A Study to Evaluate the Effect of Belimumab on Vaccine Responses in Subjects With Systemic Lupus Erythematosus (SLE): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate the Effect of Belimumab on Vaccine Responses in Subjects With Systemic Lupus Erythematosus (SLE) (NCT01597492), a randomized phase 4 study. The primary outcome is Number of Participants With Positive Antibody Responses to at Least One of the 23 Pneumococcal Vaccine Serotypes 4 Weeks Post-vaccination.

2. Clinical Question

The central comparison concerns the timing of vaccination relative to belimumab treatment, not belimumab versus no belimumab. Both groups are described as receiving belimumab; one was vaccinated early and the other later. The outcome is a binary participant-level result: whether a participant met the registry definition of a positive response for at least one serotype.

The registry’s response definition uses a 2-fold or greater increase from pre-vaccination antibody levels. When pre-vaccination levels were unquantifiable, a post-vaccination level >=0.6 micrograms per milliliter counted as positive. The posted primary outcome therefore asks whether at least one serotype crossed the defined response threshold, rather than measuring the size of antibody changes across all serotypes.

The clinical question is limited to these response criteria and assessment times. The registry record does not report a separate comparison with participants who did not receive belimumab, nor does it establish what would happen under other vaccination schedules or in other populations.

3. Trial Design

NCT01597492 is a Phase 4, randomized, open-label, parallel-group study of belimumab and pneumococcal vaccine responses in participants with systemic lupus erythematosus. The registry lists an enrollment of 79 and a completed study. The study began on 2012-05-31, with primary completion on 2015-09-01. Human Genome Sciences Inc., a GSK Company, is listed as the sponsor.

Participants were assigned to Belimumab Plus Early Vaccination or Belimumab Plus Late Vaccination. The registry reports no masking. In the early cohort, vaccination occurred at Day 0 and titers were measured on Day 28, before the first dose of belimumab. In the late cohort, vaccination occurred at Week 24; titers were measured on Day 196, or Week 28, before the last belimumab dose. The late-cohort evaluability description requires at least 5 of the 7 belimumab doses through Week 24, vaccination at Week 24, and titers at Week 28.

The primary outcome counts participants with a positive antibody response to at least one of the 23 pneumococcal serotypes four weeks after vaccination. The posted outcome counts were 32 of 33 participants in the early group and 40 of 41 in the late group. These are the denominators and counts reported for the primary outcome; the registry’s vaccine-titer analyses are described as using the as-treated population of participants who received at least one dose of belimumab.

The registry record does not report the original sample-size justification, detailed randomization procedures, allocation concealment, or the trial’s full statistical analysis plan. Those omissions limit what can be concluded about design choices beyond the recorded randomized, open-label, parallel-group structure.

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 79 participants were enrolled, of which 34 were randomised to the early cohort and 45 were randomized to the late cohort. All randomized participants received at least 1 dose of vaccine and/or belimumab and were included in the Intent-to-Treat (ITT) Population.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: As-treated Population: all participants who received at least one dose of belimumab. All analyses of vaccine titers were performed on the As-treated Population.

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 Number of Participants With Positive Antibody Responses to at Least One of the 23 Pneumococcal Vaccine Serotypes 4 Weeks Post-vaccination, measured over Four weeks after vaccination.

The primary endpoint counts participants with a positive antibody response to at least one of the 23 pneumococcal vaccine serotypes four weeks after vaccination. It is a binary endpoint: each evaluable participant is counted as having met or not met the response definition. A participant does not need to respond to every serotype to meet this endpoint.

The registry defines a positive response as a 2-fold or greater increase from pre-vaccination levels. If a pre-vaccination antibody level could not be quantified, the definition instead uses a post-vaccination level >=0.6 micrograms per milliliter. The timing of the four-week assessment differs by cohort in calendar terms: Day 28 in the early group and Day 196, or Week 28, in the late group.

