Study of GSK1278863 to Reduce Ischemic Events in Patients Undergoing Thoracic Aortic Aneurysm Repair: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of GSK1278863 to Reduce Ischemic Events in Patients Undergoing Thoracic Aortic Aneurysm Repair (NCT01920594), a randomized phase 2 study. The primary outcome is Change From Baseline to Peak in Cerebrospinal Fluid (CSF) S100 Beta Within 48 Hours Following Descending Thoracic Aorta/Thoracoabdominal Aortic Aneurysm (DTA/TAAA) Repair.

2. Clinical Question

The registry describes a Phase II study of GSK1278863 in patients undergoing elective descending thoracic aorta or thoracoabdominal aortic aneurysm repair. The clinical question was whether treatment was associated with a different change in cerebrospinal-fluid S100 beta, a biomarker linked in the record to neurologic injury after this surgery. The measure was the change from baseline to the peak value within 48 hours following repair.

The study also recorded adverse events and other outcomes, including cerebrospinal-fluid biomarkers and laboratory measures. The primary result is a comparison of a biomarker measurement, not a direct count of ischemic events or neurologic injuries. Interpretation therefore depends on the biomarker contrast and its uncertainty, while the wider clinical context includes the recorded safety and secondary outcomes.

3. Trial Design

The registry identifies a randomized, double-blind, parallel-group, placebo-controlled Phase II trial sponsored by GlaxoSmithKline. The record describes 57 enrolled participants. The study status is completed. The study dates recorded are 2013-10-31 for start and 2014-10-08 for primary completion.

The planned treatment assignment was 1:1, with participants receiving GSK1278863 or matching placebo. The brief summary describes stratification according to whether repair was surgical or endovascular, and states that endovascular repair was limited to 50% of the planned study population. It describes approximately 160 subjects as the planned enrollment. These are design descriptions, distinct from the recorded enrollment of 57.

The active regimen was a 300 mg loading dose on Day-1, 12 +/- 4 hours before planned surgery, followed by 100 mg once daily for 4 days starting on Day 0. The placebo arm received matching placebo on the corresponding schedule. Treatment was described as starting before planned repair and continuing through postoperative day 3.

The registry calls the study double-blind and also labels it sponsor open. Further details about who remained masked, how allocation was generated or concealed, and how the planned stratification was implemented are not reported in the registry record. The trial's original sample-size justification and its full statistical analysis plan are also not reported in the registry record. The record describes participation as expected to last approximately 4 to 8 weeks from screening to follow-up.

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 55 participants were randomized in the study and were included in All Subjects Population. Participants were stratified according to intervention type (surgical or endovascular repair, with the latter limited to 50% of the total study population).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Pharmacodynamic (PD) population comprised of all participants from whom PD data was available. For the change from Baseline assessment, only those with both evaluable Baseline and post-dose values (so that the change could be calculated) were included in the 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.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline to Peak in Cerebrospinal Fluid (CSF) S100 Beta Within 48 Hours Following Descending Thoracic Aorta/Thoracoabdominal Aortic Aneurysm (DTA/TAAA) Repair, measured over Baseline (Day 0) to 48 hours following DTA/TAAA repair.

The primary endpoint was change from baseline to peak in cerebrospinal-fluid S100 beta within 48 hours following descending thoracic aorta or thoracoabdominal aortic aneurysm repair. The unit was nanograms per liter. Baseline was defined as Day 0, and the change was calculated as the post-baseline value minus the baseline value. The primary summary was the mean change, accompanied by a standard deviation.

The registry describes S100 beta as a cerebrospinal-fluid biomarker that rises significantly in participants with neurologic injury following this surgery. That rationale makes the biomarker relevant to the study's ischemia hypothesis, but the endpoint remains a laboratory measurement rather than a direct clinical event. A difference in the biomarker does not, by itself, establish a difference in neurologic outcomes or in the broader set of ischemic events named in the study question.

