Effect of Secukinumab on Radiographic Progression in Ankylosing Spondylitis as Compared to GP2017 (Adalimumab Biosimilar): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effect of Secukinumab on Radiographic Progression in Ankylosing Spondylitis as Compared to GP2017 (Adalimumab Biosimilar) (NCT03259074), a randomized phase 3 study. The primary outcome is Percentage of Participants With no Radiographic Progression at Week 104 (Multiple Imputation) (Full Analysis Set).

2. Clinical Question

The trial focused on structural change in the spine rather than only on symptoms. The primary endpoint classified a participant as having no radiographic progression when the change from baseline in mSASSS was <= 0.5. Results were reported for three groups: AIN457 150 mg/Placebo, AIN457 300 mg, and GP2017 40mg.

The arm label AIN457 150 mg/Placebo is the registry's wording. The registry record does not explain in the available information how the matching placebo relates to treatment allocation over the study, so the label should not be expanded into a more detailed treatment sequence. The clinical question is limited to the endpoint and time point registered: absence of radiographic progression between baseline and Week 104.

The registry also includes secondary outcomes involving mSASSS change, new syndesmophytes, MRI edema scores, and clinical response measures. Those outcomes address related but distinct aspects of disease and should not be treated as interchangeable with the primary radiographic classification.

3. Trial Design

This was a randomized, parallel-group Phase 3 study with single masking. The registry reports enrollment of 859 participants and a completed study. The listed start date is 2017-11-30, and the primary completion date is 2021-11-12. The sponsor was Novartis Pharmaceuticals.

The three primary-result groups were AIN457 150 mg/Placebo, AIN457 300 mg, and GP2017 40mg. Their reported analysis-set sizes were 287, 286, and 286, respectively. The intervention descriptions state that AIN457 150 mg and a matching placebo were administered subcutaneously at Baseline and Weeks 1, 2, 3 and 4, followed by dosing every 4 weeks until Week 100. AIN457 300 mg was administered on the same stated schedule, and GP2017 40 mg was administered at Baseline and every 2 weeks until Week 102.

The primary outcome was assessed between baseline and Week 104. It was analyzed using multiple imputation in the full analysis set. The registry lists eight secondary outcomes, including imaging, radiographic, and clinical response measures. The numerical secondary results available in the registry record cover selected outcomes and subsets; detailed numerical results for every named secondary outcome are not reported in the registry record.

The registry record does not report the original sample-size justification, the assumptions used to design the trial, or details of any interim analyses. It also does not provide enough information to reconstruct all aspects of randomization, masking, or treatment handling. The design description should therefore be read as the registered design information, not as a complete protocol account.

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: A total of 859 subjects were randomized to treatment at 171 sites in 30 countries in Europe, North America, South America, and Asia

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With no Radiographic Progression at Week 104 (Multiple Imputation) (Full Analysis Set), measured over Baseline and at Week 104.

The primary endpoint was the percentage of participants with no radiographic progression at Week 104. Radiographic change was assessed with the modified Stoke Ankylosing Spondylitis Spine Score, or mSASSS, which scores vertebral corners in the lumbar and cervical spine. The registry describes scores of 0 for normal, 1 for erosion, sclerosis, or squaring, 2 for a syndesmophyte, and 3 for a bridging syndesmophyte. The total score ranges from 0-72.

No radiographic progression was defined as a change from baseline in mSASSS of <= 0.5. The outcome is therefore a binary classification: each participant is counted as either meeting or not meeting that threshold. The reported percentages describe the proportion classified as having no progression, not the average amount of score change.

General statistical context: a threshold-based endpoint can be easier to interpret as a participant-level outcome than a mean score change, but it compresses information. Participants with different score changes can fall into the same category, and measurements near the threshold can affect classification. The registry's secondary mSASSS change outcome retains a different form of information by reporting score change rather than the percentage below a threshold.

Structural spine change was the focus of the trial's stated purpose. The endpoint is not a direct measure of every clinical feature of ankylosing spondylitis, and it should be interpreted as a radiographic outcome at the specified assessment time.

