GSK1550188 A 52 Week Study of Belimumab Versus Placebo in the Treatment of Subjects With Systemic Lupus Erythematosus (SLE) Located in Northeast Asia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for GSK1550188 A 52 Week Study of Belimumab Versus Placebo in the Treatment of Subjects With Systemic Lupus Erythematosus (SLE) Located in Northeast Asia (NCT01345253), a randomized phase 3 study. The primary outcome is Percent of Participants Achieving Systemic Lupus Erythematosus (SLE) Responder Index (SRI) Response Rate at Week 52 for Double-blind Phase..

2. Clinical Question

The question is about the proportion meeting a composite response definition at a fixed assessment time, not about complete absence of lupus activity, long-term disease control, or a single symptom. The study also recorded secondary measures, including other response thresholds and time to a severe flare. Those measures offer additional perspectives, but they do not replace the primary Week 52 comparison.

3. Trial Design

This was a randomized, quadruple-masked, parallel-group Phase 3 study. The registry lists an enrollment of 709 and the status as completed. Participants were assigned to belimumab 10 mg/kg or placebo. The study was described as evaluating belimumab in addition to standard therapy compared with placebo in people with SLE in Northeast Asia over a 52-week period. Specific details of the standard therapy and the allocation procedure are not reported in the registry record.

The primary outcome was assessed at Week 52 in the double-blind phase. The primary analysis population was defined as randomized participants who received at least one dose of study treatment, with participants from site 086485 excluded. The outcome-specific arm counts are 217 for placebo and 446 for belimumab 10 mg/kg. Other outcomes have different listed counts; reasons for those differences are not reported in the registry record.

The registry lists a study start date of 2011-05-23 and a primary completion date of 2015-09-15. It also lists an open-label phase, including a secondary SRI response outcome, and reports safety information for both double-blind and open-label phases. The duration and entry rules for the open-label phase are not reported in the registry record.

Randomization is intended to make treatment groups comparable in expectation, while masking is intended to reduce the influence of treatment knowledge on participant, clinician, and assessment decisions. The registry labels the study quadruple-masked, but does not specify which roles were masked. The original sample-size justification, allocation ratio, details of concealment, interim analyses, and full statistical analysis plan are not reported in the registry record.

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 707 par. were randomized, 705 received at least 1 dose of investigational product (Safety Population), 677 par. were included in the primary efficacy population (MITT Population) of which 663 were evaluable for the primary endpoint. The participant flow and baseline characteristics are presented for MITT Population (Pop).

Double-blind (DB) Phase (52 Weeks)

Reasons for not completing this period:

Open-label Phase (Up to Year 5 Week 44)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Modified Intention-to-Treat (MITT) Population: all participants who were randomized and treated with at least one dose of study treatment, with exclusion of participants from the site 086485.

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 Percent of Participants Achieving Systemic Lupus Erythematosus (SLE) Responder Index (SRI) Response Rate at Week 52 for Double-blind Phase., measured over Week 52.

The primary endpoint is the percentage of participants achieving an SLE Responder Index response at Week 52 of the double-blind phase. It combines evidence of improvement with rules intended to prevent a response classification when specified disease activity worsens. A response requires at least a 4-point reduction from baseline in the SELENA SLEDAI score, no worsening in physician global assessment (PGA), and no new BILAG A organ-domain score or two new BILAG B organ-domain scores compared with baseline.

The registry describes no worsening in PGA as an increase of less than 0.30 points from baseline. SELENA SLEDAI scores range from 0, suggesting no lupus activity, to a maximum calculable score of 105. PGA ranges from 0, no activity, to 3, severe activity; BILAG has no range specified in the record.

As a composite, SRI response asks whether a participant meets several conditions together, rather than whether only one score improves. This can make the endpoint more clinically interpretable than an improvement score alone because the response definition also accounts for selected signs of worsening. It remains a threshold-based classification at a specified assessment time: it does not describe every change in disease activity or the course of disease outside the assessment window. The registry’s reason for selecting this endpoint is not reported in the registry record.

7. Results Posted in the Registry

• Belimumab 10 mg/kg: 53.8% (240/446) versus 40.1% (87/217) on control; risk difference 13.7 percentage points (95% CI 5.7 to 21.7), risk ratio 1.34 (1.12–1.61); unadjusted two-proportion z-test p = <0.001.

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.

Percent of Participants With >=4 Point Reduction From Baseline in SELENA SLEDAI Score at Week 52 for Double-blind Phase.

