A Study in Participants With Rheumatoid Arthritis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study in Participants With Rheumatoid Arthritis (NCT01202773), a randomized phase 3 study. The primary outcome is Percentage of Participants With American College of Rheumatology 20% (ACR20) Response.

2. Clinical Question

The study was designed to evaluate efficacy and safety in a population whose rheumatoid arthritis had not responded adequately to one or more TNF-α inhibitors. The primary outcome is a response proportion, so it describes how many participants met a predefined composite threshold rather than the average change in a single symptom or measurement. The registry record also lists secondary outcomes covering ACR response, joint counts, pain, global assessments, and physical function.

The posted percentages are descriptive results for the trial’s assigned groups. Recalculated unadjusted comparisons give a separate way to express the observed differences and their uncertainty. Neither the percentages nor these comparisons establish that a regimen caused a difference. The record does not report a complete statistical analysis plan, and the endpoint population and missing-data rules affect how the percentages should be understood.

3. Trial Design

This was a Phase 3, multicenter, randomized, parallel-group study of LY2127399 in participants with moderate to severe rheumatoid arthritis and an inadequate response to one or more TNF-α inhibitors. The official title describes the study as double-blind and placebo-controlled; the registry’s masking classification is triple. The record does not explain how the masking classifications relate to one another. Eli Lilly and Company is listed as the sponsor.

The study included three randomized groups: 120 mg LY2127399 every 4 weeks, 90 mg LY2127399 every 2 weeks, and placebo every 2 weeks. The active-treatment descriptions report a loading dose at initiation. The planned treatment period was 24 weeks, followed by a 48-week post-treatment follow-up period. The primary outcome was assessed from baseline through Week 24. The record describes Week 16 response-based treatment handling and defines nonresponders and participants who discontinued treatment as nonresponders from that point onward for the primary endpoint.

The registry reports 456 participants enrolled. The primary endpoint summaries include 153 participants in the 120 mg group, 148 in the 90 mg group, and 155 in the placebo group. The safety summaries use different group sizes for the randomized treatment period. These counts reflect the populations used for different posted summaries and should not be assumed to represent identical analysis populations.

The study began in 2011-01, with primary completion in 2013-03, and its status is terminated. The reason for termination and the date or circumstances of that decision are not reported in the registry record. The original design’s sample-size rationale, planned interim analyses, and detailed statistical methods are also 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.

Before assignment: Study had a treatment period (Weeks 0-24) and a post-treatment follow-up period (24-48 weeks in length). All participants were assessed for nonresponse at Week 16 with non-responders (NR) defined as participants with \<20% improvement from baseline in both tender and swollen joint counts.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All randomized participants with evaluable ACR20 responder data. If participant's CRP was missing, last post-baseline value was used. If ACR was missing after carrying forward CRP, last post-baseline ACR response was used. Data after Week 16 for Week 16 NR were not included.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: All randomized participants, including participants who did not take the assigned treatment, did not receive the correct treatment, or otherwise did not follow the protocol.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With American College of Rheumatology 20% (ACR20) Response, measured over Baseline through Week 24.

The primary endpoint was the percentage of participants with an American College of Rheumatology 20% response, or ACR20, from baseline through Week 24. ACR20 is a composite response definition: it required at least a 20% improvement in both the 68-joint tender joint count and the 66-joint swollen joint count, plus at least a 20% improvement in three of five additional criteria. Those criteria covered participant and physician global assessments, HAQ-DI, joint pain, and C-reactive protein.

A composite endpoint combines information from several domains of disease activity and function into a single responder classification. This can capture a broader pattern than relying on one measurement alone, but it also means that the result depends on meeting the full specified combination of criteria. A participant who does not meet that definition is not counted as an ACR20 responder, even if some individual measures changed.

The registry defines the percentage as the number of ACR20 responders divided by the number treated. It further specifies that participants who were nonresponders at Week 16, or who discontinued study treatment at any time for any reason, were counted as nonresponders from that point forward, including at Week 24. For missing C-reactive protein, the last post-baseline value was used; if the ACR response remained missing after that step, the last post-baseline ACR response was used. Data after Week 16 for Week 16 nonresponders were not included. These rules matter because the reported response percentage depends on both the response definition and the handling of missing or post-discontinuation information.