Measurements were collected at baseline and within 48 hours after repair to assess peak change. The primary analysis population was the pharmacodynamic population: participants with pharmacodynamic data, and, for this change assessment, evaluable values both at baseline and after dosing. This definition means that participants without both measurements could not contribute a calculated change. The number excluded for that reason and the reasons for any missing measurements are not reported in the registry record.

7. Results Posted in the Registry

• GSK1278863 300 mg Loading + 100 mg QD: mean 2,748 versus 566.78 on control; difference 2,181 (95% CI -268.58 to 4,631), standardized effect d = 0.47; z-test p = 0.081.

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.

Number of Participants With Clinical Chemistry Parameters of PCI

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • GSK1278863 300 mg Loading + 100 mg QD: 70.4% (19/27) versus 64.3% (18/28) on control; risk difference 6.1 percentage points (95% CI -18.6 to 30.8), risk ratio 1.09 (0.76–1.58); unadjusted two-proportion z-test p = 0.631.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between GSK1278863 300 mg Loading + 100 mg QD and the control. The interval gives the range of effects that remain compatible with the data.

Number of Participants With Hematology Parameters of PCI

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • GSK1278863 300 mg Loading + 100 mg QD: 92.6% (25/27) versus 57.1% (16/28) on control; risk difference 35.4 percentage points (95% CI 14.6 to 56.3), risk ratio 1.62 (1.16–2.27); unadjusted two-proportion z-test p = 0.003.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for GSK1278863 300 mg Loading + 100 mg QD than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline to Peak in CSF Biomarker Erythropoietin Within 48 Hours Following DTA/TAAA Repair

• 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:
  • GSK1278863 300 mg Loading + 100 mg QD: mean 1.34 versus 0.61 on control; difference 0.73 (95% CI -51.10 to 52.56), standardized effect d = 0.01; z-test p = 0.978.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between GSK1278863 300 mg Loading + 100 mg QD and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists 23 secondary outcomes, spanning adverse events, vital signs, electrocardiography, laboratory measures, and cerebrospinal-fluid biomarkers. The secondary material does not form a single consistent confirmation of the primary biomarker result: it combines outcomes with different scales and purposes, and the reported comparisons do not all point to the same statistical interpretation. Safety laboratory findings and CSF biomarker measures should not be treated as interchangeable evidence.

Some secondary outcomes are counts of participants meeting criteria, while others measure change in a continuous biomarker. Their estimates therefore answer different questions and have different sources of uncertainty. The sponsor-reported analyses are not listed for the two binary secondary outcomes with detailed comparisons, while the erythropoietin result has a sponsor ANCOVA analysis alongside a very different simple re-calculation. The record does not report model details that would resolve that difference.

Unadjusted secondary results carry less weight because testing multiple outcomes increases the chance of apparently notable findings arising by chance. A multiplicity procedure can address that issue, but none is reported in the registry record. The secondary endpoints may help describe the breadth of measurements collected and identify patterns for further study, but they should not override the primary result or be presented as definitive evidence of benefit or harm. The record does not report a hierarchy or prespecified strategy for interpreting the secondary outcomes.

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

• Number of Participants With Adverse Events (AEs) and Serious Adverse Events (SAEs)
• Number of Participants With Vital Signs of Potential Clinical Importance (PCI)
• Number of Participants With Abnormal Electrocardiography (ECG) Parameters
• Number of Participants With Clinical Chemistry Parameters of PCI
• Number of Participants With Hematology Parameters of PCI
• Change From Baseline in Area Under Curve (AUC) for CSF S100 Beta to 48 Hours
• Change From Baseline in AUC for CSF GFAP to 48 Hours
• Change From Baseline to Peak in CSF Biomarker Erythropoietin Within 48 Hours Following DTA/TAAA Repair

9. Statistical Interpretation

10. Statistical Methods Explained

What does the mean difference describe?

