Comparison Of 6 CP-690,550 Doses Vs.Placebo, Each Combined With Methotrexate, For The Treatment Of Rheumatoid Arthritis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparison Of 6 CP-690,550 Doses Vs.Placebo, Each Combined With Methotrexate, For The Treatment Of Rheumatoid Arthritis (NCT00413660), a randomized phase 2 study. The primary outcome is Percentage of Participants Achieving American College of Rheumatology 20% (ACR20) Response at Week 12.

2. Clinical Question

The registry record describes a Phase 2B, randomized, double-blind, placebo-controlled, multicenter study. It states that the study treatment was administered for 6 months and was intended to examine effectiveness and safety in this rheumatoid arthritis population. The central comparison is between each CP-690,550 dose plus methotrexate and placebo plus methotrexate; the endpoint is not a comparison of either regimen with methotrexate alone.

The primary result is a binary response outcome: each participant either met the ACR20 criteria at Week 12 or did not. The posted arm percentages were 47.14 for CP-690,550 1 mg, 55.88 for 3 mg, 56.34 for 5 mg, 58.11 for 10 mg, 56.0 for 15 mg, 56.25 for 20 mg, and 36.23 for placebo. These are observed percentages under the registry’s stated analysis approach, not estimates of an individual participant’s chance of response in another setting.

There are six dose-versus-placebo comparisons. Each has its own estimated risk difference and confidence interval. Those estimates describe the posted group-level contrast, while the interval conveys uncertainty in that contrast. The registry record does not report a multiplicity-adjusted analysis or a broader dose-selection conclusion. The clinical question is therefore best read as a set of dose-specific comparisons at the stated endpoint and time point, rather than as a definitive ranking of doses or a claim about outcomes beyond the recorded measures.

3. Trial Design

The study is recorded as a randomized, parallel-group, placebo-controlled Phase 2 trial with quadruple masking. The official title characterizes it as a Phase 2B, double-blind, multicenter study of six dose regimens, each combined with methotrexate, administered for 6 months. The registry lists enrollment as 509 and status as completed.

The study population was described as adults with active rheumatoid arthritis who had an inadequate response to methotrexate alone. The randomized treatment groups consisted of CP-690,550 1 mg, 3 mg, 5 mg, 10 mg, 15 mg, and 20 mg, alongside placebo. Each regimen was combined with methotrexate. The arm descriptions identify BID for the 1 mg, 3 mg, 5 mg, and 10 mg regimens, 15 mg without a frequency specified there, and 20 mg QD. The placebo description says dummy tablets.

The primary endpoint was ACR20 response at Week 12, although the official title describes treatment over 6 months. The primary population was the Full Analysis Set, defined as randomized participants who received at least one dose of study treatment. Missing values for the primary endpoint were imputed using Baseline Observation Carried Forward (BOCF). The record does not provide a detailed account of the randomization sequence, allocation concealment, or the practical procedures used to maintain masking; these details are not reported in the registry record.

The registry records the study start as 2007-01 and primary completion as 2008-08. It lists rheumatoid arthritis as the condition and Pfizer as sponsor. In addition to the primary outcome, the record includes secondary measures involving ACR20, ACR50, and ACR70 responses; an ACR-n response curve measure; tender joint count; and swollen joint count, including changes at specified follow-up visits. The registry record does not report the original sample-size justification, the planned statistical analysis in detail, or the timing and conduct of interim analyses.

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.

Only the first 6 of 11 groups are shown.

Up To Week 12

Reasons for not completing this period:

Post Week 12

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set (FAS) included all randomized participants who received at least 1 dose of study treatment. Missing values were imputed using Baseline Observation Carried Forward (BOCF).

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.

Only the first 6 of 8 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants Achieving American College of Rheumatology 20% (ACR20) Response at Week 12, measured over Week 12.

The primary endpoint is the percentage of participants achieving an American College of Rheumatology 20% response (ACR20) at Week 12. It is a binary outcome: a participant either meets the response criteria or does not. The record defines response as at least 20% improvement in tender joint count and swollen joint count, together with at least 20% improvement in at least three of five remaining ACR core measures.

Those five measures are participant assessment of pain, participant global assessment of disease activity, physician global assessment of disease activity, self-assessed disability using the disability index of the Health Assessment Questionnaire, and C-Reactive Protein. Combining joint counts with these additional measures makes ACR20 a composite assessment of change across several aspects of rheumatoid arthritis rather than a measure of a single symptom or laboratory result. The endpoint is used to classify whether a participant reached a specified response threshold; it does not describe the size of improvement among participants who responded or the distribution of changes among all participants.

