Study of Efficacy, Safety and Tolerability of ACZ885 (Canakinumab) in Pediatric and Young Adult Patients With Sickle Cell Anemia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Efficacy, Safety and Tolerability of ACZ885 (Canakinumab) in Pediatric and Young Adult Patients With Sickle Cell Anemia (NCT02961218), a randomized phase 2 study. The primary outcome is Change From Baseline of 4- Week Average Daily Pain Measured by Visual Analog Score (VAS) Over the Period of Week 8 to 12.

2. Clinical Question

The registry describes a randomized, quadruple-masked, parallel-group study. Its primary endpoint focuses on pain intensity averaged across days, rather than a single pain rating. That repeated daily measurement is intended to represent the participant’s average pain experience over the assessment period and reduce the influence of an unusually high or low rating on one day.

The clinical question is therefore about the contrast in change between groups over the specified period. It is not simply whether pain scores changed within either group: both groups can change over time, and the relevant comparative estimate is the difference between their changes. The registry reports group-level changes and a sponsor analysis of the net mean difference. It does not report a treatment effect that establishes a clinical benefit.

3. Trial Design

This was a phase 2, randomized, placebo-controlled, parallel-design study sponsored by Novartis Pharmaceuticals. The registry lists quadruple masking. It records an enrollment of 49 and a completed status. The study start is recorded as 2017-04-05, with primary completion on 2019-06-27.

The two principal groups were ACZ885 and placebo. ACZ885 is identified in the record as canakinumab, administered subcutaneously. The intervention descriptions specify monthly dosing and weight-based dosing for patients or subjects weighing 40 kg or less. The placebo description also specifies monthly dosing. Further details about allocation procedures, concealment, adherence, or how masking was maintained are not reported in the registry record.

The primary endpoint was change from baseline in the four-week average daily pain visual analog score over Week 8 to 12. The primary description says daily pain intensity was recorded once each evening on an 11-point scale from 0 to 10, with higher ratings corresponding to more intense pain. Baseline was derived from the screening-period average; the stated screening period could last up to 28 days and included at least one week of daily e-diary recording. The primary analysis population was the pharmacodynamics analysis set: subjects with available pharmacodynamics data who had received any study drug.

The registry also lists secondary outcomes covering pain ratings over repeated four-week intervals and changes in inflammatory, blood-cell, hemoglobin, reticulocyte, and bilirubin measures. It records 15 outcome measures and 6 analyses. The detailed protocol, original sample-size justification, and the full analysis plan are not reported in the registry record. The posted design information therefore establishes the broad comparison and endpoint structure, but not all operational details needed to reconstruct trial conduct.

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: Subjects who met the eligibility criteria at screening underwent evaluation of baseline clinical and biomarker assessments prior to first dose administration.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Pharmacodynamics (PD) analysis set: included all subjects with available PD data, who received any study drug.

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 Change From Baseline of 4- Week Average Daily Pain Measured by Visual Analog Score (VAS) Over the Period of Week 8 to 12, measured over Baseline (upto 28 days prior to start of treatment), Week 8 to 12.

The primary endpoint measures change in average daily pain intensity, using a visual analog score, from baseline to the Week 8 to 12 period. Participants recorded pain once each evening on an 11-point numerical rating scale ranging from 0 to 10. A higher rating meant more intense pain; 0 represented no pain and 10 represented the worst pain. The endpoint is the average across the specified period, not an isolated daily score.

Baseline was based on the average daily pain ratings during screening. The registry says the screening period could last up to 28 days and required daily e-diary recording for at least one week. The later value was averaged over Week 8 to 12, and the analysis considered change from baseline. This construction helps describe the pattern across repeated diary entries rather than relying on a single observation. In general, averaging repeated ratings can make a measure less sensitive to day-to-day fluctuation, although it does not remove all uncertainty or missing-data concerns.

Pain is a patient-reported experience, so a daily rating gives participants a direct way to record its intensity. The measure is relevant to a study in sickle cell anemia because the registry identifies pain as the outcome under assessment. The scale describes intensity; it does not by itself capture every aspect of pain, its causes, or its effect on daily life. The registry’s primary endpoint is specifically an average daily score over the stated weeks, and conclusions should remain tied to that measure and assessment period.