7. Results Posted in the Registry

• AIN457 300 mg: 66.8% (191/286) versus 66.2% (190/287) on control; risk difference 0.6 percentage points (95% CI -7.1 to 8.3), risk ratio 1.01 (0.90–1.13); unadjusted two-proportion z-test p = 0.883.
• GP2017 40mg: 65.7% (188/286) versus 66.2% (190/287) on control; risk difference -0.5 percentage points (95% CI -8.2 to 7.3), risk ratio 0.99 (0.88–1.12); unadjusted two-proportion z-test p = 0.906.

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 mSASSS at Week 104 (Multiple Imputation) (Full Analysis Set)

• 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 300 mg: mean 0.55 versus 0.54 on control; difference 0.01 (95% CI -0.48 to 0.50), standardized effect d = 0.00; z-test p = 0.968.
  • GP2017 40mg: mean 0.72 versus 0.54 on control; difference 0.18 (95% CI -0.31 to 0.67), standardized effect d = 0.06; z-test p = 0.470.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AIN457 300 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 GP2017 40mg and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Participants Without New Syndesmophytes by mSASSS Between Baseline and Week 104 (Multiple Imputation) (Syndesmophyte Subset)

• 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 300 mg: 53.9% (110/204) versus 56.9% (120/211) on control; risk difference -3.0 percentage points (95% CI -12.5 to 6.6), risk ratio 0.95 (0.80–1.13); unadjusted two-proportion z-test p = 0.545.
  • GP2017 40mg: 53.3% (113/212) versus 56.9% (120/211) on control; risk difference -3.6 percentage points (95% CI -13.0 to 5.9), risk ratio 0.94 (0.79–1.11); unadjusted two-proportion z-test p = 0.460.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AIN457 300 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 GP2017 40mg and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in MRI Berlin Sacroiliac (SI) Joint Edema Score (Observed Data) (MRI Subset)

• 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 300 mg: mean -1.38 versus -1.53 on control; difference 0.15 (95% CI -0.15 to 0.45), standardized effect d = 0.12; z-test p = 0.328.
  • GP2017 40mg: mean -1.71 versus -1.53 on control; difference -0.18 (95% CI -0.48 to 0.11), standardized effect d = -0.14; z-test p = 0.227.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AIN457 300 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 GP2017 40mg and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover several related dimensions: mSASSS score change, the percentage without new syndesmophytes, MRI edema scores, and clinical response or disease-activity classifications. Together they broaden the outcome profile beyond the primary binary radiographic classification. However, the measures differ in scale, population subset, and analysis approach, so they cannot be combined into a single overall effect without a prespecified synthesis.

Across the secondary results with numerical values reported, the estimates do not establish a clear separation between the treatment groups. The mSASSS change comparisons have uncertainty intervals spanning zero, and the reported percentages without new syndesmophytes are also similar across groups. MRI edema score results are based on an MRI subset and represent score change, not radiographic progression. This overall pattern is consistent with the primary comparisons in the limited sense that the available estimates do not show a precise group difference; it does not prove that all outcomes or effects are equivalent.

The registry names additional secondary outcomes, including ASAS20, ASAS40, partial remission, and inactive disease responses. Detailed numerical results for every named secondary outcome are not reported in the registry record. Conclusions about those outcomes should not be inferred from the radiographic endpoint or from the subset of secondary results with numeric estimates.

General statistical context: testing multiple outcomes and contrasts increases the chance of obtaining apparently notable results by chance if no adjustment or hierarchy is used. The registry record does not report a multiplicity adjustment strategy for these secondary comparisons. Accordingly, unadjusted secondary results carry less confirmatory weight than a prespecified primary analysis and are best viewed as supportive or exploratory evidence, with attention to their own analysis populations and missing-data methods.