7. Results Posted in the Registry

• 120 mg LY2127399: 17.6% (27/153) versus 20.0% (31/155) on control; risk difference -2.4 percentage points (95% CI -11.1 to 6.4), risk ratio 0.88 (0.55–1.40); unadjusted two-proportion z-test p = 0.597.
• 90 mg LY2127399: 24.3% (36/148) versus 20.0% (31/155) on control; risk difference 4.3 percentage points (95% CI -5.0 to 13.7), risk ratio 1.22 (0.80–1.86); unadjusted two-proportion z-test p = 0.365.

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.

American College of Rheumatology Percent Improvement (ACR-N)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • 120 mg LY2127399: mean -9.03 versus -12.72 on control; difference 3.69 (95% CI -7.40 to 14.78), standardized effect d = 0.07; z-test p = 0.514.
  • 90 mg LY2127399: mean -8.03 versus -12.72 on control; difference 4.69 (95% CI -6.55 to 15.93), standardized effect d = 0.09; z-test p = 0.413.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between 120 mg LY2127399 and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between 90 mg LY2127399 and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline to Week 24 in Tender Joint Count (68 Joint Count)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • 120 mg LY2127399: mean -6.37 versus -6.56 on control; difference 0.19 (95% CI -3.66 to 4.04), standardized effect d = 0.01; z-test p = 0.923.
  • 90 mg LY2127399: mean -6.08 versus -6.56 on control; difference 0.48 (95% CI -3.40 to 4.36), standardized effect d = 0.03; z-test p = 0.808.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between 120 mg LY2127399 and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between 90 mg LY2127399 and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline to Week 24 in Swollen Joint Count (66 Joint Count)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • 120 mg LY2127399: mean -6.02 versus -5.41 on control; difference -0.61 (95% CI -3.33 to 2.11), standardized effect d = -0.05; z-test p = 0.660.
  • 90 mg LY2127399: mean -5.64 versus -5.41 on control; difference -0.23 (95% CI -2.97 to 2.51), standardized effect d = -0.02; z-test p = 0.870.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between 120 mg LY2127399 and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between 90 mg LY2127399 and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists a range of secondary outcomes addressing ACR response, ACR-N, tender and swollen joint counts, pain, global assessments, and physical function. Detailed numeric summaries are available for ACR-N and the two joint-count measures. Across those available summaries, the unadjusted between-group intervals include zero, and the posted comparisons do not establish a clear separation from placebo.

These secondary outcomes provide context for the primary ACR20 result because they examine related but distinct aspects of rheumatoid arthritis response. ACR-N is a continuous improvement measure, while joint-count outcomes describe average changes in specific counts. Pain, global assessments, and HAQ-DI cover other patient or clinician assessments and function. Agreement across measures can sometimes support a coherent interpretation, but the outcomes are not interchangeable: a change in a mean does not directly answer whether participants met the ACR20 responder definition.

The available secondary pattern does not overturn the primary result. It likewise does not independently confirm a treatment-placebo difference. Several listed secondary outcomes do not have detailed numerical comparisons reported in the registry record, limiting assessment of whether findings were consistent across the full outcome set.

Unadjusted secondary results carry less weight because many outcomes and comparisons create multiple opportunities for apparently notable findings. The registry record does not report a prespecified hierarchy, multiplicity adjustment, or complete analysis plan. Without those details, secondary results are best treated as descriptive or exploratory rather than as standalone confirmatory evidence. They should be interpreted alongside the primary endpoint definition, missing-data rules, and the uncertainty in the comparative estimates.