7. Results Posted in the Registry

• ACZ885: mean -0.45 versus -0.37 on control; difference -0.08 (95% CI -0.30 to 0.14), standardized effect d = -0.20; z-test p = 0.477.

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 in the Concentration of Hemoglobin From Baseline to Week 12

• 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 deviations.
• Result:
  • ACZ885: mean -0.97 versus 1.11 on control; difference -2.08 (95% CI -6.25 to 2.09), standardized effect d = -0.32; z-test p = 0.328.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between ACZ885 and the control. The interval gives the range of effects that remain compatible with the data.

Change in the Reticulocyte Count From Baseline to Week 12

• 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 deviations.
• Result:
  • ACZ885: mean -6.58 versus 25.36 on control; difference -31.94 (95% CI -68.49 to 4.62), standardized effect d = -0.57; z-test p = 0.087.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between ACZ885 and the control. The interval gives the range of effects that remain compatible with the data.

Change in the Concentration of Bilirubin From Baseline to Week 12

• 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 deviations.
• Result:
  • ACZ885: mean 5.05 versus -1.95 on control; difference 7.00 (95% CI -3.12 to 17.12), standardized effect d = 0.43; z-test p = 0.175.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between ACZ885 and the control. The interval gives the range of effects that remain compatible with the data.

The listed secondary outcomes extend beyond pain to include inflammatory markers and blood measures. The registry names changes in high-sensitivity C-reactive protein, white blood cell count, neutrophil count, monocyte count, hemoglobin, reticulocyte count, and bilirubin, as well as pain ratings over repeated four-week intervals. The detailed numerical summaries available for three laboratory outcomes show estimates with confidence intervals that cross zero: hemoglobin, reticulocyte count, and bilirubin. Their directions are not uniform, and the registry does not report sponsor analyses for those three comparisons.

These secondary outcomes provide context for the broader set of measurements collected, but they do not replace the primary pain endpoint. Their results do not resolve the uncertainty in the primary comparison, which also has an interval spanning zero. Nor should the various secondary measures be treated as independent confirmations: they concern different outcomes and may be related, and the registry does not report a multiplicity adjustment plan.

When many outcomes or comparisons are examined, the chance of finding an apparently notable result by chance can increase. Unadjusted secondary results therefore generally carry less confirmatory weight than a prespecified primary analysis, particularly when the analysis plan and multiplicity handling are not reported. The registry records 15 outcome measures and 6 analyses, but does not report which secondary comparisons were prespecified for formal testing, how missing secondary data were handled, or whether results were adjusted for multiple comparisons. The secondary evidence is best read as exploratory context rather than as a separate basis for a definitive conclusion.

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

• Change From Baseline of Average Daily Pain VAS Over 4 Weeks Intervals up to Week 24
• Change in the Concentration of High Sensitivity C-Reactive Protein (hsCRP) From Baseline to Week 12
• Change in the Concentration of White Blood Cell (WBC) Count From Baseline to Week 12
• Change in the Concentration of Absolute Count of Neutrophils From Baseline to Week 12
• Change in the Concentration of Absolute Count of Blood Monocytes From Baseline to Week 12
• Change in the Concentration of Hemoglobin From Baseline to Week 12
• Change in the Reticulocyte Count From Baseline to Week 12
• Change in the Concentration of Bilirubin From Baseline to Week 12

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a numerical score, and the registry reports group means and standard deviations. A mean summarizes the central value of a set of observations. A standard deviation describes how dispersed individual observations are around that mean. It is not the same as the uncertainty around the mean or the uncertainty around a treatment contrast.

What does a change-from-baseline comparison estimate?

Each group’s change describes how its average score at follow-up differs from its baseline average. A comparison of those changes asks whether the amount of change differs between groups. This distinction matters because improvement or worsening within one group does not by itself show that the intervention caused a difference: changes can also occur in the comparison group. In this study, the posted group changes were -0.45 for ACZ885 and -0.37 for placebo. The unadjusted difference in those changes was -0.08.

