Efficacy at 24 Weeks and Safety, Tolerability and Long Term Efficacy of Secukinumab (AIN457) in Patients With Active Rheumatoid Arthritis (RA) and an Inadequate Response to Anti-Tumor Necrosis Factor α (Anti-TNFα) Agents (CAIN457F2309 and CAIN457F2309E1): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy at 24 Weeks and Safety, Tolerability and Long Term Efficacy of Secukinumab (AIN457) in Patients With Active Rheumatoid Arthritis (RA) and an Inadequate Response to Anti-Tumor Necrosis Factor α (Anti-TNFα) Agents (CAIN457F2309 and CAIN457F2309E1) (NCT01350804), a randomized phase 3 study. The primary outcome is Percentage of Participants Achieving an American College of Rheumatology Response 20 (ACR20)..

2. Clinical Question

3. Trial Design

The registry record describes NCT01350804 as a completed Phase 3 randomized, quadruple-masked, parallel-group study sponsored by Novartis Pharmaceuticals. The condition was rheumatoid arthritis. Enrollment is listed as 551. The recorded start is 2011-09 and primary completion is 2015-02.

The core study compared two secukinumab regimens with placebo and abatacept. The secukinumab arms began with intravenous AIN457 at 10mg/kg at baseline, Weeks 2 and 4, then used either 75 mg or 150 mg subcutaneously from Week 8, with injections every 4 weeks. Abatacept was administered intravenously at a dose from 500 to 1000 mg based on weight.

The placebo arrangement included treatment switching: participants received matching placebo until Week 16 or Week 24, depending on responder status. Nonresponders switched to active treatment at Week 16 and responders at Week 24. Participants who did not respond to abatacept at Week 16 were to be reassigned to secukinumab at Week 24 after an 8 week washout period. These treatment changes matter when interpreting outcomes beyond the stated primary time point, because later group exposure was not necessarily unchanged.

The record distinguishes the core study from a four-year extension. Its brief summary states that the core study was completed and that the extension was prematurely terminated after the core study’s week 24 primary endpoint analysis. The reason given is that abatacept had numerically higher efficacy than secukinumab. Details of the trial’s original sample-size justification, planned interim analyses, and full statistical analysis plan are 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: Abatacept responders at Week16 continued on abatacept and non-responders at Week 16 were re-randomized to begin either AIN457 75mg or AIN457 150mg at Week 24. Of the 551 core study participants, 254 participants entered the extension study.

Core Study

Reasons for not completing this period:

Extension Study, Weeks 52 - 260

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set: the full analysis set was comprised of all randomized participants (excluding mis-randomized participants) who were assigned to study treatment.

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants Achieving an American College of Rheumatology Response 20 (ACR20)., measured over week 24.

The primary endpoint was the percentage of participants achieving an American College of Rheumatology response 20 at week 24. ACR20 was a composite response: participants needed at least 20% improvement in tender 68-joint count and swollen 66-joint count, plus improvement in at least 3 of 5 additional measures. Those measures covered patient assessments of pain and disease activity, physician assessment of disease activity, disability measured with the HAQ-DI, and an acute-phase reactant, hsCRP or ESR.

General explanation: a composite endpoint combines several aspects of a condition into one response definition. This can represent a broader clinical change than a single measure, but the result depends on meeting the specified combination of criteria. ACR20 is used to classify whether an individual meets a defined improvement threshold; it is not a measure of how large the improvement was beyond that threshold.

The registry reports the week 24 response using non-responder imputation. In general, this approach treats participants without an observed response under the specified analysis rules as nonresponders. The record’s endpoint description identifies the method, but further details of how particular missing or post-switch observations were handled are not reported in the registry record.

7. Results Posted in the Registry

• AIN457 10mg/kg - 75 mg: 28.3% (39/138) versus 18.1% (25/138) on control; risk difference 10.1 percentage points (95% CI 0.3 to 20.0), risk ratio 1.56 (1.00–2.43); unadjusted two-proportion z-test p = 0.046.
• AIN457 10mg/kg - 150 mg: 30.7% (42/137) versus 18.1% (25/138) on control; risk difference 12.5 percentage points (95% CI 2.5 to 22.6), risk ratio 1.69 (1.10–2.61); unadjusted two-proportion z-test p = 0.015.
• Abatacept: 42.8% (59/138) versus 18.1% (25/138) on control; risk difference 24.6 percentage points (95% CI 14.2 to 35.1), risk ratio 2.36 (1.58–3.53); unadjusted two-proportion z-test p = <0.001.