The registry gives the primary population as the Full Analysis Set: randomized participants who received at least one dose of study treatment. Missing endpoint values were imputed using Baseline Observation Carried Forward. That approach carries the baseline observation forward for a missing value; it is an analysis convention, not an observed Week 12 measurement. The posted percentages therefore need to be interpreted in light of this stated population and missing-data rule.

ACR20 is one of several response and disease-measure outcomes listed in the registry, alongside ACR50, ACR70, joint counts, and other measures. ACR20 at Week 12 is the designated primary endpoint; it should not be conflated with those secondary outcomes or with outcomes over the full 6-month treatment period.

7. Results Posted in the Registry

• CP-690,550 1 mg: 47.1% (33/70) versus 36.2% (25/69) on control; risk difference 10.9 percentage points (95% CI -5.4 to 27.2), risk ratio 1.30 (0.87–1.94); unadjusted two-proportion z-test p = 0.192.
• CP-690,550 3 mg: 55.9% (38/68) versus 36.2% (25/69) on control; risk difference 19.7 percentage points (95% CI 3.3 to 36.0), risk ratio 1.54 (1.06–2.25); unadjusted two-proportion z-test p = 0.021.
• CP-690,550 5 mg: 56.3% (40/71) versus 36.2% (25/69) on control; risk difference 20.1 percentage points (95% CI 3.9 to 36.3), risk ratio 1.55 (1.07–2.26); unadjusted two-proportion z-test p = 0.017.
• CP-690,550 10 mg: 58.1% (43/74) versus 36.2% (25/69) on control; risk difference 21.9 percentage points (95% CI 5.9 to 37.8), risk ratio 1.60 (1.11–2.32); unadjusted two-proportion z-test p = 0.009.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary because it classifies each participant as an ACR20 responder or non-responder. A natural summary is the proportion of participants classified as responders in each arm. A between-arm risk difference subtracts the placebo response proportion from the active-dose response proportion. Positive estimates indicate a higher observed response proportion in the dose group than in placebo; a value of zero represents no difference on this scale. The risk difference is an absolute measure, unlike a relative risk, which compares proportions as a ratio.

The posted response percentages and comparison estimates answer related but distinct questions. The arm percentage describes the response proportion within that group. The risk difference describes the contrast with placebo. For example, the posted estimate for CP-690,550 10 mg is 0.219, with a confidence interval from 0.059 to 0.378. That estimate is not a percentage-point figure as printed; converting it to another scale would change its displayed form. The registry also reports relative risks and odds ratios, but the primary comparison values discussed in the results are risk differences.

A confidence interval gives a range of values compatible with the data under the calculation’s assumptions and method. It represents uncertainty in an estimated effect, rather than a range containing a specified proportion of individual participant outcomes. For a risk difference, zero is the value corresponding to no difference. Intervals that include zero are compatible with both a positive and a negative contrast at the interval’s stated confidence level; intervals entirely on one side of zero indicate that the estimate is more precise relative to that null value under the method used. The record does not state the confidence level or detailed method used for these intervals, so those specifics are not reported in the registry record.

The p-values reported with the comparisons are unadjusted. A p-value measures how unusual data at least as incompatible with a null hypothesis would be under that hypothesis and the assumed statistical model. It is not the probability that the null hypothesis is true, nor a measure of effect size. The p-values do not account for the fact that six dose groups are each compared with placebo, based on the information reported in the registry. Whether and how the original analysis addressed multiple comparisons is not reported in the registry record.

The Full Analysis Set is defined as all randomized participants who received at least one dose of study treatment. This definition is the population used for the primary outcome and differs from an analysis that includes every randomized participant regardless of treatment receipt. The registry states that missing values were imputed using Baseline Observation Carried Forward. In general, an imputation rule replaces missing outcome information with a value determined by a stated procedure. BOCF carries a participant’s baseline observation forward. It allows an analysis to include participants with missing follow-up values, but its implications depend on why values are missing and on how plausible the carried-forward assumption is. The registry record does not provide the amount or pattern of missing primary outcome data, so their influence cannot be assessed from the reported summary alone.

Several doses are tested against the same placebo group. This creates a multiplicity issue: when many hypotheses are tested, the chance of at least one apparently unusual result can increase if each comparison is interpreted separately without adjustment. An adjustment can change inferential thresholds or intervals, depending on the planned method. The registry record does not report a multiplicity strategy or detailed statistical analysis plan. The unadjusted estimates and p-values can be described as posted, but they should not be represented as multiplicity-controlled findings.

