Safety and Efficacy of Intravenous Natalizumab in Acute Ischemic Stroke: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy of Intravenous Natalizumab in Acute Ischemic Stroke (NCT02730455), a randomized phase 2 study. The primary outcome is Percentage of Participants With Composite Global Measure of Functional Disability Excellent Outcome at Day 90.

2. Clinical Question

The study also aimed to explore dose and exposure response and clinical effects across independence, daily activities, neurologic function, quality of life, cognition, safety, and tolerability. The registry lists additional outcomes, but the primary question centers on the Day 90 composite. The reported group percentages describe the analyzed participants, while the sponsor's logistic-regression results are a separate, adjusted analysis. Because those analyses do not align straightforwardly with the unadjusted comparisons, they should not be treated as interchangeable summaries.

3. Trial Design

This was a completed phase 2 study of intravenous natalizumab in acute ischemic stroke. The registry reports 277 participants enrolled. The design was randomized, quadruple-masked, and parallel-group, with placebo as the comparator. The listed groups were placebo, natalizumab 300 mg IV, and natalizumab 600 mg IV. Each group was assigned a single baseline infusion, administered within one of two windows relative to last known normal: within 9 hours or between 9-24 hours.

The primary endpoint was assessed at Day 90 and defined as an excellent outcome on a composite global measure of functional disability. Participants had to meet both the specified modified Rankin Scale threshold and the specified Barthel Index threshold. The primary analysis population was the Modified Intent-to-Treat population, defined in the registry as randomized participants who received the entire infusion; participants accidentally enrolled because conditions mimicked stroke symptoms were excluded. The record gives analyzed group counts for the primary endpoint, but the reason for differences between enrollment and analyzed counts is not reported in the registry record.

The study's stated primary objective was to assess clinical effects on functional independence and activities of daily living. Secondary objectives included exploring dose and exposure response and other clinical, cognitive, quality-of-life, and safety outcomes. The registry lists eight secondary outcomes, while its posted results include outcome information for some, but not all, of those measures. The original sample-size justification, planned allocation ratio, detailed protocol for the treatment windows, and statistical analysis plan are not reported in the registry record. The dates recorded for study start and primary completion are 2016-07-18 and 2017-11-20, respectively.

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 277 participants with acute ischemic stroke were randomized into the study (94 participants in the placebo group, 91 participants in the natalizumab 300 milligram (mg) group, and 92 participants in the natalizumab 600 mg group).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Modified Intent-to-Treat (MITT) population: all randomized participants who had received entire infusion of study treatment. Participants who were accidentally enrolled based on conditions that mimicked stroke symptom at presentation were excluded from MITT population. Number analyzed is the number of participants with data available for analysis.

5. Baseline Characteristics

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

Population: Intent-to-treat (ITT) population included all participants who were randomized.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With Composite Global Measure of Functional Disability Excellent Outcome at Day 90, measured over Day 90.

The primary endpoint was the percentage of participants with an excellent outcome on a composite global measure of functional disability at Day 90. To qualify, a participant needed a modified Rankin Scale score of 0 or 1 and a Barthel Index score >=95. Both conditions were required, so the measure represents a combined threshold rather than success on either scale alone.

The modified Rankin Scale describes independence after stroke across grades from 0 to 6, with 0 corresponding to no symptoms and 6 to death. The Barthel Index sums scores across 10 activities of daily living and mobility, including feeding, transfers, grooming, toileting, bathing, walking, stairs, dressing, and continence. Its total ranges from 0 to 100, with higher scores indicating greater independence. Combining the scales captures both a broad assessment of disability and performance in everyday activities. A stringent composite can be useful when the clinical question concerns a favorable functional state, but it also means that a participant must meet both component thresholds to count as having an excellent outcome.

The primary endpoint is binary for each participant: the composite threshold is either met or not met. The reported percentages therefore summarize the proportion meeting the complete definition in each group. The endpoint does not describe every aspect of recovery, and it does not by itself explain which component of the composite accounts for a group difference. The registry does not report component-level primary results in the record.