In general, a threshold-based endpoint makes a complex immune measurement easier to summarize as a participant count. It also discards information about the size of responses and about responses to individual serotypes beyond whether at least one met the criterion. The endpoint is therefore useful for answering the registry’s specific question about achieving a qualifying response, but it does not describe the full pattern of antibody titers.

The registry also defines which participants were evaluable in each cohort. The posted outcome should be interpreted in that context; the record does not report a separate endpoint summary for participants excluded from those evaluability descriptions.

7. Results Posted in the Registry

• Belimumab Plus Late Vaccination: 97.6% (40/41) versus 97.0% (32/33) on control; risk difference 0.6 percentage points (95% CI -6.9 to 8.1), risk ratio 1.01 (0.93–1.09); unadjusted two-proportion z-test p = 0.876.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a binary endpoint count?

The primary outcome is binary because it classifies each participant as having or not having a positive response under the stated threshold. The reported counts are 32 of 33 in the early group and 40 of 41 in the late group. This summarizes whether a qualifying response occurred for at least one of the 23 serotypes. It does not preserve the magnitude of each person’s antibody change, or show how many serotypes each participant responded to beyond the minimum needed to meet the endpoint.

In general, binary outcomes can be summarized as counts, proportions, and contrasts between groups. A count is tied to its denominator: the same number of responses could imply different proportions in groups of different sizes. For that reason, the denominators are essential context for reading the response counts.

What does the risk difference describe?

The registry reports a risk difference comparing late vaccination with early vaccination. In general, this is the difference between the two groups’ outcome proportions, expressed on an absolute scale. A value of zero corresponds to equal proportions. The reported confidence interval crosses zero, so the data are compatible with no difference and with differences in either direction. The point estimate alone should not be read as a definitive group effect.

The registry also reports a risk ratio and an odds ratio. In general, a risk ratio compares the proportions with the outcome, whereas an odds ratio compares the odds. A ratio of 1 corresponds to no relative difference. These measures can be useful alongside the risk difference, but they answer different questions and their values are not directly interchangeable. The reported risk ratio is 1.006 and the reported odds ratio is 1.25; both intervals include 1.

How should the p-value be understood?

The posted p-value is 0.876. In general, a p-value describes how unusual the observed data, or data more extreme, would be under a specified null model and analysis procedure. It is not the probability that the null hypothesis is true, and it does not measure the probability that a result will reproduce. It also cannot establish that two groups are equivalent. Interpretation depends on the model, endpoint, and analysis decisions; those details are not fully reported in the registry record.

The record lists no analysis records, while it does post a group comparison summary. The full inferential procedure behind that summary, including any covariate adjustment or alternative analysis, is not reported in the registry record. No adjustment should therefore be assumed.

Why does the response threshold matter?

The endpoint classifies a response using a 2-fold or greater increase from pre-vaccination levels, with a post-vaccination threshold of >=0.6 micrograms per milliliter for unquantifiable pre-vaccination levels. This rule turns antibody measurements into a yes-or-no outcome. In general, thresholding can aid interpretation, but findings can depend on the chosen definition. The registry record gives the definition used, but it does not report alternative thresholds or sensitivity analyses using other definitions.

The primary outcome is also an “at least one” endpoint across 23 serotypes. It does not mean that each participant responded to all serotypes. It can classify a participant as positive after a qualifying response in one serotype, even if response patterns differ across the others. The registry record does not report the detailed serotype-specific results in the registry record for this analysis.

What does uncertainty mean in this comparison?

In general, a confidence interval describes a range of values compatible with the data under the statistical method used. It reflects sampling uncertainty, not every possible source of error. The reported risk-difference interval spans negative and positive values, indicating uncertainty about both direction and magnitude. The interval should be considered together with the point estimate and p-value, rather than treated as a separate verdict.