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

• Change From Baseline in mSASSS at Week 104 (Multiple Imputation) (Full Analysis Set)
• Percentage of Participants Without New Syndesmophytes by mSASSS Between Baseline and Week 104 (Multiple Imputation) (Syndesmophyte Subset)
• Change From Baseline in MRI Berlin Sacroiliac (SI) Joint Edema Score (Observed Data) (MRI Subset)
• Change From Baseline in Berlin Modification of ASspiMRI-a Edema Score (MRI Subset)
• Percentage of Responders for Assessment of SpondyloArthritis International Society 20 (ASAS20)
• Percentage of Responders for Assessment of SpondyloArthritis International Society 40 (ASAS 40)
• Percentage of Responders for Assessment of SpondyloArthritis International Society With a Partial Remission Response (Full Analysis Set)
• Percentage of Participants With Assessment of SpondyloArthritis International Society for Inactive Disease Response (Observed Data) (Full Analysis Set)

9. Statistical Interpretation

10. Statistical Methods Explained

What does the primary percentage represent?

The primary endpoint is a binary participant-level outcome: the participant either met the registered definition of no radiographic progression or did not. The arm percentages estimate the proportion meeting that definition. A risk difference expresses the difference between two such proportions on the proportion scale. A value of zero would correspond to equal estimated proportions; a positive or negative value indicates which group has the larger estimated proportion according to the comparison's ordering.

The recalculated risk differences use the posted arm-level information and are unadjusted. They are a simple comparison, not a reproduction of the sponsor's logistic-regression model. This distinction matters because a model-based marginal difference can depend on the model and its handling of covariates, while the unadjusted calculation is based on the overall arm summaries. The detailed covariates and full model specification are not reported in the registry record.

What does a confidence interval tell the reader?

General statistical explanation: a confidence interval describes the uncertainty around an estimate under the statistical method used. A narrower interval indicates greater precision under that method; a wider interval indicates more uncertainty. It is not a range of individual participant outcomes, and it does not assign a probability that the true effect lies in the particular interval after the data have been observed.

For a difference in proportions, an interval containing zero means the posted data are compatible with no difference under the calculation, as well as with nonzero differences represented by the interval. That is not proof of no effect. Conversely, an interval excluding zero would not by itself show that a difference is clinically important. The registry does not report a threshold for clinical importance for this endpoint.

How should the p-values be read?

General statistical explanation: a p-value measures how unusual data at least as incompatible with a specified null model would be if that model and the analysis assumptions held. It is not the probability that the null hypothesis is true, nor is it a measure of the size or importance of an effect. The unadjusted recalculated p-values are 0.883 for the AIN457 300 mg comparison with AIN457 150 mg/Placebo and 0.906 for the GP2017 comparison with AIN457 150 mg/Placebo. The sponsor's posted p-values come from logistic regression and are reported separately.

The registry identifies the sponsor analyses as logistic regression with a marginal difference parameter. The exact covariate set, model diagnostics, and detailed calculation of the marginal estimates are not reported in the registry record. The recalculation uses the observed binary outcome summaries and does not attempt to infer or reproduce those unavailable details.

Why does the endpoint use a threshold?

A threshold converts score change into a classification. That can make the result readable as the share of participants meeting a defined radiographic criterion. However, classification discards detail about how far a score changed once the threshold is crossed. General statistical context: dichotomizing a continuous or score-based measure can reduce information and can make results sensitive to the precise cutoff. Here, the registered definition is a change in mSASSS of <= 0.5.

The mSASSS change outcome is a separate secondary measure. Its reported arm values are least-squares means with standard errors, rather than percentages. It addresses average score change, which is not the same estimand as the primary proportion meeting the no-progression definition. The MRI edema outcomes and clinical response outcomes are also distinct measures, so agreement or disagreement across them should be interpreted as a pattern across related endpoints, not as repeated measurements of one identical quantity.

What role does multiple imputation have?

The primary endpoint is identified as using multiple imputation. General statistical explanation: multiple imputation addresses missing outcome information by creating multiple completed datasets under a specified imputation model, analyzing each, and combining results to reflect uncertainty due to missingness. The validity of that approach depends on the model and assumptions about why data are missing. The imputation model, number of imputations, variables included, and sensitivity analyses are not reported in the registry record.

Some secondary outcomes are also identified as using multiple imputation, while other outcomes are identified as observed-data analyses. Those analysis labels indicate that not all outcomes necessarily use the same missing-data approach. The registry record does not report enough detail to determine the influence of missingness on each result or to compare approaches fully.