7. Results Posted in the Registry

• Natalizumab 300 mg IV: 42.0% (37/88) versus 54.4% (49/90) on control; risk difference -12.4 percentage points (95% CI -27.0 to 2.2), risk ratio 0.77 (0.57–1.05); unadjusted two-proportion z-test p = 0.098.
• Natalizumab 600 mg IV: 40.4% (36/89) versus 54.4% (49/90) on control; risk difference -14.0 percentage points (95% CI -28.5 to 0.5), risk ratio 0.74 (0.54–1.02); unadjusted two-proportion z-test p = 0.061.

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.

Percentage of Participants With Excellent Outcome in mRS Score at Day 90

• 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:
  • Natalizumab 300 mg IV: 28.9% (24/83) versus 40.7% (35/86) on control; risk difference -11.8 percentage points (95% CI -26.0 to 2.5), risk ratio 0.71 (0.47–1.08); unadjusted two-proportion z-test p = 0.108.
  • Natalizumab 600 mg IV: 25.6% (21/82) versus 40.7% (35/86) on control; risk difference -15.1 percentage points (95% CI -29.1 to -1.1), risk ratio 0.63 (0.40–0.99); unadjusted two-proportion z-test p = 0.038.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Natalizumab 300 mg IV and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Natalizumab 600 mg IV than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percentage of Participants With Excellent Outcome in BI Score at Day 90

• 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:
  • Natalizumab 300 mg IV: 54.3% (44/81) versus 67.4% (58/86) on control; risk difference -13.1 percentage points (95% CI -27.8 to 1.6), risk ratio 0.81 (0.63–1.03); unadjusted two-proportion z-test p = 0.082.
  • Natalizumab 600 mg IV: 54.3% (44/81) versus 67.4% (58/86) on control; risk difference -13.1 percentage points (95% CI -27.8 to 1.6), risk ratio 0.81 (0.63–1.03); unadjusted two-proportion z-test p = 0.082.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Natalizumab 300 mg IV 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 Natalizumab 600 mg IV and the control. The interval gives the range of effects that remain compatible with the data.

Stroke Impact Scale-16 (SIS-16) Score Using a Repeated Measures Mixed Effects Model at Day 90

• 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:
  • Natalizumab 300 mg IV: mean 66.96 versus 76.21 on control; difference -9.25 (95% CI -19.44 to 0.94), standardized effect d = -0.28; z-test p = 0.075.
  • Natalizumab 600 mg IV: mean 68.10 versus 76.21 on control; difference -8.11 (95% CI -17.74 to 1.52), standardized effect d = -0.26; z-test p = 0.099.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Natalizumab 300 mg IV 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 Natalizumab 600 mg IV and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover several dimensions of stroke recovery and study experience: excellent outcomes on the modified Rankin Scale and Barthel Index, Stroke Impact Scale-16 scores, cognitive assessment, neurologic function, adverse events, serious adverse events, and dose response. This breadth is consistent with the stated aim to explore effects beyond the primary composite. It also means that secondary outcomes address different questions and use different measurement scales; they should not be collapsed into a single overall finding.

Where results are posted, the secondary estimates do not form a clear, consistent pattern of evidence that resolves the primary discrepancy. The registry includes both binary outcomes and a repeated-measures score, with sponsor analyses for some comparisons. The results for the remaining listed outcomes are not reported in the registry record. These endpoints can provide context about specific functional, cognitive, neurologic, and safety measures, but they are not substitutes for the prespecified primary endpoint.

Unadjusted secondary p-values and estimates generally carry less confirmatory weight than a primary analysis because multiple outcomes and comparisons create more opportunities for chance findings. Whether the trial used a multiplicity adjustment, a hierarchy of testing, or another strategy to control interpretation is not reported in the registry record. Accordingly, a nominal secondary result should be viewed as exploratory unless the analysis plan establishes otherwise. Secondary outcomes are most informative when read alongside their measurement definitions, analysis populations, interval estimates, and the primary result, rather than as isolated evidence.