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

• Percentage of Participants With American College of Rheumatology 50% (ACR50) and 70% (ACR70) Response
• American College of Rheumatology Percent Improvement (ACR-N)
• Change From Baseline to Week 24 in Tender Joint Count (68 Joint Count)
• Change From Baseline to Week 24 in Swollen Joint Count (66 Joint Count)
• Change From Baseline to Week 24 in Participant's Assessment of Pain [Visual Analog Scale (VAS)]
• Change From Baseline to Week 24 in Participant's Global Assessment of Disease Activity (VAS)
• Change From Baseline to Week 24 in Physician's Global Assessment of Disease Activity (VAS)
• Change From Baseline to Week 24 in Health Assessment Questionnaire-Disability Index (HAQ-DI)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary at the participant level: each participant is classified as an ACR20 responder or nonresponder. The arm-level result is a percentage. The registry reports the percentages in the three groups, while the comparative risk differences and confidence intervals described in this analysis are unadjusted recalculations from posted summary counts. They should not be confused with a sponsor-reported model or an adjusted analysis. The discrepancy flag is false, so there is no identified conflict requiring a sponsor analysis to take precedence.

What does a risk difference describe?

A risk difference expresses the difference between two event proportions. For these comparisons, the event is achieving an ACR20 response, and placebo is the reference group. A positive estimate means the response proportion was higher in the LY2127399 group in the posted summary; a negative estimate means it was lower. The estimate is a point summary, not a statement about certainty. The confidence interval describes a range of values compatible with the data and the calculation’s assumptions.

How should confidence intervals and p-values be read?

In general, a confidence interval conveys the uncertainty around an estimated effect. An interval that includes zero for a difference means that zero difference is among the values compatible with the data under the interval procedure. It can also include effects in either direction. A p-value summarizes how unusual data at least as discrepant as those observed would be under a specified null model. It does not give the probability that the null hypothesis is true, nor does it measure the size or clinical importance of an effect. The p-values reported for the ACR20 contrasts are unadjusted recalculations.

There is no reported multiplicity adjustment for the primary and secondary outcomes in the registry record. When multiple outcomes or comparisons are examined, the chance of obtaining at least one apparently notable result can increase. General statistical practice therefore treats unadjusted secondary comparisons cautiously, especially when the analysis plan and hierarchy of outcomes are not reported. No particular adjustment method can be attributed to this study based on the registry record.

What do the secondary numeric summaries represent?

For ACR-N and the joint-count outcomes, the registry reports least-squares means and standard errors. A least-squares mean is a model-based adjusted mean, rather than necessarily the simple arithmetic average of individual observations. The registry record does not provide enough detail to identify the adjustment model or its covariates. A standard error describes the estimated uncertainty of a mean, not the spread of individual participant measurements. The comparative mean differences and associated intervals in the displayed results are recalculated from posted summaries; the pooled standard deviations used for planning illustrations are approximate.

For changes in joint counts, negative values indicate a decrease from baseline in the measured count. The group-level least-squares means describe the estimated average change for each group, while the mean difference contrasts two groups. A group’s within-group change is not itself evidence of a treatment-versus-placebo difference. The relevant comparative interpretation depends on the between-group estimate and its uncertainty.

How are missing values handled for ACR20?

The primary endpoint description specifies a particular sequence for missing data. If C-reactive protein was missing, the last post-baseline value was used. If ACR response was still missing after carrying forward C-reactive protein, the last post-baseline ACR response was used. Week 16 nonresponders were not assessed using data after Week 16 for the endpoint, and participants who discontinued treatment were counted as nonresponders from discontinuation onward. These are important parts of the endpoint definition, not incidental details. The registry record does not report sensitivity analyses using alternative missing-data assumptions.

The primary endpoint population is described as randomized participants with evaluable ACR20 responder data. That description differs from simply assuming that every randomized participant contributed complete endpoint information. It is important to distinguish randomized enrollment from the population contributing to a particular summary. The registry record does not report a full flow of exclusions and missing observations for every outcome.

What do the planning illustrations mean?

The sample-size illustrations use a two-sided significance level of 0.05, equal allocation between two groups, and a normal approximation. The effect to detect is set at the stated fraction of the observed effect. For binary outcomes, the listed values are participants per arm. These are generic planning illustrations based on the observed effect, not the trial’s own design calculation. The original sample-size assumptions and justification are not reported in the registry record.