The direction of that contrast depends on the group order. The stated recalculation expresses ACZ885 relative to placebo, so a negative value means the ACZ885 group had a more negative average change. Because higher pain scores indicate more intense pain, a negative change represents a decline in average score from baseline. The comparison is still an estimate, not a definitive statement about the experience of every participant.

Why use a repeated-measures model?

Participants contributed repeated daily pain ratings, and the primary outcome was averaged over a later assessment period. A repeated-measures model is a general approach for analyzing observations made over time while accounting for their structure as repeated measurements from the same participant. Such observations are not necessarily independent: readings from one person may be related to one another. A model can represent that dependence and use the repeated observations according to its specified assumptions.

The registry states that the primary change was analyzed using a Bayesian model for repeated measures. It does not report the model’s complete specification, prior distributions, covariates, handling of missing diary entries, or other details needed to reproduce the analysis. The model-based estimate and interval should therefore be read as the sponsor analysis reported in the registry, not as a fully reproducible analysis description.

How does the recalculation differ?

The unadjusted recalculation uses the posted arm-level means and standard deviations to describe a simple contrast. It does not reproduce the repeated-measures model, incorporate its structure, or adjust for other variables. The recalculated mean difference is -0.08 and its unadjusted p-value is 0.477. The reported model-based net mean difference is -0.07, with a 90% confidence interval from -1.03 to 0.85 and p = 0.55. These results have similar point estimates, but the model-based result is the registry’s reported analysis and is the more relevant reference for understanding the sponsor’s stated primary analysis.

A p-value summarizes how unusual the observed data, or data more extreme under a specified test, would be if the test’s null model were correct. It does not give the probability that the null hypothesis is true, and it does not measure clinical importance. The reported and recalculated p-values are not identical because they come from different analyses. The record does not report enough detail to reconstruct every modeling choice or test calculation.

What does a confidence interval contribute?

A confidence interval describes uncertainty around an estimated effect under the method used to construct it. In general, a wider interval indicates less precision, while a narrower interval indicates greater precision. The reported 90% interval runs from -1.03 to 0.85 and includes zero, so it allows both directions of the group contrast as well as values close to no difference. It should not be interpreted as a range containing every plausible individual response or as a guarantee about the true effect.

The registry also reports standard deviations for each arm. These describe variability in observed group scores, whereas the interval for the treatment contrast reflects uncertainty in the estimated difference. The two quantities answer different questions. The small analysis group sizes and the spread of pain scores are relevant context for why estimates may remain uncertain, but no specific causal explanation for the observed uncertainty is reported in the registry record.

Why do repeated daily data and missingness matter?

Repeated diary records can provide a more representative summary than one rating, but the analysis can still depend on which entries are available and how missing entries are handled. The registry describes the baseline diary period and identifies the pharmacodynamics analysis set, but does not report the amount or pattern of missing data or the model’s missing-data assumptions. It is therefore not possible to determine from the registry record whether missingness differed between groups or how sensitive the result was to alternative assumptions.

More generally, an analysis population is part of the definition of an estimate. The primary population included subjects with available pharmacodynamics data who received any study drug. That description is not the same as a complete account of every randomized participant’s inclusion in every analysis. The registry record does not report all analysis-population details or explain exclusions from the primary endpoint analysis.

How should the evidence be framed?

The randomized, placebo-controlled design is intended to support a comparative assessment, while the outcome, analysis population, and model define the particular estimate being reported. The primary result is not a standalone claim that can be separated from its uncertainty and methods. The point estimate is near zero, the reported interval spans both directions, and the model-based p-value does not indicate a clear statistical distinction. Taken together, these features describe limited precision for the primary comparison rather than proof of equality or proof of a treatment effect.