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

• Percentage of Participants With Excellent Outcome in mRS Score at Day 90
• Percentage of Participants With Excellent Outcome in BI Score at Day 90
• Stroke Impact Scale-16 (SIS-16) Score Using a Repeated Measures Mixed Effects Model at Day 90
• Montreal Cognitive Assessment (MoCA) Score at Day 90
• Change From Baseline in National Institute of Health Stroke Scale (NIHSS) Score at Day 90
• Number of Participants Experiencing Adverse Events (AE)
• Number of Participants Experiencing Serious Adverse Events (SAE)
• Percentage of Participants With Dose Response at Day 90

9. Statistical Interpretation

10. Statistical Methods Explained

What is being compared? The primary outcome is binary: each participant either meets the composite excellent-outcome definition at Day 90 or does not. A percentage in an arm expresses how common that outcome was among the participants analyzed in that arm. The unadjusted comparisons summarize the difference between the observed proportions in each natalizumab group and placebo. These simple comparisons do not account for covariates in the way a regression model can.

What does a risk difference mean? A risk difference is the difference between two event proportions, expressed on the probability scale. A value below zero means the event proportion in the treatment group is lower than the corresponding proportion in the comparison group under the stated subtraction order. It is an absolute contrast, not a relative ratio. In this trial, the recalculated risk differences are -0.124 for natalizumab 300 mg IV and -0.14 for natalizumab 600 mg IV. Their confidence intervals extend across zero, so the unadjusted estimates do not rule out no difference at the interval's stated level.

What does an odds ratio mean? An odds ratio compares the odds of an event between groups. Odds and probabilities are related but are not the same quantity. The registry reports logistic-regression odds ratios for the primary comparisons, along with confidence intervals and p-values. Regression can incorporate model terms, but the covariates and coding details for these analyses are not reported in the registry record. Since the registry flags a discrepancy between sponsor results and recalculations, the sponsor's analysis takes precedence. The odds ratios should not be translated into a claim about which group had a better outcome without resolving the outcome coding and model details.

How should confidence intervals be read? In general, a confidence interval conveys the range of estimates compatible with the data and the method used, under repeated use of that method. It is not a guarantee that the true value lies inside a particular interval, nor does it measure the probability that a specific hypothesis is true. For a risk difference, an interval containing zero is compatible with no difference. For an odds ratio, an interval containing 1 is compatible with equal odds. The posted primary recalculation intervals include zero, while the sponsor's 600 mg odds-ratio interval excludes 1. Those results are based on different analyses and cannot be substituted for one another.

What does a p-value add? In general, a p-value describes how unusual data at least as incompatible with a specified null model would be if that model and its assumptions were correct. It does not measure the size or clinical importance of an effect, and it is not the probability that the null hypothesis is true. A p-value also does not resolve conflicts between analyses, establish that a model is correctly specified, or account for multiple outcomes unless the analysis addresses multiplicity. The sponsor reports p = 0.086 for the 300 mg comparison and p = 0.031 for the 600 mg comparison. These values belong to the sponsor's logistic-regression analyses.

What is the role of the recalculation? The recalculated values are simple, unadjusted comparisons derived from posted summary statistics. They are useful for checking the scale and direction of the observed percentages and for understanding uncertainty around a straightforward group contrast. They are not a replacement for the sponsor's analysis, particularly when a discrepancy is flagged. The reported sponsor odds ratios and their model-based intervals should remain distinct from the recalculated risk differences and their intervals.

What about continuous secondary outcomes? For the Stroke Impact Scale-16, the registry reports arm means and standard deviations, alongside sponsor estimates from a repeated-measures mixed-effects model. In general, a mean summarizes the center of a numeric score and a standard deviation describes spread among participants. A mixed-effects repeated-measures model can use repeated observations while representing within-participant correlation and estimating adjusted group contrasts. The registry reports adjusted mean differences and intervals, but the full model specification is not reported in the registry record. These model-based estimates therefore should be quoted as reported rather than reconstructed from the arm means.

What do planning illustrations represent? The planning figures use observed effects as inputs and illustrate the participant counts that would correspond to specified power levels under stated assumptions. They are not observed power and do not indicate the probability that this completed study detected an effect. The original trial design assumptions and original sample-size justification are not reported in the registry record. The illustrative assumptions are described separately in the sample-size reading, and results for each endpoint should be understood as conditional on those assumptions.