The trial’s own design assumptions, including any planned effect size or sample-size justification, are not reported in the registry record. A separate planning illustration uses a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It varies the effect to detect by using the stated fraction of the observed effect. These are assumptions for the planning illustration, not evidence about the trial’s original design. The resulting participant counts are planning quantities, not outcomes from the trial.

General statistical context: binary response analyses are often affected by sparse data when event counts are small, and estimates can then be less stable. This is a general consideration, not a reported diagnosis of a problem in this trial. Methods for binary outcomes may also depend on the chosen scale, adjustment variables, missing-data assumptions, and whether comparisons were planned in advance. Those choices are not described in enough detail in the registry record to reconstruct the sponsor’s full analysis.

For safety, the registry presents counts of participants affected by serious adverse events and other adverse events, with at-risk counts for the listed groups. These are descriptive counts, not a comparative safety model. Some groups are labeled as switching from one treatment to another, so the figures should be read according to the named group and its reported at-risk denominator. A count alone does not adjust for duration of exposure or establish that a treatment caused an event. Details of formal comparative safety analyses are not reported in the registry record.

10. Confidence Intervals and Precision

Precision concerns how narrowly the data constrain an estimated effect. The primary comparisons include risk-difference confidence intervals, which show the uncertainty around each point estimate under the method used. For CP-690,550 1 mg, the estimate was 0.109 and its interval ran from -0.054 to 0.272. Because this interval spans zero, it is compatible with a range of contrasts that includes no difference as well as contrasts on either side of zero.

For the other dose groups, the posted intervals were 0.033 to 0.360 for 3 mg, 0.039 to 0.363 for 5 mg, 0.059 to 0.378 for 10 mg, 0.038 to 0.357 for 15 mg, and 0.043 to 0.357 for 20 mg. These intervals are all above zero, but they remain broad relative to their point estimates. Their width indicates that the estimated contrasts are not pinned down to a narrow range. It is not appropriate to read the point estimates as exact treatment effects.

All groups have finite observed sample sizes, and the response percentages are estimates based on those groups. More information or a larger study can sometimes improve precision, but precision also depends on event rates, design, analysis, and missing data. The registry record does not report the confidence-interval method or confidence level, and the reported intervals do not by themselves address multiplicity across the six dose comparisons.

The point estimates are similar in scale for the 3 mg through 20 mg groups, but the intervals overlap substantially. Overlap is not itself a formal test of differences between doses, and these placebo comparisons cannot establish that the doses have the same effect. Nor can the results identify a uniquely best dose without a prespecified and appropriate comparative analysis. The precision shown in the posted estimates is limited to the endpoint, population, time point, and imputation approach stated in the record.

11. 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 planning figures are participants per arm for 80% power and participants per arm for 90% power. They are not total participants across both arms and do not represent the enrolled group sizes. The registry’s planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect: 100%, 75%, or 50%.

In plain language, the 80% power column gives an illustrative number of participants per arm for a design with an 80% chance of detecting the specified effect under those assumptions, if that effect were the true effect and the model assumptions held. The 90% power column gives the corresponding illustrative number for a design with a 90% chance under the same conditions. These are design calculations, not a description of the probability that the completed trial detected a real effect. This is not observed power.

The rows that use a smaller fraction of the observed effect require more participants per arm than the rows based on the full observed effect. That pattern reflects a general feature of power calculations: smaller differences are harder to distinguish from random variation, so a larger sample is generally needed to detect them with the same planned power. The counts are illustrative and depend on the assumptions listed, including equal allocation and the normal approximation.

The observed-effect input comes from the posted comparison, and the planning figures therefore should not be treated as an independent forecast or as the trial’s original design justification. The registry record does not report the original sample-size assumptions or its justification. The illustrations also do not incorporate details not listed in their assumptions, such as any multiplicity adjustment, missing-data patterns, or additional analysis features.

12. 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 = 0.362; treatment proportion p₁ = 0.471; average p̄ = 0.417.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6973 = 1.3666.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6930 = 0.5832.
4. n = (1.3666 + 0.5832)² / (0.1091)² = 319.4, rounded up to 320 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 <- 0.36231884057971                          # control proportion
p1 <- 0.471428571428571                          # 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(413660)
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 = 320 (80% power), 428 (90% power)
 75% of observed effect: n per arm = 563 (80% power), 754 (90% power)
 50% of observed effect: n per arm = 1,254 (80% power), 1,678 (90% power)
power.prop.test: 319.4 
Simulated power at n = 320 per arm: 0.795 (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.

