the full statistical plan are not reported in the registry record. The result should therefore be read as a defined comparison with measurable uncertainty and important limits, rather than as a universal claim about treatment effects.
18. Why This Trial Matters Statistically
This trial is statistically informative because it is a randomized, double-blind, parallel-group Phase 3 comparison with three groups and a clearly defined primary endpoint. The endpoint translates several clinical assessments into a binary response at a specified follow-up time, allowing the response proportion to be compared across assigned groups. Reporting arm percentages with confidence intervals makes both the estimates and their uncertainty visible.
The record also illustrates why analysis methods matter. A simple unadjusted risk difference based on group counts is not the same quantity as a Cochran-Mantel-Haenszel weighted difference. Both can be useful descriptions, but they answer the comparison using different approaches. Reading the sponsor analysis alongside the unadjusted recalculation helps distinguish a reported analysis from a summary-statistic check, without treating the two estimates as interchangeable.
The breadth of listed secondary outcomes raises a further statistical issue: multiple outcomes can provide useful context, but their interpretation depends on the testing plan. Because that plan is not reported in the registry record, secondary results deserve more cautious interpretation than a single prespecified primary result would receive under a known multiplicity strategy. The record also leaves missing-data methods and the trial's original sample-size justification unreported.
Finally, the status is recorded as terminated, while the reason is not reported in the registry record. That status is important context, but it does not explain the observed results. Taken together, the trial highlights the value of separating endpoint definition, randomized comparison, uncertainty, analysis method, multiplicity, and missing-data information when interpreting clinical trial statistics.
19. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
- Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.
The registry record does not report the reason the study was terminated. It also does not report the trial's original sample-size justification, the full statistical analysis plan, or any interim-analysis plan. The sample-size illustration is based on the observed effect and stated planning assumptions; it cannot fill in those missing design details.
Although the primary analysis population is defined, the record does not describe how missing Week 24 endpoint data were handled, how discontinuations were treated, or whether sensitivity analyses were conducted. The record also does not provide the stratification details needed to reconstruct the sponsor's Cochran-Mantel-Haenszel weighted analyses. The multiple secondary outcomes are listed, but a multiplicity adjustment or testing hierarchy is not reported.
The safety summaries are descriptive counts and listed terms. They do not establish that an event was caused by a study treatment, and the record excerpt does not give the timing or clinical circumstances needed to interpret individual events. Some safety denominators differ from primary endpoint denominators, so those populations should not be conflated.
These limits concern the information reported in the registry record; they do not show that an unreported procedure was absent. They constrain independent assessment of the analysis choices, robustness to missing data, and the confirmatory status of secondary findings. The primary result remains specific to the stated endpoint, population, groups, and time point.
20. Related Tutorials
21. Related Statistical Calculators
22. Sources
- ClinicalTrials.gov: A Study of Ocrelizumab Compared to Placebo in Patients With Active Rheumatoid Arthritis Continuing Methotrexate Treatment (STAGE) — NCT00406419.
- Registry record last updated 2020-11-27; status last verified 2020-11.
- Results first posted 2020-11-27.
- Data retrieved from ClinicalTrials.gov on 2026-11-09.
- Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
23. Record Summary
The registry record describes NCT00406419, titled “A Study of Ocrelizumab Compared to Placebo in Patients With Active Rheumatoid Arthritis Continuing Methotrexate Treatment (STAGE).” It identifies a randomized, double-blind, parallel-group Phase 3 trial with enrollment of 1015 participants and status recorded as terminated.
Participants with active rheumatoid arthritis and an inadequate response to methotrexate were assigned to placebo, ocrelizumab 200 mg, or ocrelizumab 500 mg, with continued methotrexate. The primary endpoint was the percentage with an ACR20 response at Week 24. The posted percentages were 35.7%, 56.9%, and 54.5%, respectively, with group-specific 95% confidence intervals. Sponsor-reported analyses used the Cochran-Mantel-Haenszel method.
The record lists secondary clinical, disease-activity, joint-count, and radiographic outcomes, along with safety counts. The reason for termination, the original sample-size justification, multiplicity plan, and missing-data methods are not reported in the registry record.