In general, planning calculations estimate how many participants may be needed to detect a specified effect under stated assumptions. They do not establish that a study with that size will produce a particular result, and they do not correct for unreported features of a trial’s design or analysis. The displayed figures should therefore be read as conditional illustrations, not as reconstruction of the sponsor’s original plan.

The registry record reports no analysis entries in its analysis listing. The numerical outcomes and arm summaries are present, but detailed sponsor models, covariate adjustments, prespecified contrast methods, and multiplicity handling are not reported in the registry record. Where these details are absent, interpretation should remain tied to the posted summaries and their stated definitions rather than assuming a protocol or model that is not documented.

11. Confidence Intervals and Precision

The ACR20 estimates are not highly precise. For 120 mg LY2127399 versus placebo, the recalculated risk difference is -0.024 and its 95% confidence interval runs from -0.111 to 0.064. For 90 mg LY2127399 versus placebo, the estimate is 0.043 and the interval runs from -0.050 to 0.137. Each interval crosses zero and extends into both negative and positive values. Thus, the results allow for uncertainty about both the direction and size of the difference.

A confidence interval is not a range containing the true effect with certainty. In general, it is a measure of statistical uncertainty constructed using a specified procedure and assumptions. Wider intervals indicate less precise estimates than narrower intervals. These intervals are based on unadjusted recalculations from the posted responder counts, not on a sponsor-reported adjusted analysis.

The posted group percentages are close enough that the point estimates should not be read without their intervals. The observed values alone do not show whether a modest difference is real or reflects sampling variability. Conversely, intervals that cross zero do not prove that there is no difference or that the groups are equivalent. The record does not report an equivalence or noninferiority framework.

The detailed secondary comparisons also have confidence intervals spanning zero. Their uncertainty reinforces the need to avoid strong conclusions from the point estimates, particularly given the number of outcomes and the absence of a reported multiplicity approach. Precision is limited not only by the widths of the intervals but also by missing information about planned models, outcome hierarchy, and analysis populations. The registry record does not report additional analyses that would narrow or otherwise characterize this uncertainty.

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 illustrations are planning calculations based on the observed effect, not reconstructions of the study’s original design. For the binary ACR20 endpoint, figures are participants per arm, rather than a total across both groups. The stated 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 the stated fraction of the observed effect. The original trial’s own design assumptions and sample-size justification are not reported in the registry record.

The 80% power column means the illustrated number of participants per arm under which a study would have an 80% chance of detecting the selected effect at the stated significance level, if the planning assumptions and effect were correct. The 90% column describes the corresponding planning target for a 90% chance under those assumptions. These are design concepts, not guarantees about an actual future result. A higher power target generally requires more participants for the same effect and assumptions.

The rows for planning for 100%, 75%, and 50% of the observed effect illustrate how the assumed effect size changes the calculated requirement. Smaller effects need more participants to detect under the same assumptions. In some comparisons, the observed effect is small, so the resulting planning figures are very large. Those figures do not imply that such a study is feasible or appropriate; they show the consequence of the stated calculation inputs.

These values are not observed power. They are conditional planning illustrations and do not explain why the trial enrolled its actual participants. The record’s enrollment was 456, while the original sample-size target and its justification are not reported.

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 = 0.200; treatment proportion p₁ = 0.176; average p̄ = 0.188.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5528 = 1.0835.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5526 = 0.4651.
4. n = (1.0835 + 0.4651)² / (-0.0235)² = 4,331.4, rounded up to 4,332 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 <- 0.2                          # control proportion
p1 <- 0.176470588235294                          # 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(1202773)
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 = 4,332 (80% power), 5,799 (90% power)
 75% of observed effect: n per arm = 7,794 (80% power), 10,433 (90% power)
 50% of observed effect: n per arm = 17,742 (80% power), 23,750 (90% power)
power.prop.test: 4331.4 
Simulated power at n = 4,332 per arm: 0.802 (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 reports a primary outcome and numerous secondary outcomes. The registry record does not report how the statistical plan addressed multiplicity: it does not describe an adjustment procedure, a testing hierarchy, or which comparisons were intended to support confirmatory conclusions. When several outcomes are tested, unadjusted p-values do not account for the increased opportunity for chance findings across the full set. This is why unadjusted secondary results warrant particular caution.