13. Multiplicity, Missing Data and Analysis Populations

The primary analysis population is the Full Analysis Set, defined in the registry as randomized participants who received at least one dose of study treatment. This definition is important when interpreting the arm percentages: it is not explicitly described as including randomized participants who never received study treatment. The posted primary endpoint is ACR20 at Week 12.

The registry states that missing primary endpoint values were imputed using Baseline Observation Carried Forward. Under BOCF, baseline observations are carried forward for missing values. This is a simplifying assumption about unobserved follow-up outcomes, not a direct measurement at Week 12. Without information on the amount and reasons for missingness, the effect of this rule on the estimates cannot be determined. The amount and pattern of missing primary data are not reported in the registry record.

The record lists six dose-versus-placebo comparisons for the primary endpoint and reports unadjusted p-values. When multiple comparisons are examined, interpreting each p-value as if it were the sole planned test can increase the risk of false-positive findings across the set. A prespecified multiplicity strategy may address that issue, but the registry record does not report whether an adjustment was used, what method it would have used, or whether the reported values correspond to any adjusted analysis. Thus, the posted p-values should be described as unadjusted rather than as evidence from a documented multiplicity-controlled procedure.

The registry includes secondary outcome measures, including ACR50 and ACR70 responses, ACR-n, and joint-count outcomes. It does not provide a complete statistical analysis plan for the primary or secondary outcomes in this record. The planned hierarchy of hypotheses, any adjustment across secondary endpoints, covariate adjustments, sensitivity analyses, and handling of missing secondary outcomes are not reported in the registry record. These omissions limit assessment of whether the reported findings arose from a prespecified analytical strategy.

The record also does not report interim-analysis procedures or a sample-size justification for the trial’s original design. Those omissions do not alter the posted descriptive results, but they restrict evaluation of the design and inferential context. The planning illustrations are separate calculations and do not fill in the missing information about the original protocol.

14. 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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: uterine haemorrhage (4.8%), chest pain (2.0%), cardiac failure (1.8%), pneumonia (1.8%), urinary tract infection (1.8%).

The registry reports serious adverse event counts and other adverse event counts for named groups, with an at-risk count for each. For example, the CP-690,550 1 mg group lists 1 participant affected by a serious adverse event among 49 at risk, and 28 affected by other adverse events among 49 at risk. The placebo group lists 0 affected by serious adverse events among 51 at risk, and 29 affected by other adverse events among 51 at risk. These are group-specific counts, not adjusted comparisons.

Other listed serious-event counts include 4 among 55 at risk for CP-690,550 3 mg, 4 among 71 at risk for 5 mg, 1 among 74 at risk for 10 mg, 6 among 75 at risk for 15 mg, and 4 among 67 at risk for 20 mg. The registry also lists groups whose names indicate a switch to CP-690,550 5 mg; their figures should be read as separate named groups rather than merged with the original dose groups.

The record lists serious-event terms with percentages, including uterine haemorrhage at 4.8%, chest pain at 2.0%, and cardiac failure, pneumonia, urinary tract infection, and arthralgia each at 1.8%. The registry does not report deaths in the safety data. Counts and listed terms do not establish causation or account for exposure time, and formal comparative safety analyses are not reported in the registry record.

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

The posted data describe ACR20 response at Week 12 in a randomized study of CP-690,550 dose regimens combined with methotrexate versus placebo combined with methotrexate. The observed response percentages were higher in each CP-690,550 arm than in the placebo arm. The risk-difference estimates were positive for all six dose comparisons, while the confidence interval for the 1 mg comparison included zero and the intervals for the other dose comparisons did not.

Those summaries do not establish that every dose has a clinically important effect, that one dose is best, or that the doses are equivalent. The registry record does not report a clinical threshold for an important ACR20 difference, an equivalence margin, or a formal dose-ranking analysis. The p-values are unadjusted, and the record does not report a multiplicity strategy for the several comparisons. Interpretations based on individual p-values should therefore remain cautious.

The results apply to the endpoint, time point, population, and missing-data rule specified in the record. They concern ACR20 at Week 12 in the Full Analysis Set, with missing primary values imputed using BOCF. They do not, by themselves, describe longer-term disease control, quality of life, all components of the composite separately, or outcomes in people who differ from the study population. Other secondary outcomes are listed in the registry, but they are not substitutes for the primary result.

The study’s safety counts are descriptive. They can indicate how many participants in the named groups were recorded as affected, but they do not alone show whether events were caused by treatment or how the overall balance of benefit and harm compares. The record does not include enough information to reconstruct a full comparative safety analysis.