• 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:
  • Belimumab 10 mg/kg: 55.7% (249/447) versus 42.2% (92/218) on control; risk difference 13.5 percentage points (95% CI 5.5 to 21.5), risk ratio 1.32 (1.11–1.57); unadjusted two-proportion z-test p = 0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Belimumab 10 mg/kg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percent of Participants With SRI7 Response at Week 52 for Double-blind Phase.

• 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:
  • Belimumab 10 mg/kg: 32.4% (119/367) versus 23.5% (43/183) on control; risk difference 8.9 percentage points (95% CI 1.1 to 16.7), risk ratio 1.38 (1.02–1.86); unadjusted two-proportion z-test p = 0.030.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Belimumab 10 mg/kg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Time to First Severe SLE Flare Index (SFI) Flare Over 52 Weeks for Double-blind Phase.

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor (Regression, Cox); it was not recalculated.
• Result:
  • Placebo vs Belimumab 10 mg/kg: hazard ratio 0.50 (95% CI 0.34–0.73), reported p < 0.001 (as reported in the registry).
• Interpretation:
  • The 95% confidence interval excludes no difference (a ratio of 1), so the hazard of the event is statistically lower for Placebo vs Belimumab 10 mg/kg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes cover more than one aspect of SLE response and course: a SELENA SLEDAI improvement threshold, a higher SRI response threshold, time to first severe flare, and outcomes extending into the open-label phase. Taken together, the posted double-blind results generally point in the same direction as the primary result: the response measures favor the belimumab comparison, and the reported time-to-event result is consistent with a lower hazard for severe flare. These outcomes are related to the primary result, but each uses its own definition, population, or analysis scale.

That broad directional pattern adds context, but it does not make the secondary outcomes equivalent to the primary endpoint. A stricter response threshold can classify fewer participants as responders; time to an event asks a different question from response status at Week 52. The open-label phase is also distinct from the randomized, double-blind comparison, so its results should not be treated as another direct randomized comparison with placebo.

Secondary findings generally carry less evidential weight when several outcomes are examined and the record does not state whether or how multiplicity was controlled. Multiple analyses increase the chance that at least one result will appear notable under a null pattern. The registry record lists five posted secondary outcomes, but the statistical hierarchy, any prespecified testing sequence, and any multiplicity adjustment are not reported in the registry record. Secondary results can therefore support coherence and generate questions, but should not be treated as independent confirmation of the primary result or as replacements for it.

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

• Percent of Participants With >=4 Point Reduction From Baseline in SELENA SLEDAI Score at Week 52 for Double-blind Phase.
• Percent of Participants With SRI7 Response at Week 52 for Double-blind Phase.
• Number of Days of Daily Prednisone Dose <=7.5 mg/Day and/or Reduced by 50 Percent From Baseline Over 52 Weeks for Double-blind Phase.
• Time to First Severe SLE Flare Index (SFI) Flare Over 52 Weeks for Double-blind Phase.
• Percent of Participants Achieving SLE SRI Response Rate for Open-label (OL) Phase

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is binary at the participant level: a participant either meets the SRI response definition at Week 52 or does not. For a binary outcome, a percentage describes the observed fraction classified as having the event—in this case, response. The percentages are the most direct way to see the outcome in each group, while a comparison measure summarizes how the groups differ.

What does the risk difference describe?

The risk difference is the response proportion in one group minus the response proportion in the comparison group, expressed in the recalculation as a proportion. The reported estimate is 0.137. Its 95% confidence interval is 0.057 to 0.217. A risk difference is an absolute comparison of proportions and retains the outcome’s original scale. It is not the same as an odds ratio or a relative risk. The recalculation uses the posted summary counts and is unadjusted; it does not account for covariates or other features of an individual-level analysis.

What do the odds ratio and relative risk mean?

Odds are the chance of an event divided by the chance of no event. An odds ratio compares those odds between groups. A relative risk compares the event proportions themselves. Both are relative measures, whereas the risk difference is an absolute measure. Their numerical values do not have the same interpretation, and an odds ratio should not be described as a percentage-point change. The registry reports a logistic-regression odds ratio for the primary analysis. The record does not give further model details, such as covariates or handling of site effects.

How should the p-value be read?

A p-value describes how unusual results at least as incompatible with a specified null model would be if that model and the test assumptions were correct. It is not the probability that the null hypothesis is true, and it does not measure the probability that the result occurred by chance in a general sense. The unadjusted recalculated p-value is <0.001; the sponsor-reported logistic-regression p-value is < 0.001. The analysis method and estimand differ, so those values need not be identical. Neither p-value by itself describes effect size, precision, clinical importance, or bias.