The record also does not report a complete analysis plan, including the planned model for each endpoint, covariate adjustments, or the method used for primary comparisons. Detailed analyses are not listed in the registry analysis listing. The recalculated comparisons are therefore useful descriptive summaries of posted group results, but they should not be presented as the sponsor’s prespecified inferential analysis. The discrepancy flag is false; no conflicting sponsor estimate is identified in the record.

Missing-data handling is partly described for the primary ACR20 endpoint. The record specifies carrying forward the last post-baseline C-reactive protein value when needed, then carrying forward the last post-baseline ACR response if it remains missing. It also classifies participants who discontinue treatment as nonresponders from that time onward and excludes post-Week 16 data for Week 16 nonresponders. These choices shape the endpoint definition and can affect the resulting percentages.

The record does not report sensitivity analyses that test alternative assumptions for missing data, nor does it provide a full account of missing observations across all outcomes. It describes the primary endpoint population as randomized participants with evaluable responder data, but does not fully document the analysis population for every secondary measure. The number enrolled should therefore not be assumed to equal the number contributing to each analysis.

The trial is marked terminated, but its reason for termination is not reported in the registry record. No conclusion about the relationship between termination and efficacy, safety, recruitment, or other considerations can be drawn from the status alone.

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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumothorax (4.3%), arthritis (3.0%), rheumatoid arthritis (3.0%), adenocarcinoma gastric (3.0%), myocardial infarction (2.8%).

The safety results are divided into randomized treatment, rescue, and follow-up periods. During the randomized treatment period, serious adverse events were reported for 7 of 153 participants in the 120 mg group, 6 of 147 in the 90 mg group, and 6 of 154 in the placebo group. Other adverse events were reported for 71 of 153, 51 of 147, and 61 of 154, respectively. These are counts within the listed risk groups; they should not be treated as direct comparisons of risk without accounting for the study’s definitions and exposure periods.

The rescue and follow-up summaries have their own group sizes and treatment histories. Those groups are smaller and are not equivalent to the randomized treatment groups, so their counts should not be compared as if they arose from the same randomized contrast. The registry lists serious adverse event terms including pneumothorax, arthritis, rheumatoid arthritis, gastric adenocarcinoma, myocardial infarction, and non-cardiac chest pain, with the posted percentages attached to those terms. The record does not state a denominator for that term listing in the adverse-event summary.

Deaths are not reported in the registry record for these safety groups. The safety summaries describe events recorded during the specified periods; they do not, by themselves, establish that a treatment caused an event or that the groups had equivalent safety.

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

The results describe ACR20 response proportions through Week 24 in a population with moderate to severe rheumatoid arthritis and an inadequate response to one or more TNF-α inhibitors. The posted percentages differ across the two LY2127399 regimens and placebo, but the recalculated confidence intervals for both treatment-placebo risk differences include zero and extend in both directions. The data therefore leave substantial uncertainty about the size and direction of any comparative difference.

This does not mean that the regimens are equivalent to placebo. Demonstrating equivalence would require a prespecified framework and suitable analyses, neither of which is reported in the registry record. Nor does an interval that includes zero prove that a difference is absent. The endpoint estimates are compatible with a range of possible effects, and the record does not supply a sponsor-reported adjusted analysis to resolve that uncertainty.

The primary outcome is a composite responder measure with explicit rules for joint counts and additional criteria. Its result should not be generalized automatically to every individual symptom, functional measure, or longer-term outcome. The secondary outcome summaries cover several related domains, but the detailed unadjusted comparisons do not establish a consistent difference, and the registry record does not report multiplicity handling or a full analysis plan.