Why can over-dispersion matter in other trial outcomes?

General statistical explanation: over-dispersion occurs when outcome variability exceeds what a statistical model assumes. It can make standard errors too small if ignored, leading to confidence intervals that are too narrow and p-values that are too optimistic. The primary endpoint is a binary proportion, and the listed primary analyses are not described as count models. The registry record does not report an over-dispersion assessment, so none should be inferred for this endpoint.

Overall, statistical interpretation requires keeping the estimand, analysis population, missing-data approach, and model distinct. The primary result concerns a binary radiographic classification in the full analysis set at Week 104; the unadjusted recalculation and sponsor model-based analyses are related but not interchangeable summaries.

11. Confidence Intervals and Precision

The arm-level primary percentages are 66.1%, 66.9%, and 65.6%. The registry does not give confidence intervals alongside these arm percentages in the posted primary outcome summary. Confidence intervals are available for the posted between-group comparisons, both for the sponsor's logistic-regression analyses and for the unadjusted recalculations.

For the unadjusted AIN457 300 mg versus AIN457 150 mg/Placebo comparison, the risk-difference interval runs from -0.071 to 0.083. For GP2017 40mg versus AIN457 150 mg/Placebo, it runs from -0.082 to 0.073. Each interval includes zero and spans differences in either direction. This indicates that the estimated contrasts are not very precise enough to rule out a range of possible differences, including no difference.

The sponsor's reported 95% confidence intervals also include zero: -6.63 to 9.64 for AIN457 150 mg/Placebo versus GP2017 40mg, and -6.61 to 9.95 for AIN457 300 mg versus GP2017 40mg. These intervals correspond to the sponsor's marginal-difference analyses and should be read on that analysis's scale, not substituted for the recalculated risk-difference intervals.

General statistical explanation: interval width reflects both the observed information and the method used to estimate uncertainty. A confidence interval is not a guarantee that all plausible effects are equally likely, and it does not determine clinical importance. Because the registry record does not report a clinically important difference threshold for this endpoint, the precision can be described statistically but not classified against a registered clinical threshold.

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 planning figures are illustrations based on the observed effect, not the trial's original design assumptions. For the binary primary outcome, the figures are participants per arm, rather than total participants across both arms or total events. Any curve accompanying the planning results should be read as the same kind of planning illustration: it describes how required enrollment changes as the assumed effect changes.

The listed 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 a stated fraction of the observed effect. Separate columns for 80% and 90% power indicate the number of participants per arm estimated to give those respective chances of detecting the assumed effect under the planning assumptions, if the effect and model assumptions hold.

Smaller assumed effects require more participants per arm to detect, because a smaller separation is harder to distinguish from sampling variation. The planning figures are therefore sensitive to the observed effect used as their starting point, and they should not be interpreted as a recommended design for a future study without further design assumptions and clinical context. The trial's own original sample-size justification and design assumptions are not reported in the registry record. These figures are not observed power.

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 = 0.662; treatment proportion p₁ = 0.668; average p̄ = 0.665.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6675 = 1.3083.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6675 = 0.5618.
4. n = (1.3083 + 0.5618)² / (0.0058)² = 103,563.2, rounded up to 103,564 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 <- 0.662020905923345                          # control proportion
p1 <- 0.667832167832168                          # 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(3259074)
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 = 103,564 (80% power), 138,642 (90% power)
 75% of observed effect: n per arm = 184,311 (80% power), 246,740 (90% power)
 50% of observed effect: n per arm = 415,144 (80% power), 555,759 (90% power)
power.prop.test: 103563.2 
Simulated power at n = 103,564 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. Analyses Reported by the Sponsor

The registry reports the sponsor's primary comparisons as logistic-regression analyses with marginal differences. For AIN457 150 mg/Placebo versus GP2017 40mg, the reported marginal difference is 1.51, with a 95% confidence interval from -6.63 to 9.64 and p = 0.716. For AIN457 300 mg versus GP2017 40mg, the reported marginal difference is 1.67, with a 95% confidence interval from -6.61 to 9.95 and p = 0.693.