The primary endpoint is continuous: each participant contributes a numerical change in S100 beta, and the group summaries report means and standard deviations. A mean difference compares the average values in the two groups. In the simple unadjusted re-calculation, the comparison is the active-treatment mean minus the placebo mean. A positive estimate means the active-treatment group's mean change was higher; it does not by itself establish why the values differed or whether the difference is clinically important.

The posted means are 2748.0 ng/L for GSK1278863 and 566.8 ng/L for placebo. The re-calculated unadjusted mean difference is 2181.0 ng/L. The corresponding pooled standard deviation is 4593.0, and the standardized difference is 0.475. Standardizing expresses a mean contrast relative to the spread of the measurements, which can aid comparison across scales. It does not remove uncertainty or turn the biomarker into a clinical outcome.

Why can the estimate be uncertain?

A standard deviation describes how spread out individual measurements are around a group mean. The primary endpoint's standard deviations are 6212.0 in the GSK1278863 group and 1897.0 in the placebo group. Such spread matters because the precision of a group comparison depends not only on the distance between means but also on variability and the number of observations contributing to the analysis. The primary summaries list 27 participants in each group, but the eligible analysis population is restricted to participants with both baseline and post-dose values.

The record's re-calculated standard errors are 1196.0 for GSK1278863 and 365.1 for placebo. A standard error describes uncertainty in an estimated mean, rather than variation among individual participants. The broad confidence interval around the mean difference reflects uncertainty in the group contrast. A confidence interval is a range of values reasonably compatible with the data under the statistical model; it is not a guarantee that the true value lies within its endpoints.

What does the p-value tell the reader?

The unadjusted p-value for the re-calculated primary comparison is 0.081. A p-value describes how unusual the observed data, or data more extreme, would be under a specified null model. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. The registry also reports an ANCOVA p-value of 0.082. The study's own prespecified threshold for interpreting the primary result is not reported in the registry record. The planning illustration separately specifies a two-sided significance level of 0.05; that planning assumption should not be mistaken for documentation of the trial's complete analysis plan.

How does ANCOVA differ from the simple comparison?

The registry reports an ANCOVA analysis of mean difference in final values. Analysis of covariance compares outcome values while accounting for selected covariates. The exact covariates and model specification are not reported in the registry record. The simple re-calculation uses posted summary statistics and is unadjusted; it cannot reproduce an adjusted analysis without the participant-level data and model details. Differences between the re-calculated and sponsor-reported estimates are therefore not, on their own, evidence of an error. No discrepancy is flagged in the record.

How should the secondary measures be understood?

Some secondary outcomes are counts of participants meeting a laboratory criterion, while others are continuous biomarker changes. For binary outcomes, a risk difference compares the proportions in the groups; a positive difference indicates a higher proportion in the active-treatment group when the contrast is defined as active treatment minus placebo. For continuous outcomes, a mean difference compares averages, with the same need to consider scale and variability. The reported secondary comparisons include unadjusted p-values, but the registry does not report a multiplicity adjustment or a full analysis plan for these outcomes.

For the erythropoietin outcome, the arm summaries and re-calculated contrast are extremely variable, and the re-calculated unadjusted p-value is 0.978. The sponsor-reported ANCOVA result is different in estimate and p-value. Because its model details are not reported, the record alone does not explain that divergence. Clinical chemistry and hematology outcomes are binary counts, not continuous measurements. Those outcomes cannot be interpreted using the same mean-and-standard-deviation framework as the primary endpoint.

Why can over-dispersion matter?

In general, over-dispersion means that observed variability is greater than a statistical model expects. When a model underestimates variability, its standard errors can be too small and its intervals too narrow. For continuous outcomes, large observed spread relative to the mean can also make an average less precise and more sensitive to unusually high values. This is a general statistical concern, not a claim that a particular model failure has been established in this trial. The registry does not report a distributional assessment or a sensitivity analysis for the primary measure.