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.

Change From Baseline in Disease Activity Score Utilizing CRP (DAS28-CRP)

• 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:
  • AIN457 10mg/kg - 75 mg: mean -1.47 versus -1.02 on control; difference -0.45 (95% CI -0.84 to -0.06), standardized effect d = -0.38; z-test p = 0.024.
  • AIN457 10mg/kg - 150 mg: mean -1.47 versus -1.02 on control; difference -0.45 (95% CI -0.85 to -0.05), standardized effect d = -0.38; z-test p = 0.026.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for AIN457 10mg/kg - 75 mg 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 mean is statistically lower for AIN457 10mg/kg - 150 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Stanford Health Assessment Questionnaire Disability Index (HAQ-DI)

• 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:
  • AIN457 10mg/kg - 75 mg: mean -0.30 versus -0.26 on control; difference -0.04 (95% CI -0.20 to 0.12), standardized effect d = -0.08; z-test p = 0.623.
  • AIN457 10mg/kg - 150 mg: mean -0.39 versus -0.26 on control; difference -0.13 (95% CI -0.29 to 0.03), standardized effect d = -0.26; z-test p = 0.116.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AIN457 10mg/kg - 75 mg 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 AIN457 10mg/kg - 150 mg and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants Achieving ACR50

• 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:
  • AIN457 10mg/kg - 75 mg: 11.6% (16/138) versus 9.4% (13/138) on control; risk difference 2.2 percentage points (95% CI -5.1 to 9.4), risk ratio 1.23 (0.62–2.46); unadjusted two-proportion z-test p = 0.556.
  • AIN457 10mg/kg - 150 mg: 16.8% (23/137) versus 9.4% (13/138) on control; risk difference 7.4 percentage points (95% CI -0.6 to 15.3), risk ratio 1.78 (0.94–3.37); unadjusted two-proportion z-test p = 0.070.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AIN457 10mg/kg - 75 mg 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 AIN457 10mg/kg - 150 mg and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary outcomes cover disease activity, disability and additional ACR response thresholds. The overall pattern is mixed rather than uniform: some continuous disease-activity comparisons favor the secukinumab groups relative to placebo in the unadjusted recalculations, while the disability comparisons are less decisive. The ACR50 response percentages are higher in the secukinumab groups than in placebo, but the recalculated intervals include the null value for both comparisons. Abatacept’s posted values are also part of the descriptive context, but the recalculated comparisons listed for these secondary endpoints are against placebo.

These outcomes add context to the primary ACR20 result because they assess related but different aspects of rheumatoid arthritis: disease activity scores, functional disability and a more demanding response threshold. They should not be treated as interchangeable confirmations. A participant can meet one endpoint definition without meeting another, and continuous scale changes answer a different question from the percentage crossing a response threshold. No single secondary pattern overturns the primary result or supplies a full account of longer-term outcomes.

Unadjusted secondary results carry less weight because the registry lists multiple outcomes and comparisons, and does not report a multiplicity adjustment for them. General explanation: when many hypotheses are tested, the chance of observing at least one apparently notable result by chance can increase. Without a reported testing hierarchy or correction, individual secondary p-values should be read as exploratory evidence rather than definitive confirmation. The registry record also does not report a complete statistical analysis plan that would clarify which secondary outcomes were prespecified as confirmatory, how missing data were handled for each, or how treatment switching affected later assessments.

The results are therefore best understood as a set of complementary signals with different scales, populations and analysis methods, not as a single combined verdict. Their interpretation depends on the endpoint definition and posted analysis for each outcome. The registry record does not report enough detail to establish a formal hierarchy among the secondary endpoints.