In general, over-dispersion matters when observed variation is greater than a statistical model expects; if ignored, standard errors can be too small and intervals too narrow. The registry’s primary outcome is a participant-level binary count, and the record does not report an over-dispersion assessment or a model that accounts for it. This does not establish that over-dispersion was present or relevant; it is a reminder that analytic details matter when interpreting model-based uncertainty.

The registry describes vaccine-titer analyses as using the as-treated population: all participants who received at least one dose of belimumab. It separately gives cohort-specific evaluability conditions for the primary response outcome. These analysis-population descriptions should not be conflated with an intention-to-treat analysis; the registry record does not report that the primary outcome was analyzed by intention to treat.

10. Confidence Intervals and Precision

The main precision issue is that the observed group contrast is accompanied by an interval that crosses zero. In general, a confidence interval represents uncertainty around an estimate under the statistical method used. For the reported risk difference, the interval includes values favoring either group as well as no difference. This means the data do not precisely establish the direction of the contrast.

The posted p-value is 0.876. Considered with the interval, it does not indicate a clear separation between the groups under the reported comparison. It should not be interpreted as proof that the groups are the same. Demonstrating equivalence or non-inferiority would require a prespecified margin and an analysis designed for that question; such a margin and design are not reported in the registry record.

The relative effect estimates are also imprecise. The risk ratio estimate is 1.006, and the odds ratio estimate is 1.25. Their reported confidence intervals include 1, the relative-measure value corresponding to no contrast. The odds ratio interval is especially broad, underscoring that its point estimate should not be interpreted in isolation.

Precision depends on the number of observations and the variability of the outcome, among other factors. In general, fewer observations can make estimates less stable, particularly when outcomes are common and there are few participants without the outcome. The posted outcome counts show that the groups are small, but the registry record does not report a formal precision target or the original sample-size justification.

Confidence intervals capture sampling uncertainty under an analysis; they do not correct for limitations such as open-label conduct, outcome definition, missing data, or differences in who was evaluable. Those matters require separate consideration. The registry record does not report enough detail to determine the impact of any missing outcomes or protocol deviations on this contrast.

11. 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, not total events. They are illustrations based on the observed effect, not a statement of the number enrolled or analyzed in the trial. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect equal to the stated fraction of the observed effect.

The 80% power column gives the planned number of participants per arm for a hypothetical study to have an 80% chance of detecting the specified effect if that effect and the assumptions used for planning were correct. The 90% power column uses a higher probability of detection under those same hypothetical conditions and therefore has larger participant requirements. These are planning quantities, not observed power for the completed study.

The displayed rows consider planning for 100%, 75%, and 50% of the observed effect. At 100% of the observed effect, the figures are 11,942 participants per arm for 80% power and 15,987 for 90% power. At 75% of the observed effect, they are 21,789 and 29,168 per arm. At 50% of the observed effect, they are 50,277 and 67,306 per arm. The pattern illustrates that smaller effects require more participants to detect under the same planning assumptions.

These figures should not be mistaken for the trial’s original design target. The registry record does not report the original sample-size assumptions or justification. The illustration is tied to the observed effect, which is itself uncertain, and it does not show that the completed study had a particular probability of finding a difference.

12. 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 = 0.970; treatment proportion p₁ = 0.976; average p̄ = 0.973.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.2306 = 0.4521.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.2306 = 0.1941.
4. n = (0.4521 + 0.1941)² / (0.0059)² = 11,941.9, rounded up to 11,942 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 <- 0.96969696969697                          # control proportion
p1 <- 0.975609756097561                          # 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(1597492)
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 = 11,942 (80% power), 15,987 (90% power)
 75% of observed effect: n per arm = 21,789 (80% power), 29,168 (90% power)
 50% of observed effect: n per arm = 50,277 (80% power), 67,306 (90% power)
power.prop.test: 11941.9 
Simulated power at n = 11,942 per arm: 0.802 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The primary endpoint concerns response to at least one of 23 serotypes, combining multiple antibody measurements into a single participant-level outcome. That endpoint definition is not the same as conducting a separate primary test for every serotype. The registry lists one outcome measure and does not list secondary outcomes in the posted results. It does not report a multiplicity plan, adjustments for multiple comparisons, or analyses of serotype-specific outcomes.