The large difference between the two primary standard deviations is another reason to focus on the interval and to avoid interpreting the mean contrast in isolation. The record does not provide participant-level values, distribution plots, or a robust alternative analysis. Without those, the influence of skewness or extreme values cannot be assessed from the posted summaries.

11. Confidence Intervals and Precision

The primary re-calculated mean difference is 2181.0 ng/L, with a confidence interval from -268.6 to 4631.0. Because the interval includes zero, the data are compatible with a range that includes a small difference in the opposite direction and a larger positive difference. The interval is broad in relation to the point estimate, indicating limited precision. It does not support treating the observed estimate as a precise measurement of the underlying contrast.

The sponsor-reported ANCOVA estimate is 2228.14, with a 95% confidence interval from -292.84 to 4749.11. This interval also includes zero and spans a wide range of possible differences. The two analyses therefore tell a similar broad story about uncertainty, although one is an unadjusted re-calculation and the other is the sponsor's adjusted analysis.

A confidence interval describes uncertainty under the assumptions of the analysis. It is not a range containing most individual participant values, nor does it show the probability that a particular treatment effect is true. A narrower interval would indicate greater precision; a broad interval leaves greater uncertainty about the size and direction of the contrast. The primary group standard deviations, 6212.0 and 1897.0, also indicate that individual changes were variable, especially in the GSK1278863 group.

The record does not report an interval for the planned trial effect or a clinical threshold for an important S100 beta difference. Consequently, the precision can be described statistically, but the registry does not establish what range would count as clinically meaningful. The interval should be read alongside the endpoint's status as a biomarker rather than a direct clinical event.

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 illustrative calculations based on the observed effect, not the trial's original design justification. For these continuous and binary outcomes, the figures are participants per arm, not total events. The calculation 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 a stated fraction of the observed effect.

The 80% power column gives the estimated participants per arm for a hypothetical study to have an 80% chance of detecting the specified effect under those assumptions, if that effect were the true one. The 90% power column gives the corresponding planning number for a 90% chance. Higher target power generally requires more participants for the same effect. Likewise, planning for 75% or 50% of the observed effect means targeting a smaller difference, which requires more participants than planning for the full observed effect.

These are planning illustrations, not observed power. They are especially sensitive to the observed effect used as their starting point: an unstable estimate from a small study can yield planning numbers that are unstable or unrepresentative. That limitation is pronounced for outcomes with very large variability or a small observed contrast, for which the illustrations can become extremely large. The registry's brief summary describes approximately 160 subjects as planned, but the original assumptions and justification for that design are not reported in the registry record. These calculations should not be treated as a reconstruction of that original plan.

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 = 2181.410; pooled standard deviation σ = 4592.855.
2. n = 2 × (1.96 + 0.842)² × 4592.855² / 2181.410² = 69.6, rounded up to 70 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 <- 2181.41                      # observed difference in means
sigma <- 4592.85534492482                     # 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(1920594)
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 = 70 (80% power), 94 (90% power)
 75% of observed effect: n per arm = 124 (80% power), 166 (90% power)
 50% of observed effect: n per arm = 279 (80% power), 373 (90% power)
power.t.test: 70.6 
Simulated power at n = 70 per arm: 0.801 (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 sponsor-reported primary analysis is ANCOVA, with a mean difference in final values of 2228.14, a 95% confidence interval from -292.84 to 4749.11, and p = 0.082. The simple re-calculation from the posted group summaries is unadjusted: it gives a mean difference of 2181.0, a confidence interval from -268.6 to 4631.0, and p = 0.081. The estimates are close in direction and the intervals lead to the same broad conclusion about uncertainty.

These are not identical analyses. The re-calculation does not adjust for covariates, while ANCOVA does; the registry does not report the exact covariates or model details. The sponsor's analysis is the trial-reported analysis, and the simple re-calculation is a supplementary check using summary statistics. No discrepancy is flagged in the registry record. The registry does not report enough model detail to reproduce the ANCOVA estimate independently.