11. Confidence Intervals and Precision

The registry’s primary model-based estimate is a net mean difference of -0.07, with a 90% confidence interval from -1.03 to 0.85. The interval extends from a possible decrease in average pain score favoring ACZ885 to a possible decrease favoring placebo, under the stated group ordering. It includes zero, so the data do not isolate a direction of difference with confidence. The interval is broad relative to the point estimate, indicating that the primary effect is not estimated precisely.

A confidence interval communicates the range of estimates compatible with the data and the statistical method at the stated confidence level. It is not a statement that all values inside are equally likely, nor does it mean that there is a 90% probability the true value lies inside this particular interval. The exact interpretation depends on the method and its assumptions. The record identifies a Bayesian repeated-measures model, but the interval is labelled as a 90% confidence interval; further details about its construction are not reported in the registry record.

The arm-level means were -0.45 for ACZ885 and -0.37 for placebo, while the unadjusted recalculated difference was -0.08. The reported model-based estimate is close to that recalculation. The group standard deviations were 0.384 and 0.402, respectively. Standard deviations describe the variation among observed scores, not the precision of the estimated difference. The model-based interval is the direct summary of uncertainty for the reported primary contrast.

Precision is relevant to interpretation independently of whether a p-value crosses a conventional threshold. The reported p-value is 0.55, and the interval includes both negative and positive differences. The registry does not report a clinically important difference threshold for this score, so the interval cannot be classified against such a threshold from the record alone. The result leaves uncertainty about the size and direction of the comparative change.

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 figures are planning illustrations based on the observed effect, not the trial’s original design assumptions. The trial’s own sample-size justification and original design assumptions are not reported in the registry record. For these illustrations, the stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning scenario uses the stated fraction of the observed effect as the effect to detect.

For the primary endpoint, the table gives participants per arm. Planning for 100% of the observed effect gives 379 participants per arm for 80% power and 507 participants per arm for 90% power. Planning for 75% of the observed effect gives 674 participants per arm for 80% power and 902 participants per arm for 90% power. Planning for 50% of the observed effect gives 1,515 participants per arm for 80% power and 2,028 participants per arm for 90% power.

In plain language, 80% power describes a design’s chance of detecting an effect of the specified size under the planning assumptions if that effect is present. The 90% power column describes the corresponding design with a higher target chance of detection. A higher power target generally requires more participants. The scenarios also illustrate that smaller effects need more participants: reducing the effect to a smaller fraction of the observed estimate increases the planning requirement.

These figures do not describe the power actually achieved by this trial and are not observed power. They are conditional planning estimates based on the observed effect, which itself is uncertain. The figures should not be read as a recommendation for a future study without considering its scientific question, outcome variability, chosen effect of interest, analysis plan, and other design assumptions.

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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.080; pooled standard deviation σ = 0.393.
2. n = 2 × (1.96 + 0.842)² × 0.393² / -0.080² = 378.7, rounded up to 379 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
delta <- -0.08                      # observed difference in means
sigma <- 0.392911554120902                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(2961218)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 379 (80% power), 507 (90% power)
 75% of observed effect: n per arm = 674 (80% power), 902 (90% power)
 50% of observed effect: n per arm = 1,515 (80% power), 2,028 (90% power)
power.t.test: 379.6 
Simulated power at n = 379 per arm: 0.794 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-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 primary analysis as a Bayesian model for repeated measures. It gives a net mean difference of -0.07, a 90% confidence interval from -1.03 to 0.85, and p = 0.55. This is the sponsor’s reported analysis and reflects the stated model-based approach to repeated pain measurements.

The unadjusted recalculation from the posted group summaries gives a mean difference of -0.08 and p = 0.477. It is a simple comparison of arm-level summary statistics, not a reproduction of the repeated-measures model. The two point estimates are close, and both indicate a small numerical difference with uncertainty about the direction. Their p-values differ because the analyses are not the same calculation.

The registry does not flag a discrepancy between its reported result and the recalculation. The sponsor analysis remains the primary reference for the trial’s stated analysis because it is the reported model-based analysis; the recalculation is a supplementary, unadjusted check. The record does not provide enough modeling detail to reproduce the sponsor’s result independently or to identify how every assumption affected the estimate.