The recalculated comparisons are unadjusted risk differences based on posted group summaries. They use a different calculation from the sponsor's logistic-regression analyses, so their estimates and intervals need not match. The registry record does not report the full model specification or enough detail to reproduce the sponsor's analysis independently.

Both approaches show uncertainty around the between-group contrasts, and the sponsor's reported confidence intervals include zero. The recalculation is useful as a transparent check of the summary data, but it does not replace the sponsor's model-based analysis. Neither a p-value above a conventional threshold nor an interval including zero establishes equivalence; the registry record does not report a prespecified equivalence or non-inferiority margin.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies one primary endpoint: the percentage of participants with no radiographic progression at Week 104. It also lists multiple secondary outcomes and reports more than one comparison. When many outcomes or comparisons are assessed, the chance of at least one seemingly unusual result can increase. General statistical context: a prespecified testing hierarchy or multiplicity adjustment can control that issue, but the registry record does not report whether such a strategy was planned or applied. The registered primary result should therefore remain distinct from the wider set of secondary findings.

The primary endpoint is labeled as analyzed using multiple imputation in the full analysis set. That identifies the broad approach and population, but not its full implementation. The imputation model, assumptions about missing data, number of imputed datasets, auxiliary variables, and sensitivity analyses are not reported in the registry record. The effect of missing outcomes on the primary estimate therefore cannot be assessed from the available detail alone.

Secondary outcomes use different analysis populations and methods. The syndesmophyte outcome is reported in a syndesmophyte subset, and MRI outcomes are reported in an MRI subset. Some outcomes are described as multiple-imputation analyses, while others use observed data. These differences mean that sample sizes and missing-data handling are not uniform across outcomes. The registry record does not report detailed reasons for missing assessments or enough information to assess whether missingness could have differed between groups.

The full analysis set is named for the primary result, but its precise membership rules and how participants with protocol deviations were handled are not reported in the registry record. The record also does not report the original statistical analysis plan, the prespecified primary contrast hierarchy, or interim-analysis details. Those omissions limit assessment of how closely the posted analysis follows a prespecified plan and how the reported secondary findings should be weighted.

16. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: crohn's disease (1.1%), colitis (1.0%), osteoarthritis (1.0%), pancreatitis acute (0.7%), pneumonia (0.7%).

The registry reports serious adverse events among 286 participants at risk in the AIN457 150 mg/Placebo group, with 40 affected; among 285 participants at risk in the AIN457 300 mg group, with 29 affected; and among 285 participants at risk in the GP2017 40 mg group, with 32 affected. These are counts of participants reported as affected, not counts of all serious-event episodes.

For other adverse events, the corresponding at-risk groups were 286, 285, and 285, with 106, 107, and 99 affected, respectively. Deaths reported were 1, 1, and 3, respectively. These figures describe the registry's reported counts and should not be read as evidence that a particular treatment caused an event. The registry record does not provide the event-level timing, exposure-adjusted rates, or causal assessments needed for a more detailed comparison.

The registry also lists serious-event terms including Crohn's disease, colitis, osteoarthritis, acute pancreatitis, pneumonia, and pyelonephritis, with percentages attached to the terms. Those terms provide a limited description of reported events, but they do not replace a full safety analysis with definitions, follow-up time, and treatment attribution.

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

The primary result describes the percentage of participants meeting a specific radiographic definition at Week 104. The posted group percentages are close, and the available between-group intervals include zero. That pattern means the registry results do not provide a precise estimate of a difference between groups. It does not establish that the groups are identical, nor does it establish that any difference is clinically unimportant.

The endpoint measures structural change in the spine using mSASSS. It does not by itself summarize pain, function, quality of life, MRI inflammation, or every other aspect of ankylosing spondylitis. Those issues appear in some of the secondary outcomes, but they have different definitions and analysis populations. A conclusion about the primary radiographic endpoint should not be expanded into a general claim about all outcomes.