Why can over-dispersion matter? In general, over-dispersion occurs when variation is greater than a statistical model expects. If ignored, standard errors can be too small and intervals too narrow. The primary endpoint is binary and the posted analyses include logistic regression, but the registry does not report whether over-dispersion was assessed or how any such issue was handled. It would therefore be inappropriate to infer that a particular correction was used.

11. Confidence Intervals and Precision

The primary unadjusted risk-difference estimate for natalizumab 300 mg IV is -0.124, with a confidence interval from -0.270 to 0.022. For natalizumab 600 mg IV, the estimate is -0.14, with a confidence interval from -0.285 to 0.005. Both intervals include zero. This means the unadjusted data are compatible with no difference as well as with values on either side of zero. The interval limits also show that the estimates are not precise enough to settle the direction of the unadjusted contrast.

The sponsor's logistic-regression results have a different scale. The 300 mg comparison has an odds ratio of 0.64 and a 95% confidence interval from 0.38 to 1.07. The 600 mg comparison has an odds ratio of 0.57 and a 95% confidence interval from 0.34 to 0.95. The first interval includes 1; the second does not. These intervals describe model-based odds-ratio estimates and should not be read as intervals for the unadjusted risk differences.

In general, greater interval width indicates less precision, while a narrower interval indicates a more tightly estimated quantity under the analysis assumptions. Precision does not establish that an estimate is unbiased or clinically important. The registry record does not report the full adjustment specification, details needed to reconcile the sponsor and recalculated results, or the original design assumptions. The discrepancy therefore affects interpretation as well as precision: the separate estimates answer related, but not demonstrably identical, statistical questions.

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 figures are participants per arm for the binary-outcome comparisons, not total events. The illustrations use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. They use the observed effect as the starting point and show planning for 100%, 75%, and 50% of that observed effect. The trial's own original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column describes the number of participants per arm that would, under those assumptions, give an 80% chance of detecting the specified effect if that effect were the true one. The 90% power column gives the corresponding number for a 90% chance. These are planning illustrations, not observed power. For example, planning for 100% of the observed primary effect gives 254 participants per arm for 80% power and 340 participants per arm for 90% power in the 300 mg comparison. For the 600 mg comparison, the corresponding figures are 199 and 266 participants per arm.

The illustrations also show that smaller target effects require larger planned group sizes: the 75% and 50% effect assumptions produce larger participant counts than the 100% assumption. This is a general feature of study planning, since subtle differences are harder to distinguish from random variation than larger differences. The figures depend on the observed effect, which can itself be uncertain, and should not be treated as proof that the trial was or was not adequately sized. They do not reconstruct the study's actual design process.

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 = 0.544; treatment proportion p₁ = 0.420; average p̄ = 0.482.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7067 = 1.3850.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.7012 = 0.5902.
4. n = (1.3850 + 0.5902)² / (-0.1240)² = 253.8, rounded up to 254 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 <- 0.544444444444444                          # control proportion
p1 <- 0.420454545454545                          # 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(2730455)
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 = 254 (80% power), 340 (90% power)
 75% of observed effect: n per arm = 453 (80% power), 606 (90% power)
 50% of observed effect: n per arm = 1,020 (80% power), 1,364 (90% power)
power.prop.test: 253.8 
Simulated power at n = 254 per arm: 0.791 (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 sponsor logistic-regression results for the primary comparisons and separately flags a discrepancy with unadjusted recalculations. The sponsor reports an odds ratio of 0.64 for the 300 mg comparison, with a 95% confidence interval from 0.38 to 1.07 and p = 0.086. For the 600 mg comparison, the reported odds ratio is 0.57, with a 95% confidence interval from 0.34 to 0.95 and p = 0.031. The registry classifies both analyses as superiority analyses.

The recalculations use the posted group summaries to estimate unadjusted risk differences. They are not adjusted logistic-regression results, and the odds ratios are not the same measure as risk differences. The registry record does not report enough detail about outcome coding or model adjustment to explain fully why the analyses diverge. In light of the flagged discrepancy, the sponsor's own analysis takes precedence. The recalculated values remain useful as a descriptive check, but should not replace the sponsor's reported estimates or be blended with them into a single conclusion.