The study’s termination is a registry status, not an explanation of why it ended. The reason is not reported in the registry record, so it cannot be linked to efficacy or safety findings. The safety data are also period-specific summaries and do not establish causation or comparative safety on their own.

In short, the posted data show the observed response proportions and event counts under the stated definitions. Their uncertainty, missing analysis details, and endpoint-specific population rules limit the conclusions that can be drawn. They do not support a treatment recommendation or a definitive claim about efficacy or safety.

17. Why This Trial Matters Statistically

This trial is statistically informative as an example of why a response percentage and a comparative effect estimate answer related but different questions. The arm percentages describe the proportion meeting the ACR20 definition in each group. The risk differences quantify the observed contrast with placebo, while confidence intervals convey how uncertain that contrast is. Looking only at the point estimates would omit the substantial uncertainty around the comparisons.

The endpoint’s composite structure also matters. ACR20 combines joint counts with several additional clinical, functional, and laboratory criteria, and classifies participants using a threshold. A threshold outcome can be easy to communicate, but it compresses information: changes that fall short of the defined response are not counted as responses. The secondary continuous outcomes offer complementary information, though their unadjusted comparisons and incompletely reported analysis details limit confirmatory interpretation.

The registry illustrates the importance of transparent analysis reporting. It describes primary missing-data rules and gives arm-level outcomes, but does not report the full statistical plan, sample-size justification, multiplicity strategy, or reason for termination. Without these details, readers cannot determine whether the posted comparisons match all prespecified analyses or how the sponsor intended to handle multiple outcomes.

The planning illustrations further distinguish observed results from design assumptions. They show how required sample sizes depend on the chosen effect and planning targets, but they do not explain the trial’s original enrollment decision. Together, these features make the record useful for understanding estimation, uncertainty, and reporting limits rather than for drawing a definitive comparative conclusion.

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 limitations are specific to the available record. First, the study is terminated, and the reason for termination is not reported in the registry record. The status cannot be interpreted as evidence about the trial’s results or the reason it ended.

Second, the primary endpoint is measured through Week 24, while the study description also includes a 48-week post-treatment follow-up period. The posted primary outcome does not describe the full follow-up experience. The registry’s primary response rules also count discontinuations as nonresponses from that point onward and handle Week 16 nonresponders in a specified way. The record does not provide alternative missing-data analyses to assess how sensitive the result is to those rules.

Third, comparative estimates in the interpretation are unadjusted recalculations from posted summaries. The registry record does not report the complete sponsor analysis plan, detailed model specifications, multiplicity methods, or sample-size justification. Secondary outcomes are numerous, and detailed numerical summaries are not available for every listed measure. This limits assessment of consistency across outcomes and of which findings were intended to be confirmatory.

Fourth, the safety results are split across treatment, rescue, and follow-up periods, with differing group sizes and treatment histories. These summaries are not interchangeable randomized comparisons. Deaths are not reported in the registry record, and the adverse-event term listing does not state its denominator in the available summary.

Finally, the registry’s masking classification is triple, while the official title calls the study double-blind; the record does not explain the distinction. The available information supports description of the design and outcomes but leaves important analytical and operational details unresolved.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Study in Participants With Rheumatoid Arthritis — NCT01202773.
• Registry record last updated 2018-05-14; status last verified 2018-05.
• Results first posted 2018-05-14.
• 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

NCT01202773 is a Phase 3, multicenter, randomized, parallel-group, placebo-controlled study of LY2127399 in participants with moderate to severe rheumatoid arthritis and an inadequate response to one or more TNF-α inhibitors. The registry reports 456 participants enrolled and lists a 24-week treatment period followed by a 48-week post-treatment follow-up. The study status is terminated; the reason is not reported in the registry record.

The primary endpoint was ACR20 response from baseline through Week 24. Posted response percentages were 17.6% for 120 mg LY2127399, 24.3% for 90 mg LY2127399, and 20.0% for placebo. Unadjusted recalculated comparisons have confidence intervals that include zero. The record lists multiple secondary outcomes and period-specific safety summaries, but does not report a complete statistical analysis plan, multiplicity approach, or original sample-size rationale.