Likewise, a non-significant comparison is not a test of equivalence unless a suitable equivalence design and margin are specified. The registry record does not report such a margin. It also does not report the original sample-size justification or enough of the analysis plan to assess every design choice. The results are best read as estimates with uncertainty, alongside the stated endpoint, population, and analysis method.

The trial record is useful for understanding how the radiographic outcome was defined and what estimates were posted for each group. Its findings do not, on their own, provide a complete account of long-term safety, individual participant outcomes, or the balance of benefits and harms. No treatment decision follows from these statistics alone.

18. Why This Trial Matters Statistically

This trial is statistically notable for using a participant-level radiographic classification as its primary endpoint and reporting results at a specified long-term assessment point. The distinction between the threshold-based primary outcome and the secondary mean mSASSS change outcome illustrates how different summaries of the same general domain answer different questions. One reports the proportion meeting a threshold; the other summarizes average score change.

The record also allows readers to distinguish sponsor-reported model-based analyses from simple unadjusted recalculations. That distinction is important because analyses can target related contrasts while using different methods, and their estimates and confidence intervals may differ. A transparent reading keeps the estimand, analysis method, and uncertainty together rather than treating a single p-value as the whole result.

The study includes imaging and clinical secondary outcomes, but these should not be collapsed into one conclusion. Subsets, missing-data approaches, and multiplicity all affect how much weight to place on secondary comparisons. The absence of a reported multiplicity plan and full statistical analysis details places limits on confirmatory interpretation beyond the primary endpoint.

Finally, the planning illustrations based on observed effects demonstrate why effect-size assumptions matter to sample-size calculations. They are distinct from the original trial design and cannot establish what the study was designed to detect. Keeping planning calculations separate from observed results helps prevent a common statistical misreading: treating a post hoc sample-size illustration as evidence of the study's ability to rule out meaningful effects.

19. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The registry record leaves important details of the original design and analysis unavailable. It does not report the original sample-size justification, the trial's design effect assumptions, a full statistical analysis plan, or interim-analysis details. The randomization procedures and exact masking implementation are also not described beyond the registered labels of randomized and single masking.

The primary endpoint depends on a threshold of change in mSASSS of <= 0.5. The record does not report a clinical-importance threshold for between-group differences, nor does it state an equivalence or non-inferiority margin. The estimates and confidence intervals therefore cannot establish equivalence. The full analysis set is named, but its detailed membership rules are not reported.

Multiple imputation is identified for the primary outcome, but the imputation model and its assumptions are not reported. The available record does not provide the missing-data pattern or sensitivity analyses needed to assess how robust the primary finding is to alternative assumptions. Secondary outcomes use subsets and different analysis approaches, which limits direct comparisons across endpoints.

Some secondary outcomes named in the record do not have detailed numerical results in the available information. The registry does not report a multiplicity strategy for the numerous secondary outcomes and contrasts. Safety data are reported as counts, without the event-level details, exposure-adjusted rates, or causal assessments needed for a full comparative safety analysis. These limitations do not invalidate the posted results, but they restrict how specifically they can be interpreted.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Effect of Secukinumab on Radiographic Progression in Ankylosing Spondylitis as Compared to GP2017 (Adalimumab Biosimilar) — NCT03259074.
• Registry record last updated 2023-08-21; status last verified 2023-08.
• Results first posted 2023-03-09.
• 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

NCT03259074 is a completed randomized, single-masked, parallel-group Phase 3 study sponsored by Novartis Pharmaceuticals. It enrolled 859 participants with active ankylosing spondylitis and compared AIN457 regimens, identified in the title as secukinumab, with GP2017, an adalimumab biosimilar.

The primary outcome was the percentage with no radiographic progression at Week 104, based on change in mSASSS from baseline and analyzed with multiple imputation in the full analysis set. Reported arm percentages were 66.1% for AIN457 150 mg/Placebo, 66.9% for AIN457 300 mg, and 65.6% for GP2017 40mg. The posted comparisons are uncertain, with confidence intervals that include zero.

The record lists several secondary radiographic, MRI, and clinical outcomes and reports adverse-event counts. Important details—including the original sample-size justification, full analysis plan, multiplicity strategy, and imputation specifications—are not reported in the registry record.