The posted sponsor p-value does not establish that there is no treatment difference, and the positive point estimate does not establish that a difference exists. Both estimates should be considered with their broad confidence intervals and with the fact that the endpoint is a biomarker measurement.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists 25 outcome measures and 10 analyses overall, and it describes 23 secondary outcomes as posted. Assessing multiple outcomes creates a multiplicity concern: even when no underlying group differences exist, repeated testing can increase the chance of at least one small p-value. A statistical plan may address this through prespecified ordering, adjusted significance thresholds, or other procedures. Whether any such approach was used is not reported in the registry record.

The secondary comparisons presented as unadjusted should therefore be interpreted cautiously. Their p-values do not account for the broader set of outcomes unless an adjustment is documented. The registry record does not report whether the primary endpoint was protected by a prespecified testing hierarchy, whether secondary results were exploratory, or how the family of tests was defined. Those omissions limit conclusions from isolated secondary findings.

The primary population is described as participants with pharmacodynamic data and, for the change-from-baseline endpoint, evaluable baseline and post-dose values. The registry does not report the number of participants with missing baseline or post-dose measurements, the reasons for missingness, or an imputation method. It also does not report whether participants were analyzed according to randomized assignment for the primary endpoint. The available population definition makes clear that a complete pair of measurements was required to calculate change, but it does not describe how missing data might affect the comparison.

The record identifies randomization and double masking, but details on sequence generation, allocation concealment, operational masking, covariate selection, and the full statistical analysis plan are not reported in the registry record. These missing details matter for evaluating risk of bias and reproducibility. They do not negate the posted results, but they constrain how much can be inferred from them.

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: renal failure acute (18.5%), peripheral ischaemia (14.8%), thrombocytopenia (14.8%), multi-organ failure (14.8%), respiratory failure (11.1%).

The adverse-event record reports serious adverse events in 20 participants at risk in the GSK1278863 group and 14 participants at risk in the placebo group. Deaths were recorded for 6 participants in the GSK1278863 group and 2 in the placebo group. The record also lists other adverse events affecting 20 participants in the active-treatment group and 19 in the placebo group. These are descriptive counts; they do not establish that treatment caused an event or death.

The most frequently listed serious-event terms include renal failure acute at 18.5%, peripheral ischaemia at 14.8%, thrombocytopenia at 14.8%, multi-organ failure at 14.8%, respiratory failure at 11.1%, and spinal cord ischaemia at 7.4%. These terms describe events recorded in the trial; the registry does not establish a causal relationship to study treatment in these figures.

Safety counts should be interpreted in the context of the surgery and underlying condition described in the study record, as well as the limited enrollment. The record does not provide event timing, severity detail for each event, adjudication procedures, or an adjusted comparison of safety risks. It also does not report the causes of deaths. The counts therefore describe what was recorded, rather than providing a definitive estimate of comparative safety.

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

The primary result is a comparison of average change in a cerebrospinal-fluid biomarker between randomized groups. The posted means are numerically higher in the GSK1278863 group, but both the simple unadjusted interval and the sponsor-reported ANCOVA interval include zero. The estimates are imprecise, so the result does not resolve whether the underlying mean contrast is positive, absent, or in the opposite direction.

This is not evidence that the groups were equivalent. Establishing equivalence or non-inferiority would require a suitable design, prespecified margins, and analyses that are not reported in the registry record. Nor does the result prove that GSK1278863 changes the risk of neurologic, renal, or cardiac ischemia. The primary endpoint is S100 beta change, not a direct clinical event outcome.