In general, multiplicity arises when several hypotheses or outcomes are tested: the chance of at least one false-positive finding can increase unless the analysis accounts for the testing strategy. Whether an adjustment is needed depends on which outcomes and comparisons were planned and how confirmatory claims are intended. The registry record does not report enough of the statistical analysis plan to determine the full multiplicity strategy.

The registry describes the vaccine-titer analyses as using the as-treated population, defined as participants who received at least one dose of belimumab. For the early cohort, evaluability required vaccination at Day 0 and titers at Week 4. For the late cohort, evaluability required at least 5 of the 7 belimumab doses through Week 24, vaccination at Week 24, and titers at Week 28.

The posted primary outcome denominators are 33 in the early group and 41 in the late group, while the registry lists enrollment of 79. The record does not provide a complete explanation of the difference between enrollment and outcome denominators, nor does it report how missing antibody measurements were handled. It is therefore not possible to assess from the registry record whether missingness could have affected the comparison.

The record does not report imputation methods, sensitivity analyses for missing data, or a detailed intention-to-treat analysis of this endpoint. These are limits on interpretation, not evidence that a particular method was or was not used. The as-treated description should be kept distinct from any unreported analysis approach.

14. Safety Results

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

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: any event (5.9%), cardiac failure congestive (2.9%), pericarditis lupus (2.9%), chronic kidney disease (2.9%), glomerulonephritis proliferative (2.9%).

The registry reports serious adverse events for groups with 34 participants at risk in the early-vaccination group and 45 at risk in the late-vaccination group. It records 4 affected participants in the early group and 3 in the late group. Other adverse-event counts are 21 and 32, respectively. The safety denominators differ from the primary-outcome denominators, so these counts should be read within their own reported groups rather than combined with the vaccine-response results.

The listed serious-event terms include “Any event” at 5.9%, and cardiac failure congestive, pericarditis lupus, chronic kidney disease, glomerulonephritis proliferative, and transaminases increased, each at 2.9%. These are registry-reported terms and percentages; the record excerpt does not specify arm-specific percentages for these terms. Death counts are not reported in the registry record.

In general, adverse-event counts describe events observed and recorded during a study; they do not by themselves establish that an intervention caused an event. The reported figures do not provide a randomized comparison against a group receiving no belimumab, and event counts can be affected by follow-up and ascertainment. The registry record does not report enough detail to evaluate those factors or to make causal conclusions from the safety counts.

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

The posted results describe a high number of participants meeting the registry’s positive-response definition in both vaccination-timing groups: 32 of 33 in the early group and 40 of 41 in the late group. The estimated risk difference is close to zero, while its confidence interval includes both negative and positive values. The p-value is 0.876. Taken together, these statistics show similar observed response counts but do not precisely establish the size or direction of a difference between the groups.

The comparison is specifically early versus late vaccination among participants receiving belimumab. It does not isolate the effect of belimumab itself, because both groups received belimumab. Nor does it establish that the two schedules are equivalent: the registry record does not report an equivalence margin or a design intended to establish equivalence.

The outcome is limited to a positive antibody response to at least one serotype using the registry’s threshold and four-week assessment. It does not summarize the magnitude of antibody responses across all 23 serotypes, and it does not show the response profile for each serotype. The result also applies to the analyzed participants and timing conditions described in the registry, not automatically to other schedules or populations.

In general, a nonsignificant comparison is not evidence that no meaningful difference exists. The interval’s width, the small outcome groups, the endpoint definition, and incomplete reporting of analysis details all matter. The data support a descriptive account of the observed counts and an uncertain group contrast; they do not support a claim that one schedule is better or that the schedules are interchangeable.