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

• Change From Baseline in Disease Activity Score Utilizing CRP (DAS28-CRP)
• Change From Baseline in Stanford Health Assessment Questionnaire Disability Index (HAQ-DI)
• Percentage of Participants Achieving ACR50
• Percentage of Participants Achieving ACR20, ACR 50 and ACR 70 - Using Non-responder Imputation
• Percentage of Participants Achieving ACR20, ACR 50 and ACR 70 - Observed Data
• Change From Baseline in HAQ-DI - Using Mixed Model Repeated Measures (MMRM)
• Change From Baseline in HAQ-DI - Observed Data
• Change From Baseline in Disease Activity Score Utilizing CRP (DAS28-CRP) - Using MMRM

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary at the participant level: each participant either met the ACR20 definition or did not under the registry’s analysis rules. The registry reports percentages for each group and identifies non-responder imputation. General explanation: binary outcomes are commonly summarized as proportions, then compared using measures such as a risk difference, risk ratio or odds ratio. Each measure answers a related but distinct question, so the scale should be kept clear when reading an estimate.

What does a risk difference describe?

General explanation: a risk difference subtracts the event proportion in one group from the event proportion in another. A value above zero means the event proportion is higher in the group placed first in the comparison; a value below zero means it is lower. The recalculated primary comparisons use placebo as the reference and report risk differences of 0.101 for the 75 mg regimen and 0.125 for the 150 mg regimen. Their intervals are 0.003 to 0.200 and 0.025 to 0.226, respectively. These estimates are unadjusted recalculations from the posted group summaries.

How do odds ratios differ?

General explanation: an odds ratio compares the odds of an event between groups, whereas a risk difference compares event proportions. Odds and proportions are not interchangeable, so the numerical value of an odds ratio should not be read as a percentage-point change. The sponsor’s reported analyses used logistic regression and presented odds ratios. The posted values are 1.79 for the 75 mg group versus placebo and 2.02 for the 150 mg group versus placebo. The record lists 95% confidence intervals and p-values for those analyses. The sponsor’s model-based odds ratios are not the same estimand as the simple recalculated risk differences.

General explanation: logistic regression models the log odds of a binary outcome. It can incorporate model terms and covariates, although the specific covariates used in the sponsor’s analyses are not reported in the registry record. The reported analysis method is Regression, Logistic; details beyond that label should not be inferred. The sponsor’s results should be distinguished from simple recalculations because the methods and measures differ.

What does non-responder imputation mean?

The primary endpoint description states that week 24 ACR20 results used non-responder imputation. General explanation: imputation assigns an analysis value to observations without an observed outcome, according to a stated rule. In a non-responder-imputation analysis, missing or otherwise specified observations are treated as not meeting response. This can make the analysis conservative in some settings, but its consequences depend on why observations are missing and exactly how the rule is applied. The record does not specify all missing-data details, including how treatment switching or particular kinds of unavailable observations were handled.

The primary population is described as the full analysis set: all randomized participants, excluding mis-randomized participants, who were assigned to study treatment. This description identifies the intended analysis population but does not give all participant-level disposition details. The group sizes posted for the primary endpoint are 138 for the 75 mg regimen, 137 for the 150 mg regimen, 138 for placebo and 138 for abatacept. These are endpoint group counts, not a substitute for a full account of missing observations or analysis exclusions.

General explanation: a confidence interval is a range of values compatible with the data and the method used, under repeated-sampling assumptions. A wider interval indicates less precision; an interval that includes a null value does not establish that the groups are equal. For a risk difference, the null value is zero. For an odds ratio, the null value is one. These intervals do not capture every possible source of bias, such as outcome measurement, post-randomization treatment changes or unreported analytical decisions.

The registry also posts continuous outcomes and other response endpoints. Their methods include least-squares means with standard errors for continuous outcomes and percentages for binary outcomes. General explanation: a standard error describes the statistical uncertainty in an estimated mean, while a standard deviation describes variation among individual observations. The registry’s continuous outcome entries report standard errors, not standard deviations. Some recalculated comparisons use an approximate pooled standard deviation derived from posted summary statistics; those calculations are not the sponsor’s adjusted analysis.

Where planning illustrations are posted, they use a normal approximation and equal allocation between two groups, with a two-sided significance level of 0.05. They scale the observed effect to the stated fraction. General explanation: power is the probability, under a specified model and assumed effect, of detecting that effect using the chosen test. Planning calculations rely on assumptions and should not be confused with what the trial actually demonstrated. Details of the original trial design assumptions and any multiplicity procedure are not reported in the registry record.