More generally, a registry result is a report of what was recorded and how the outcome was defined; it is not a complete account of every protocol or analysis decision. Original sample-size assumptions, detailed statistical methods, and multiplicity handling are not reported in the registry record. The posted estimates should be read within those limits and not as a treatment recommendation.

16. Why This Trial Matters Statistically

This trial raises several important statistical issues common to multi-arm clinical studies. Six active dose regimens were compared with a shared placebo group. That structure can provide information across a dose range, but it also creates several related comparisons. The effect estimates and unadjusted p-values need to be distinguished from evidence generated under an explicitly multiplicity-controlled plan, which is not reported in the registry record.

The endpoint is a composite binary response. It combines improvement in tender and swollen joint counts with improvement in at least three other ACR core measures. This creates an interpretable responder classification, but it also compresses information: participants just meeting the threshold and participants with larger improvements are both classified as responders. The posted percentage therefore answers a threshold-based question rather than describing the full distribution of change.

The reported risk differences help express the absolute contrast between each dose group and placebo. Their confidence intervals make uncertainty visible, which is essential when comparing estimates based on finite groups. In these results, the point estimates are positive across dose groups, but the interval for the 1 mg comparison includes zero. The intervals also overlap, and the registry does not report a formal dose-to-dose analysis. These features argue against turning the observed ordering of estimates into a firm dose ranking.

Missing-data handling is another key element. The primary analysis used a defined Full Analysis Set and BOCF for missing values. That makes the analysis population and imputation rule part of the meaning of the endpoint estimates; the extent and pattern of missing data are not reported in the registry record. For readers assessing the evidence, this limits evaluation of how robust the results are to alternative assumptions.

Finally, illustrative sample-size calculations based on observed effects are not a substitute for the original design rationale. They can show how planning requirements change when the target effect changes, but the registry record does not report the original justification. Together, the multiple dose comparisons, threshold endpoint, missing-data rule, and incomplete reporting of the analysis plan make careful statistical interpretation particularly important.

17. 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 detailed statistical analysis plan, including the method used to calculate the confidence intervals, any covariate adjustment, or a multiplicity procedure for the six dose comparisons. The p-values are explicitly unadjusted. Without the original analysis plan and its decision rules, the inferential status of the collection of comparisons cannot be fully assessed.

The primary endpoint uses BOCF for missing values, but the registry does not report the number or pattern of missing Week 12 observations, reasons for missingness, or sensitivity analyses using alternative assumptions. It is therefore not possible from the listed summary results to judge how much the primary estimates depend on the imputation rule.

The reported population excludes randomized participants who did not receive at least one dose, under the registry’s Full Analysis Set definition. The record does not give enough detail to assess the difference between that analysis population and other possible analysis populations. Nor does it report the original sample-size rationale or the protocol’s assumptions about the effect to detect. The separate planning illustrations use the observed effect and should not be mistaken for the study’s own design calculations.

The primary endpoint is measured at Week 12, while the study is described as administering treatment for 6 months. The primary result alone does not establish the pattern of response over the full treatment period. Although the registry lists secondary measures, this primary-results record does not provide their complete numerical results or detailed analysis methods.

Safety information is reported as counts for named groups, including groups with treatment switches. The record does not provide a formal comparative safety analysis, exposure-adjusted estimates, or death counts. Consequently, the displayed safety summaries cannot establish comparative risk or causality. These reporting limitations are specific to what the registry record contains and should be considered alongside the endpoint and analysis limitations.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Comparison Of 6 CP-690,550 Doses Vs.Placebo, Each Combined With Methotrexate, For The Treatment Of Rheumatoid Arthritis — NCT00413660.
• Registry record last updated 2013-01-18; status last verified 2013-01.
• Results first posted 2013-01-08.
• 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.

21. Record Summary

The registry record describes a completed, randomized, parallel, quadruple-masked Phase 2 study sponsored by Pfizer. It enrolled 509 and studied six CP-690,550 dose regimens, each combined with methotrexate, against placebo combined with methotrexate in adults with active rheumatoid arthritis and inadequate response to methotrexate alone. The recorded study start was 2007-01, and primary completion was 2008-08.

The primary endpoint was ACR20 response at Week 12 in the Full Analysis Set, defined as randomized participants who received at least one dose. Missing values were imputed using BOCF. Posted response percentages ranged from 47.14 to 58.11 across the active dose groups and were 36.23 for placebo. Risk differences were positive for all dose comparisons; the 1 mg confidence interval included zero, while the others were above zero.

The registry lists secondary outcomes and descriptive safety counts, but does not report the original sample-size justification, detailed analysis plan, or multiplicity procedure. These limits matter when interpreting the precision and inferential meaning of the posted comparisons.