The safety figures are also descriptive. They report serious and other adverse-event counts in the registry’s safety groups, but they do not by themselves attribute events to treatment. Overall, the record permits a focused interpretation of the reported response endpoint while leaving broader clinical and causal questions unresolved.

16. Why This Trial Matters Statistically

This study is statistically instructive because the outcome was common in both groups, yet the group contrast remains uncertain. Looking only at the response counts can make the groups appear alike; the confidence interval adds the important qualification that the data do not tightly constrain the possible difference. Statistical interpretation depends on both the estimated contrast and its precision.

The endpoint also illustrates how a composite threshold can simplify measurement while narrowing the question. A participant counted as positive after meeting the response criterion for at least one of 23 serotypes. This creates a clear participant-level outcome, but it does not describe whether responses were broad, how large they were, or which serotypes contributed. The threshold and “at least one” rule are therefore central to what the result means.

The randomized design supports a comparison of the assigned timing groups, but the open-label structure and the absence of detailed analysis reporting limit assessment of potential bias and analytic choices. The two groups both received belimumab, so the comparison cannot answer whether belimumab changes vaccine responses relative to no belimumab. Clear identification of the estimand—the exact contrast the analysis aims to describe—is essential to avoid extending the result beyond its design.

Finally, the planning illustration shows how sample-size requirements depend on the effect one wants to detect. Smaller target effects call for more participants under the stated assumptions. That is a planning relationship, not proof that a study was adequately powered or that a non-significant comparison demonstrates no effect. The registry does not report the original design assumptions, so the illustration should remain separate from claims about the trial’s planned ability to detect a difference.

17. 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 describes a randomized, open-label study, but it does not report the randomization procedure, allocation concealment, or methods used to minimize bias. These details are relevant when judging how reliably the assigned groups represent the intended comparison.

The primary outcome denominators are smaller than the total enrollment, and the registry provides cohort-specific evaluability rules. It does not provide a complete participant-flow account or explain the handling of missing antibody measurements. The possible effect of exclusions or missingness on the estimated contrast is therefore not reported in the registry record.

The early and late groups were assessed at different study times relative to belimumab exposure. Early vaccination was at Day 0, with titers at Day 28 before the first belimumab dose; late vaccination was at Week 24 after the specified belimumab dosing, with titers at Week 28 before the last dose. The timing comparison is consequently tied to these distinct treatment and assessment contexts, rather than being a comparison of timing in isolation from all other study circumstances.

The outcome is a threshold-based response to at least one of 23 serotypes. The record does not report the serotype-specific results, response magnitudes, or alternative endpoint definitions in the registry record for this analysis. The primary outcome therefore gives a limited view of the immune-response pattern.

The registry record does not report the original sample-size justification, detailed analysis plan, multiplicity strategy, or sensitivity analyses. It also does not report a no-belimumab comparator. These limitations constrain interpretation of precision, missing-data effects, and causal attribution; they do not establish that any particular bias or analysis problem occurred.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: A Study to Evaluate the Effect of Belimumab on Vaccine Responses in Subjects With Systemic Lupus Erythematosus (SLE) — NCT01597492.
• Registry record last updated 2018-08-16; status last verified 2018-07.
• Results first posted 2016-07-25.
• 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.

21. Record Summary

NCT01597492 is a completed Phase 4, randomized, open-label, parallel-group study sponsored by Human Genome Sciences Inc., a GSK Company. It enrolled 79 participants with systemic lupus erythematosus and compared belimumab with early pneumococcal vaccination against belimumab with late vaccination.

The primary endpoint counted participants with a positive response to at least one of the 23 serotypes four weeks after vaccination. The posted counts were 32 of 33 for early vaccination and 40 of 41 for late vaccination. The reported risk-difference comparison has a p-value of 0.876, and its confidence interval crosses zero.

The result describes the specified response threshold and vaccination schedules. The registry record does not report the original sample-size justification, a full statistical analysis plan, or complete missing-data handling details.