11. Confidence Intervals and Precision

The primary percentage estimates are point estimates: they summarize the observed ACR20 response proportions under non-responder imputation. They do not show how precisely each group’s underlying response probability has been estimated. The registry record does not post confidence intervals around the individual arm percentages.

For the recalculated risk difference comparing the 75 mg regimen with placebo, the estimate is 0.101 and the interval is 0.003 to 0.200. For the 150 mg regimen versus placebo, the estimate is 0.125 and the interval is 0.025 to 0.226. Both intervals are above zero, but their widths show that the size of the difference remains uncertain. These are unadjusted recalculations, not the sponsor’s logistic-regression estimates.

The sponsor reported odds ratios of 1.79, with a 95% confidence interval from 1.0 to 3.2, for the 75 mg regimen versus placebo, and 2.02, with a 95% confidence interval from 1.1 to 3.5, for the 150 mg regimen versus placebo. The corresponding reported p-values were 0.046 and 0.015. These estimates use odds rather than risk differences; their intervals therefore should not be compared as if they were on the same scale.

General explanation: confidence intervals describe statistical uncertainty under the analysis method, not certainty about an individual participant’s outcome. They also do not account automatically for every design or analysis concern. For example, unreported multiplicity handling, missing-data assumptions and treatment switching can affect how confidently a result is interpreted. The registry record does not report enough information to assess those issues fully.

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 planning illustrations report participants per arm for 80% power and participants per arm for 90% power. They are not total event counts: the registry’s stated convention is that participant-based outcomes are shown per arm, while time-to-event displays would give total events. These illustrations use 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.

The 80% power column means the planned design would have an 80% probability of detecting the assumed effect under the calculation’s assumptions. The 90% column uses a higher target probability of detection. These are planning targets, not guarantees about a future study. For each comparison, the rows scale the observed effect to 100%, 75% or 50%. As the assumed effect becomes smaller, more participants per arm are needed to maintain the same target power. For example, the primary AIN457 10mg/kg - 75 mg versus placebo illustration gives 271 participants per arm for 80% power when using 100% of the observed effect, and 999 when using 50% of it.

These figures are not observed power. They are illustrative planning calculations based on the effect estimates in the posted results, not the trial’s original sample-size justification. The original design assumptions are not reported in the registry record. They also do not establish that the trial was adequately powered for every endpoint or comparison; the calculations depend on the stated effect and assumptions, and the record does not report whether multiplicity was accounted for in these illustrations.

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 = 0.181; treatment proportion p₁ = 0.283; average p̄ = 0.232.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5968 = 1.1698.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5925 = 0.4987.
4. n = (1.1698 + 0.4987)² / (0.1014)² = 270.5, rounded up to 271 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 <- 0.181159420289855                          # control proportion
p1 <- 0.282608695652174                          # 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(1350804)
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 = 271 (80% power), 362 (90% power)
 75% of observed effect: n per arm = 463 (80% power), 620 (90% power)
 50% of observed effect: n per arm = 999 (80% power), 1,337 (90% power)
power.prop.test: 270.5 
Simulated power at n = 271 per arm: 0.805 (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

The sponsor’s posted primary analyses used logistic regression and reported odds ratios for the two secukinumab regimens versus placebo. For the 75 mg regimen, the reported odds ratio is 1.79, with a 95% confidence interval from 1.0 to 3.2 and p = 0.046. For the 150 mg regimen, the reported odds ratio is 2.02, with a 95% confidence interval from 1.1 to 3.5 and p = 0.015.

The recalculated comparisons use risk differences, based on the posted group summaries, and are unadjusted. They are useful as a separate descriptive comparison but do not reproduce the sponsor’s regression model or its odds-ratio estimand. The registry record does not report the full model specification or covariates, so those details cannot be used to reconcile any numerical differences between the approaches. The record marks no discrepancy between posted and recalculated results; where a difference in method or measure exists, the sponsor’s analysis remains the trial’s reported analysis, while the recalculation is supplementary context.