When presenting the findings, retain the type of estimate, its interval, and the analysis source together. Avoid interpreting the direction of the sponsor odds ratios from the arm percentages alone when the relevant coding details are not reported in the registry record.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary endpoint and multiple secondary outcomes, including separate comparisons of each natalizumab dose with placebo. Testing several outcomes and contrasts increases the chance that at least one result will appear noteworthy by chance if all are interpreted as independent confirmatory tests. Statistical approaches can address this issue through adjusted significance thresholds, hierarchical testing, or other prespecified procedures. The multiplicity strategy for this study is not reported in the registry record, so the posted p-values should not be assumed to have been adjusted.

The primary population is described as Modified Intent-to-Treat: randomized participants who received the entire infusion, with participants accidentally enrolled because of conditions mimicking stroke symptoms excluded. The registry gives arm-specific analysis counts for the primary endpoint. It does not report the details of missing Day 90 outcomes, the method used to handle missing data, or whether any imputation or sensitivity analyses were performed. These omissions matter because estimates based on participants with recorded outcomes can differ from estimates under other assumptions about unavailable outcomes.

The record does not report a complete statistical analysis plan, the covariates included in the sponsor's logistic-regression models, the outcome coding used in those models, or the handling of the flagged discrepancy. It also does not report the study's original sample-size justification or planned power. Those details are needed to assess whether the posted analyses follow the intended inferential strategy and to reproduce or reconcile model-based results.

The reported results should therefore be read with attention to the named analysis population and the distinction between sponsor analyses and simple recalculations. The registry does not establish that every randomized participant contributed to every endpoint, and it does not provide enough detail to characterize the impact of missingness or multiplicity on the reported findings.

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: cerebrovascular accident (3.4%), pulmonary embolism (3.4%), cerebral haemorrhage (3.3%), ischaemic stroke (3.3%), cardiac arrest (2.2%).

The registry reports serious adverse event counts for 19 of 91 participants in the placebo group, 23 of 90 in the natalizumab 300 mg IV group, and 29 of 89 in the natalizumab 600 mg IV group. It also reports counts for participants with other adverse events: 67 of 91, 69 of 90, and 65 of 89, respectively. Deaths are reported as 5, 6, and 4 in those same groups. These are group-specific counts, not adjusted comparisons.

The record lists serious adverse event terms including cerebrovascular accident and pulmonary embolism, each reported as 3.4%; cerebral haemorrhage and ischaemic stroke, each 3.3%; and cardiac arrest and cerebral artery occlusion, each 2.2%. The registry does not provide enough detail in this record to interpret these terms as comparative risks or to determine their timing, attribution, or overlap with other events.

Adverse-event counts should be read as descriptive safety information. Differences in counts alone do not establish that treatment caused or prevented an event, and the group denominators differ. The record's summary does not report a formal comparative safety analysis or details on event ascertainment. These figures therefore describe what was posted, rather than serving as a standalone causal assessment.

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

The study compares placebo with two single-dose intravenous natalizumab groups in acute ischemic stroke and reports a Day 90 composite outcome of functional independence. The posted arm percentages are 54.1% for placebo, 41.5% for natalizumab 300 mg IV, and 39.9% for natalizumab 600 mg IV. These figures describe the participants included in the primary analysis; they are not, by themselves, adjusted treatment-effect estimates.

The registry's sponsor analyses and unadjusted recalculations differ in form and do not align straightforwardly. Because the record flags a discrepancy, the sponsor's own logistic-regression analysis takes precedence. Its 600 mg comparison has a 95% confidence interval that excludes 1 and p = 0.031, while the 300 mg interval includes 1 and p = 0.086. The outcome coding and model details needed to reconcile those results with the arm percentages are not reported in the registry record. Therefore, the reported statistics do not support a simple interpretation that treats every estimate as measuring the same contrast in the same way.