15. Multiplicity, Missing Data and Analysis Populations

The registry records 15 outcome measures and 6 analyses. The primary outcome is clearly identified as change from baseline in average daily pain score over Week 8 to 12, and the record also lists several secondary pain and laboratory measures. Assessing multiple outcomes can create multiplicity: when many hypotheses are evaluated, some may appear unusual by chance even if there are no corresponding underlying effects. A prespecified approach to multiplicity can help control how confirmatory claims are made.

The multiplicity strategy is not reported in the registry record. It is not reported whether the secondary analyses were adjusted, whether a testing hierarchy was used, or whether any results were intended to be exploratory. Consequently, secondary p-values should not be interpreted as if they formed a single confirmatory testing program with a known error-control procedure. The primary result should also be understood in light of the unreported full analysis plan.

Missing-data details are limited. The primary population is described as the pharmacodynamics analysis set, including subjects with available pharmacodynamics data who received any study drug. The registry does not report the number or pattern of missing pain diary entries, the number excluded from the primary analysis, the model’s assumptions about missing observations, or sensitivity analyses addressing departures from those assumptions. These details matter because repeated-measures estimates can depend on the availability and distribution of observations over time.

The registry record does not report the full analysis-population definitions for every outcome or explain how participants with incomplete follow-up were handled. It also does not report baseline group characteristics or a full account of protocol deviations. These omissions limit assessment of comparability and robustness from the registry record alone. They do not establish that the analyses were inappropriate; they mark information that is unavailable for an independent statistical appraisal.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: sickle cell anaemia with crisis (50.0%), pneumonia (9.1%), acute chest syndrome (8.3%), asthenia (5.0%), pyrexia (5.0%).

The safety record lists serious adverse events and other adverse events by recorded group, along with deaths. In the ACZ885 group, serious adverse events affected 11 of 25 participants, and other adverse events affected 19 of 25; deaths were 0. In the placebo group, serious adverse events affected 15 of 24 participants, and other adverse events affected 20 of 24; deaths were 0.

The registry also lists groups labelled ACZ885 / ACZ885 and Placebo / ACZ885. For ACZ885 / ACZ885, serious adverse events affected 11 of 22 participants, other adverse events affected 17 of 22, and deaths were 0. For Placebo / ACZ885, serious adverse events affected 11 of 20 participants, other adverse events affected 18 of 20, and deaths were 0. The registry record does not explain the relationship between these group labels and the principal treatment groups, so they should be retained as recorded rather than combined or reclassified.

Seriousness is a regulatory classification and is not synonymous with severity. The listed serious-event terms include sickle cell anaemia with crisis, pneumonia, acute chest syndrome, asthenia, pyrexia, and cholelithiasis. The registry gives percentages for these terms, but their event-specific denominators and further details are not reported in the registry record. These counts describe reported events in the recorded groups; they do not establish comparative safety or attribute an event to a study drug.

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

The primary result is a comparison of average daily pain-score change from baseline over Week 8 to 12. The observed arm means were -0.45 for ACZ885 and -0.37 for placebo, and the unadjusted mean difference was -0.08. The sponsor’s repeated-measures analysis estimated a net difference of -0.07, with a 90% confidence interval from -1.03 to 0.85 and p = 0.55. Together, these statistics show a small numerical difference with substantial uncertainty about its size and direction.

The result does not establish that canakinumab changed pain outcomes relative to placebo. It also does not establish that the treatments were equivalent or that any difference is absent. A finding of equivalence requires a prespecified equivalence margin and an analysis designed for that purpose; neither is reported in the registry record. The interval includes zero and values in both directions, so it supports caution rather than a categorical conclusion.

The interpretation applies to the measured score, the stated assessment period, and the analysis population described in the record. It should not be generalized automatically to other ages, patient groups, dosing approaches, outcomes, or periods. The registry record does not report a threshold defining a clinically important difference on the pain scale, so statistical estimates cannot be translated into a threshold-based judgment of clinical importance from the record alone.