Why use a composite response?

The SRI definition joins improvement in SELENA SLEDAI with limits on worsening in PGA and BILAG. A composite can capture multiple dimensions of a condition in one endpoint, but the result depends on all components and thresholds in its definition. A participant who improves on one component may not qualify if another criterion is not met. Conversely, the overall response percentage does not show how often each component independently changed. Component-level results and the handling of individual endpoint data are not reported in the registry record.

What does the time-to-event analysis add?

A time-to-event outcome concerns how long it takes for an event to occur, while accounting for the fact that some participants may not have an observed event before follow-up ends. The secondary severe-flare analysis is reported using a Cox regression hazard ratio of 0.50, with a 95% confidence interval from 0.34 to 0.73 and p< 0.001. A hazard ratio compares event rates over follow-up under the model; it is not a direct comparison of the percentages who ever experienced an event, nor does it state a difference in days. The registry does not report the median time values or further details of censoring and model assumptions.

General statistical principle: standard methods for binary and time-to-event outcomes rely on assumptions. Independence of participants, correct outcome classification, suitable model specification, and appropriate handling of missing or censored observations can affect inference. For binary outcomes, differences in baseline risk can make absolute and relative measures tell different aspects of the same comparison. For time-to-event models, the proportional-hazards assumption is often relevant; whether it was assessed is not reported in the registry record.

11. Confidence Intervals and Precision

A confidence interval gives a range of values compatible with the observed data under the method’s assumptions. It communicates uncertainty in an estimate; it does not guarantee that the true value lies inside the interval, and it does not include every possible source of bias or measurement error. A narrower interval indicates greater statistical precision for the particular estimate, while a wider interval indicates more uncertainty.

For the primary unadjusted risk difference, the estimate is 0.137 and the 95% confidence interval is 0.057 to 0.217. Both interval limits are above zero. The interval conveys uncertainty in the absolute comparison of response proportions and is more informative about the range of plausible effects than a p-value alone. It does not establish whether the observed difference is clinically important, because a threshold for clinical importance is not reported in the registry record.

The sponsor-reported primary odds ratio is 1.99, with a 95% confidence interval from 1.40 to 2.82. This interval is on the odds-ratio scale, not the risk-difference scale, so the two intervals should not be compared as if they measured the same quantity. The reported logistic-regression estimate and the unadjusted recalculation also arise from different methods.

Secondary confidence intervals likewise describe uncertainty for their specific measures. For example, the reported hazard ratio for time to first severe SLE flare has a 95% confidence interval from 0.34 to 0.73. That interval is about a time-to-event model parameter, not the primary endpoint’s response proportion. Confidence intervals do not resolve concerns about unreported analysis choices, multiplicity, missing data, or the selected analysis population.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size illustrations use the observed effect as a starting point and show planning scenarios for detecting a stated fraction of that effect. For the binary outcomes, the figures are participants per arm. For the time-to-event outcome, the figures are total events, not participants. The rows distinguish planning for 100%, 75%, or 50% of the observed effect. Smaller effects require more participants for the binary outcomes, or more total events for the time-to-event outcome, to reach the same target power.

The listed planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The target power columns describe the chance, under the planning assumptions and effect being targeted, of detecting that effect using the specified test if it is truly present. The 80% power column therefore corresponds to a lower target probability of detection than the 90% power column; neither is a guarantee for a future study. These are planning illustrations based on the observed effect, not observed power.

For example, the primary-outcome planning row for 100% of the observed effect lists 207 participants per arm for 80% power and 276 participants per arm for 90% power. The row for 50% of the observed effect lists 819 participants per arm for 80% power and 1,096 participants per arm for 90% power. These are illustrations, not evidence about the original trial’s design. The trial’s own sample-size assumptions and justification are not reported in the registry record.

For time-to-event planning, the corresponding unit is total events. The row for 100% of the observed effect lists 66 total events for 80% power and 88 total events for 90% power. These figures cannot be read as participant counts. The planning values do not incorporate any additional assumptions beyond those stated in the registry record.