The findings do not establish a general conclusion beyond this trial, and they should not be read as medical advice. A phase 2 result, an observed group percentage, or a p-value alone cannot establish the size, reliability, or clinical importance of an effect. Interpretation also depends on the analysis population, missing-data handling, multiplicity, model specification, and the relation between the sponsor analysis and descriptive results. Several of these details are not reported in the registry record.

The record supports a careful account of the measured outcomes and their uncertainty, not a definitive statement about treatment effectiveness. Secondary outcomes add context but do not remove the need to understand the primary-analysis discrepancy and the unreported statistical details.

18. Why This Trial Matters Statistically

This trial illustrates why statistical interpretation depends on more than comparing percentages. The primary outcome is a composite requiring both a modified Rankin Scale threshold and a Barthel Index threshold. The group percentages offer an accessible description, while risk differences and logistic-regression odds ratios express comparisons on different scales and may reflect different modeling choices. Those quantities must be kept distinct.

The record also illustrates how analysis populations affect interpretation. The primary population excludes randomized participants who did not receive the entire infusion and participants enrolled after conditions mimicking stroke symptoms were identified. Thus, the primary estimates describe a defined subset rather than automatically representing every person enrolled. The registry provides analyzed counts but does not report complete details about missing outcomes or sensitivity analyses.

A further statistical issue is the flagged difference between the sponsor's own analysis and simple recalculations. The sponsor's results take precedence, but the record does not provide enough detail to reconcile them fully with the raw arm summaries. A reader should preserve that distinction rather than selecting whichever result appears simpler or more persuasive.

Finally, the study includes several secondary outcomes and dose comparisons. Without a reported multiplicity strategy, isolated secondary findings have limited confirmatory force. The sample-size illustrations also show how planning assumptions depend on effect size, but they are not the trial's original design calculations. Together, these issues demonstrate the value of transparent analysis plans, population definitions, model specifications, and uncertainty reporting when interpreting clinical-trial 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 trial-specific limitations affect interpretation of the posted results. The primary analysis population is defined by receipt of the entire infusion and excludes participants accidentally enrolled because conditions mimicked stroke symptoms. The record reports the population definition and arm analysis counts, but it does not describe the outcomes or characteristics of excluded participants or explain the differences between enrollment and endpoint counts.

The sponsor's primary logistic-regression results do not align straightforwardly with the displayed arm percentages and unadjusted recalculations, and the registry flags this discrepancy. The model's covariates, outcome coding, and other details needed to reproduce the analysis are not reported in the registry record. This leaves a limit on independent interpretation of the sponsor estimates, even though those estimates take precedence over simple recalculations.

The registry does not report the original sample-size justification, the trial's planned power, the complete statistical analysis plan, multiplicity procedures, or missing-data methods. It also does not provide detailed results for every listed secondary outcome. These omissions make it difficult to judge the planned inferential hierarchy, the robustness of findings to missing outcomes, and the confirmatory status of secondary analyses.

The primary outcome is a composite, so the overall percentage does not identify which component drove the result. The record does not include component-level primary results. Safety information is descriptive in the posted record, without a reported formal comparative analysis or sufficient detail to assess attribution and timing. These limitations do not invalidate the reported figures, but they restrict the conclusions that can be drawn from them.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Safety and Efficacy of Intravenous Natalizumab in Acute Ischemic Stroke — NCT02730455.
• Registry record last updated 2019-01-08; status last verified 2018-12.
• Results first posted 2019-01-08.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

NCT02730455 is a completed phase 2 randomized, quadruple-masked, parallel-group, placebo-controlled study of intravenous natalizumab for acute ischemic stroke. The registry reports 277 participants enrolled and lists placebo, natalizumab 300 mg IV, and natalizumab 600 mg IV groups. Treatment was a single baseline infusion in one of two windows relative to last known normal.

The primary endpoint was the percentage with an excellent Day 90 composite outcome requiring modified Rankin Scale 0 or 1 and Barthel Index >=95. The registry reports arm percentages and sponsor logistic-regression analyses, while also flagging a discrepancy with unadjusted recalculations. The sponsor's own analysis takes precedence. Secondary and safety outcomes broaden the scope of the record, but several analysis details—including multiplicity handling, missing-data methods, model specification, and original sample-size justification—are not reported in the registry record.