Secondary laboratory outcomes add information about other measured domains, but their unadjusted results do not settle the primary question. Safety counts provide a description of reported events, not a causal comparison. Overall, the posted results leave uncertainty about the comparative effect and do not by themselves support a broad claim about efficacy, safety, or tolerability beyond the specific observations recorded.

18. Why This Trial Matters Statistically

This study is statistically informative because it places a patient-reported, repeatedly measured pain outcome at the center of a randomized, placebo-controlled comparison. Daily ratings can capture variation across time, while an average over a defined period can provide a more stable summary than an isolated rating. At the same time, the design makes the analysis dependent on how repeated observations, within-person correlation, and incomplete diary records are handled.

The reported model-based analysis and simple recalculation give similar point estimates, but the reported confidence interval is broad and crosses zero. That combination illustrates an important distinction: a numerical estimate can point in one direction without providing precise evidence that the true comparative effect lies in that direction. The p-value alone cannot convey this uncertainty or determine whether an effect would be important to patients.

The registry also lists several secondary measures, making multiplicity and prespecification relevant to interpretation. Without a reported multiplicity strategy, secondary findings should be treated cautiously and should not be elevated to confirmatory evidence. Likewise, the absence of the full statistical plan, missing-data details, and original sample-size rationale limits the ability to assess how well the posted analysis answers the original trial question.

The planning illustrations show how the number of participants required can increase when the effect of interest is smaller or the desired power is higher. They are not a reconstruction of the original design, and they are not evidence about the trial’s actual power. The main statistical lesson is that design assumptions, analysis choices, uncertainty, and the hierarchy of outcomes all shape what a trial result can support.

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.

The registry record does not report the original sample-size justification, the complete statistical analysis plan, or the design assumptions used when the trial was planned. The planning figures are based on the observed effect and stated illustrative assumptions, rather than on the trial’s original design. They cannot establish whether the study was designed to detect a clinically meaningful difference.

The primary outcome depends on repeated daily diary reports. The record describes the measurement scale and screening-period approach, but does not report diary completion rates, missing-data patterns, or sensitivity analyses. It also does not provide the full repeated-measures model specification, including its covariates and assumptions. These gaps make independent reproduction and evaluation of robustness difficult.

The registry identifies the primary analysis population but does not report all analysis-population details or participant flow through each analysis. It does not report baseline characteristics in the registry record for this account, so the degree of observed balance between groups cannot be evaluated. Details about allocation concealment, adherence, treatment discontinuation, and protocol deviations are not reported in the registry record.

The primary interval includes zero and extends in both directions, limiting precision about the comparative effect. A clinically important difference threshold is not reported, so the clinical meaning of particular score differences cannot be determined from the registry record alone. Several secondary outcomes are listed, but no multiplicity strategy is reported. The safety record gives group counts and selected serious-event terms, while event-specific denominators and fuller descriptions are not reported in the registry record.

These limitations concern the information available for interpreting the posted results. They do not demonstrate that the trial procedures or analyses were flawed. They indicate where the registry record does not support a more detailed assessment.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study of Efficacy, Safety and Tolerability of ACZ885 (Canakinumab) in Pediatric and Young Adult Patients With Sickle Cell Anemia — NCT02961218.
• Registry record last updated 2026-01-13; status last verified 2025-12.
• Results first posted 2020-12-17.
• 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

The registry describes NCT02961218, a completed phase 2 randomized, quadruple-masked, parallel study sponsored by Novartis Pharmaceuticals. It enrolled 49 pediatric and young adult patients with sickle cell anemia and compared monthly ACZ885 (canakinumab) with placebo. The primary endpoint was change from baseline in average daily pain visual analog score over Week 8 to 12.

The posted primary group means were -0.45 for ACZ885 and -0.37 for placebo. The sponsor reported a Bayesian repeated-measures net mean difference of -0.07, with a 90% confidence interval from -1.03 to 0.85 and p = 0.55. The interval spans zero and both directions, leaving the comparative effect uncertain. Secondary measures and adverse-event counts provide additional descriptive context, but the registry does not report key details of the full analysis plan, multiplicity handling, or missing-data approach.