The sponsor’s results compare each secukinumab group with placebo. They do not provide a reported logistic-regression comparison against abatacept in the primary analyses listed. The record’s brief summary separately notes numerically higher efficacy for abatacept than secukinumab at the core study’s week 24 analysis. That summary is descriptive and does not supply a formal direct-comparison estimate.

15. Multiplicity, Missing Data and Analysis Populations

The primary ACR20 outcome is identified clearly, and its week 24 analysis is stated to use non-responder imputation. The record also lists several secondary outcomes and more than one analysis approach for some outcomes. However, the registry record does not report a multiplicity procedure, such as a testing hierarchy or adjustment across endpoints and comparisons. Without that information, the unadjusted p-values should not be treated as if they were corrected for the number of statistical questions considered.

General explanation: multiplicity matters because repeated testing creates more opportunities for a result to appear statistically notable by chance. A prespecified strategy can control how evidence is assessed across multiple comparisons. The available record does not establish whether a strategy was used outside the details posted. It is therefore not possible to label secondary findings as confirmatory based on the registry entries alone.

The record does specify the primary analysis population as the full analysis set, consisting of randomized participants assigned to study treatment and excluding mis-randomized participants. It also states non-responder imputation for the primary response. Further details about missing outcomes, reasons for missingness, sensitivity analyses, and handling of treatment discontinuation or switching are not reported in the registry record.

Those omissions matter because placebo participants could switch to active treatment at Week 16 or Week 24 based on responder status, and abatacept nonresponders had a reassignment pathway after a washout period. The primary endpoint is at week 24, but the record does not fully explain how each post-randomization change was represented in every analysis. Nor does it report whether analyses followed an intention-to-treat principle beyond the stated full analysis set definition.

General explanation: missing-data assumptions can influence an estimate when the probability of an unobserved outcome is related to participant status or treatment response. Non-responder imputation applies a defined rule, but it does not by itself reveal the amount or pattern of missingness. The posted record should thus be read as documenting the stated primary method, not as resolving every question about missing data, multiplicity or analysis populations.

16. 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: osteoarthritis (1.8%), foot deformity (0.9%), intervertebral disc protrusion (0.9%), rheumatoid arthritis (0.9%), cardiac failure (0.9%).

The safety record lists serious adverse events and other adverse events by safety group. For the groups labeled Any AIN457 75 mg, Any AIN457 150 mg, placebo and abatacept, the numbers at risk for serious events are 218, 215, 139 and 137, respectively. The corresponding numbers listed as affected by serious events are 30, 28, 7 and 9. For other adverse events, the listed affected counts are 144, 146, 51 and 78, with the same respective at-risk group counts.

These are counts, not comparative risk estimates, and the treatment group labels reflect the registry’s safety grouping. The listed serious-event terms include osteoarthritis at 1.8%, foot deformity at 0.9%, intervertebral disc protrusion at 0.9%, rheumatoid arthritis at 0.9%, cardiac failure at 0.9% and back pain at 0.9%. The registry record does not report death counts in these entries. The counts alone do not establish that a treatment caused an event or provide a complete assessment of comparative safety.

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

The primary result describes the proportion of participants meeting a defined ACR20 response at week 24 under non-responder imputation. The posted percentages were higher in both secukinumab groups than in placebo, and the simple unadjusted risk-difference intervals for those comparisons are above zero. The sponsor’s logistic-regression analyses also report odds ratios above one for both secukinumab comparisons with placebo. These findings describe the posted comparisons; they do not establish what would happen for an individual participant.

The record also shows a higher posted ACR20 percentage for abatacept than for either secukinumab regimen. The brief summary explicitly characterizes abatacept efficacy as numerically higher at the core study’s week 24 primary endpoint analysis. The registry entries do not provide a formal direct comparison of abatacept against each secukinumab regimen in the reported primary analyses, so the descriptive ordering should not be mistaken for a statistically tested head-to-head conclusion.

General explanation: a statistically notable comparison does not establish clinical importance by itself. The size of an effect, its precision, the endpoint’s meaning, the analysis assumptions and the balance of harms all matter. ACR20 captures crossing a specified response threshold; it does not measure the full magnitude of change, all dimensions of disease or long-term durability.

The record does not establish long-term efficacy or safety from the terminated extension. It also does not report enough detail to resolve the consequences of multiplicity, missing data or post-randomization treatment changes. The serious-event and other-event counts are descriptive safety information, not causal estimates. No medical recommendation follows from these registry results alone.