The secondary outcomes add descriptive information about laboratory findings, other biomarkers, and safety events. They do not establish a consistent treatment effect across clinical outcomes, and unadjusted secondary comparisons are vulnerable to multiplicity. The adverse-event counts also do not demonstrate causation. Taken together, the registry record supports a cautious statistical conclusion: it reports a positive primary point estimate with substantial uncertainty, alongside additional findings that require careful interpretation.

The record does not report the trial's original sample-size justification, complete statistical analysis plan, or missing-data handling. Those gaps prevent a full assessment of design assumptions and the robustness of the reported comparisons. The study can inform discussion of how a biomarker was assessed in this surgical population, but the posted result should not be treated as a definitive answer to the broader clinical hypothesis.

18. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates the distinction between a point estimate and the precision of that estimate. The primary mean difference is positive, but its confidence interval includes zero and spans a wide range. Reporting the point estimate alone would obscure how uncertain the comparison remains.

It also illustrates why analysis populations and endpoints matter. A change score can be calculated only when baseline and post-dose values are available, and the record defines the primary population accordingly. The biomarker is relevant to the study rationale, but it is not the same as a direct measure of ischemic events. Statistical interpretation must keep that endpoint distinction in view.

The record also offers a useful comparison between an unadjusted re-calculation and the sponsor's ANCOVA result. Their estimates are similar in broad terms, but they are not interchangeable: adjustment depends on model details that are not reported. Finally, the range of secondary measures highlights multiplicity and the danger of treating one isolated result as decisive when many outcomes are assessed.

These features make the record a case study in calibrated interpretation. It encourages attention to the uncertainty interval, the analysis population, the distinction between biomarker and clinical outcomes, and the limits imposed by unreported methods. Those considerations are central to evaluating clinical trial results without overstating what the data establish.

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 recorded enrollment was 57, whereas the brief summary describes approximately 160 subjects as planned. The registry does not explain the difference, provide the original sample-size justification, or state whether the trial was stopped early or changed its planned enrollment. Any such explanation would be speculation. The available record therefore cannot establish why the completed enrollment differed from the planned description.

The primary analysis required both baseline and post-dose measurements, but the number of participants missing either measurement and the reasons are not reported in the registry record. The record does not describe imputation, sensitivity analyses, or the effect of missing measurements on the analysis population.

The primary measure is a biomarker endpoint, and the posted values show substantial variability. The registry does not provide participant-level data, distribution assessments, or robustness analyses, so the effect of skewness or unusually large values cannot be evaluated. The sponsor ANCOVA model's covariates and implementation details are also not reported, limiting independent reproduction of the adjusted estimate.

Further limitations include unreported details on random sequence generation and allocation concealment, the absence of a reported multiplicity strategy, and limited information on adverse-event attribution and timing. These gaps restrict assessment of the strength and reproducibility of the results. They do not establish that the methods were inadequate; they mean the registry record does not document the details needed to evaluate them fully.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study of GSK1278863 to Reduce Ischemic Events in Patients Undergoing Thoracic Aortic Aneurysm Repair — NCT01920594.
• Registry record last updated 2017-12-07; status last verified 2017-11.
• Results first posted 2017-12-07.
• 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

This completed randomized, double-blind, parallel-group Phase II trial compared GSK1278863 with matching placebo in patients undergoing descending thoracic aorta or thoracoabdominal aortic aneurysm repair. The registry records enrollment of 57 and describes approximately 160 subjects as planned.

The primary endpoint was change in cerebrospinal-fluid S100 beta from baseline to peak within 48 hours after repair. The re-calculated unadjusted mean difference was 2181.0 ng/L, with a confidence interval from -268.6 to 4631.0. The sponsor-reported ANCOVA estimate was 2228.14, with a 95% confidence interval from -292.84 to 4749.11. Both intervals include zero.

The record also lists secondary outcomes and adverse events, but the multiplicity approach, full statistical plan, and several analysis details are not reported in the registry record. The primary result is a biomarker comparison with substantial uncertainty and should not be equated with a direct estimate of clinical ischemic events.