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 = 0.401; treatment proportion p₁ = 0.538; average p̄ = 0.470.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7058 = 1.3833.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6991 = 0.5884.
4. n = (1.3833 + 0.5884)² / (0.1372)² = 206.5, rounded up to 207 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 <- 0.400921658986175                          # control proportion
p1 <- 0.538116591928251                          # 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(1345253)
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 = 207 (80% power), 276 (90% power)
 75% of observed effect: n per arm = 367 (80% power), 490 (90% power)
 50% of observed effect: n per arm = 819 (80% power), 1,096 (90% power)
power.prop.test: 206.5 
Simulated power at n = 207 per arm: 0.813 (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’s reported primary analysis is logistic regression, with an odds ratio of 1.99, a 95% confidence interval from 1.40 to 2.82, and p< 0.001. The recalculation uses the posted arm-level counts to estimate an unadjusted risk difference of 0.137, with a 95% confidence interval from 0.057 to 0.217 and p < 0.001. The record does not flag a discrepancy between the posted analysis and the recalculation.

These are not duplicate estimates on the same scale. Logistic regression reports an odds ratio, whereas the recalculation compares response proportions by risk difference. The registry does not provide enough model details to determine whether the regression included covariates or other adjustments, so no such detail should be inferred. Differences between the estimates and p-values can reflect the distinct estimands and methods rather than an error.

The sponsor-reported regression result is the registry’s own stated analysis. The unadjusted recalculation is a complementary summary based on posted counts and should be read as such. Neither summary answers questions about missing-data handling or multiplicity that are not described in the registry record.

15. Multiplicity, Missing Data and Analysis Populations

The primary analysis population is identified as the Modified Intention-to-Treat population: all randomized participants treated with at least one dose, excluding participants from site 086485. That definition clarifies who was included in the posted primary outcome analysis, but it is not the same as a complete description of the analysis plan. The registry record does not report the reasons for excluding the site or the number of participants excluded from it.

The record lists multiple secondary outcomes and analyses, but multiplicity procedures are not reported in the registry record. It is not reported in the registry record whether secondary hypotheses were ordered, whether a family-wise error approach was used, or whether secondary p-values were adjusted. Consequently, the secondary p-values should be interpreted with caution rather than as if each were the only test conducted.

Missing-data methods are not reported in the registry record. The record does not state how participants without a Week 52 assessment were handled, whether any imputation was used, or whether sensitivity analyses explored alternative assumptions. The observed analysis counts differ across outcomes, but reasons for those differences are not reported in the registry record. Without the missing-data rules and outcome-specific disposition, it is not possible to assess from the registry results alone how incomplete follow-up may have affected the estimates.

The original statistical analysis plan, interim analyses, stopping rules, and sample-size justification are not reported in the registry record. The sponsor’s logistic-regression method is named, but details of its specification are not reported. These omissions limit assessment of prespecification, analysis flexibility, and robustness. They do not negate the posted estimates, but they constrain how fully those estimates can be evaluated.

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: lupus nephritis (2.8%), pyrexia (1.7%), herpes zoster (1.4%), osteonecrosis (1.4%), pneumonia (0.9%).

The registry reports serious adverse event counts and other adverse event counts separately for the double-blind and open-label phases. In the double-blind phase, 43 participants with serious adverse events were listed among 235 at risk in the placebo group, and 58 among 470 at risk in the belimumab 10 mg/kg group. Other adverse events were reported for 108 of 235 and 207 of 470, respectively. The registry also lists 1 death in the placebo group and 0 in the belimumab group during the double-blind phase.

For the open-label phase, the registry lists 96 participants with serious adverse events among 424 at risk, 263 with other adverse events among 424 at risk, and 1 death. This phase has a different treatment context and should not be combined with the double-blind comparison as though it were a randomized placebo comparison. The registry lists lupus nephritis (2.8%), pyrexia (1.7%), herpes zoster (1.4%), osteonecrosis (1.4%), pneumonia (0.9%), and lung infection (0.9%) among the top serious adverse event terms.

These are descriptive counts, not proof that an event was caused by treatment. The registry record does not provide event-level timing, severity detail beyond the serious-event category, or causal attribution for these counts. Comparisons should retain the phase and at-risk denominator context.

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

The primary result describes the proportion of participants in the defined analysis population who met the SRI response criteria at Week 52. The posted percentages were 40.1 for placebo and 53.8 for belimumab 10 mg/kg. The unadjusted risk-difference estimate and its confidence interval indicate an observed separation between groups in this study, with uncertainty quantified on the risk-difference scale. The sponsor-reported logistic-regression analysis also reports an odds ratio above 1, with its confidence interval above 1.