18. Why This Trial Matters Statistically

This trial is statistically informative because it includes placebo and active-comparator groups, reports two secukinumab regimens, and identifies a week 24 primary response endpoint. The posted results permit readers to distinguish descriptive response percentages from comparative estimates and to compare the sponsor’s logistic-regression odds ratios with unadjusted risk-difference recalculations.

The active comparator adds context beyond a placebo comparison. The registry’s summary says abatacept had numerically higher efficacy than secukinumab at the primary analysis, and the posted ACR20 percentages show that descriptive ordering. This underlines why a positive comparison against placebo does not automatically establish equivalence or superiority to another active treatment. The listed sponsor primary analyses compare secukinumab with placebo, not directly with abatacept.

The record also illustrates how endpoint definition and missing-data methods shape interpretation. ACR20 is a threshold-based composite, and non-responder imputation determines how the primary response result is formed. The placebo switching rules and abatacept reassignment pathway make the treatment history relevant, while the registry record does not fully report the analytical handling of those changes.

Finally, the trial highlights the value of separating confirmatory conclusions from secondary patterns and exploratory recalculations. Unadjusted secondary p-values, planning illustrations based on observed effects, and descriptive safety counts each answer different questions. The registry’s limited detail on multiplicity, missingness and the full model specification sets boundaries on what can be inferred from the posted results.

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.

Several limitations are specific to this registry record. The sponsor’s reported primary analyses use logistic regression, but the full model specification and covariates are not reported. The record also does not provide a direct sponsor analysis comparing either secukinumab regimen with abatacept, even though abatacept is an active comparator and its posted ACR20 percentage is higher.

Placebo participants could switch to active treatment according to responder status, and abatacept nonresponders had a later reassignment pathway. These changes complicate interpretation beyond the primary time point. The registry does not describe in full how switching, discontinuation or missing observations were handled in each analysis. The primary endpoint does identify non-responder imputation, but further missing-data details and sensitivity analyses are not reported in the registry record.

The record lists multiple secondary outcomes and comparisons without reporting a multiplicity procedure or a complete hierarchy of confirmatory endpoints. Secondary p-values and unadjusted recalculations should therefore be interpreted cautiously. The sample-size illustrations are derived from observed effects and stated planning assumptions, not from the trial’s original design justification. The original sample-size assumptions are not reported in the registry record.

The extension study was prematurely terminated after the week 24 core analysis, limiting what this record can say about the planned longer-term evaluation. Safety information is presented as counts, and death counts are not reported in the registry record. These registry-level limitations do not show that the analyses were improperly conducted; they define what can be assessed from the recorded information.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy at 24 Weeks and Safety, Tolerability and Long Term Efficacy of Secukinumab (AIN457) in Patients With Active Rheumatoid Arthritis (RA) and an Inadequate Response to Anti-Tumor Necrosis Factor α (Anti-TNFα) Agents (CAIN457F2309 and CAIN457F2309E1) — NCT01350804.
• Registry record last updated 2016-05-23; status last verified 2016-05.
• Results first posted 2016-05-23.
• 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.

23. Record Summary

NCT01350804 was a completed Phase 3 randomized, quadruple-masked, parallel-group study sponsored by Novartis Pharmaceuticals in people with active rheumatoid arthritis and an inadequate response to anti-TNFα agents. The core trial compared two secukinumab regimens with placebo and abatacept. Enrollment is listed as 551.

The primary endpoint was week 24 ACR20 response using non-responder imputation. Posted percentages were 28.3% for the 75 mg secukinumab regimen, 30.7% for the 150 mg regimen, 18.1% for placebo and 42.8% for abatacept. Sponsor analyses reported logistic-regression odds ratios versus placebo for both secukinumab regimens. The registry’s brief summary notes numerically higher efficacy for abatacept than secukinumab at the primary analysis and says the extension was prematurely terminated.

Secondary outcomes covered disease activity, disability and further response thresholds. Safety entries list serious and other adverse-event counts by group. The registry record does not report the original sample-size justification, a multiplicity procedure, full model details or comprehensive handling of missing data and treatment switching.