This result does not mean that every participant in the belimumab group responded, that every participant in the placebo group failed to respond, or that any individual’s course can be predicted from the group comparison. It does not establish that all components of the composite improved independently. Nor does it measure complete remission: the response definition is a specified combination of score improvement and limits on worsening.

The finding applies to the study population, intervention comparison, endpoint definition, and 52-week double-blind assessment described in the registry. It does not by itself establish what would happen in populations or settings not represented in the study, how outcomes would differ over longer follow-up, or whether every secondary outcome is reliable. The registry record does not report the full analysis plan, missing-data procedures, multiplicity handling, or a clinical-importance threshold.

Safety counts should be interpreted separately from efficacy results. They describe recorded adverse events in the specified groups and phases; they do not establish causality or resolve the balance of benefits and harms. Taken together, the posted results are statistical evidence about defined group-level outcomes, not a guarantee of individual outcome or a complete account of every aspect of SLE.

18. Why This Trial Matters Statistically

This trial illustrates why it is useful to distinguish an endpoint’s definition from its statistical comparison. The primary outcome is a composite response, not a single continuous score or an event count. Its meaning depends on the improvement threshold and the rules for worsening on PGA and BILAG. A reader needs both the response percentages and the precise definition to understand what the comparison represents.

It also demonstrates the value of reporting effect estimates with confidence intervals, rather than relying on a p-value alone. The risk difference communicates an absolute comparison of response proportions, while the sponsor’s odds ratio uses a relative odds scale. Reading the two as different estimands avoids treating them as competing versions of a single quantity. The reported interval helps show precision, while leaving questions of clinical importance and bias separate.

The difference between the registry’s logistic-regression analysis and the unadjusted recalculation is another useful statistical distinction. Summary counts permit a transparent unadjusted comparison, but cannot reproduce every possible feature of a model-based analysis. The registry’s limited model details make it important not to infer adjustments that are not documented.

Finally, multiple secondary outcomes, varying outcome populations, and separate double-blind and open-label phases show why analysis populations, missing-data procedures, and multiplicity plans matter. The posted secondary pattern can add context, but it does not remove the need for prespecified interpretation and complete reporting. These are general lessons for reading clinical trial results, not conclusions beyond the registry record.

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 registry results. The primary analysis excludes participants from site 086485, but the reason for that exclusion and the number of participants affected are not reported in the registry record. The outcome population is defined as randomized participants who received at least one dose, rather than all randomized participants, and the consequences of that choice cannot be assessed without further participant disposition information.

The counts included in the posted outcomes differ across endpoints. Reasons for the variation, including any missing assessments or eligibility differences, are not reported in the registry record. The methods used to handle missing Week 52 measurements and censored time-to-event data are also not reported. This limits assessment of whether the estimates could be sensitive to incomplete follow-up.

The registry identifies logistic regression for the primary analysis but does not report its full specification. Covariates, model diagnostics, and any adjustments are not reported in the registry record. The analysis plan, multiplicity strategy, interim analyses, and original sample-size justification are also not reported. These gaps restrict evaluation of prespecification and robustness.

The study concerns a Northeast Asian SLE population and a 52-week double-blind comparison. The registry record alone does not establish how well the results generalize beyond the studied population, setting, intervention regimen, or assessment period. The open-label results have a different design context and cannot serve as a continued randomized placebo comparison. Finally, safety counts are descriptive and do not provide causal attribution. These limitations call for careful qualification of the findings, not a claim that the observed results are invalid.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: GSK1550188 A 52 Week Study of Belimumab Versus Placebo in the Treatment of Subjects With Systemic Lupus Erythematosus (SLE) Located in Northeast Asia — NCT01345253.
• Registry record last updated 2019-10-04; status last verified 2019-09.
• Results first posted 2016-12-02.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

The registry record describes a completed, randomized, quadruple-masked, parallel-group Phase 3 study of belimumab 10 mg/kg versus placebo in participants with SLE in Northeast Asia. Enrollment is listed as 709. The primary endpoint is SRI response at Week 52 of the double-blind phase, assessed in a Modified Intention-to-Treat population that excluded participants from site 086485.

The posted response percentages were 40.1 for placebo and 53.8 for belimumab. The unadjusted recalculated risk difference was 0.137, with a 95% confidence interval from 0.057 to 0.217; the registry-reported logistic-regression odds ratio was 1.99, with a 95% confidence interval from 1.40 to 2.82. Secondary outcomes and adverse event counts are also reported, with distinct phase and population contexts. Details of missing-data handling, multiplicity, and the original design assumptions are not reported in the registry record.