Study to Evaluate Efficacy, Safety, and Immunogenicity of GlaxoSmithKline (GSK) Biologicals' Herpes Zoster Vaccine GSK1437173A: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study to Evaluate Efficacy, Safety, and Immunogenicity of GlaxoSmithKline (GSK) Biologicals' Herpes Zoster Vaccine GSK1437173A (NCT01610414), a randomized phase 3 study. The primary outcome is Number of Subjects With Confirmed Herpes Zoster (HZ) Episode.

2. Clinical Question

The registry describes a Phase 3 randomized, triple-masked, parallel-group study intended to evaluate vaccine efficacy, safety, and immunogenicity. The clinical question is about prevention of confirmed herpes zoster in a transplant-recipient population, rather than prevention in all adults or in people with other transplant types. Confirmation could be based on polymerase chain reaction or review by the HZ Ascertainment Committee, which consisted of physicians with herpes zoster expertise. The results therefore concern the registry's specified case definition and analysis population; they should not be generalized beyond those boundaries without further evidence.

3. Trial Design

This was a Phase 3 randomized, triple-masked, parallel-group trial comparing GSK1437173A with placebo. The registry lists an enrollment of 1877. The study was sponsored by GlaxoSmithKline and is recorded as completed. Its listed start date is 2012-07-13, and its primary completion date is 2016-11-04.

The vaccine and placebo were each given in two doses on a 0, 1 Month schedule. Both were administered intramuscularly in the deltoid region of the non-dominant arm. The record calls the study observer-blind and lists triple masking. It does not describe the specific masking procedures or identify which roles were masked, so those details are not reported in the registry record.

The primary endpoint was the number of participants with a confirmed herpes zoster episode. Its timeframe ran from Month 0 until the cut-off date for final analysis, with median follow-up of 21 months. The primary analysis population was a modified Total Vaccinated Cohort. The record says this excluded participants who did not receive the second vaccination, received vaccine doses or replacement not on the same group, or developed a confirmed herpes zoster case before the specified analysis point.

Other assessed outcomes included herpes zoster-associated complications, postherpetic neuralgia, pain, antibody concentrations, and adverse events. The registry record does not report the original sample-size justification, the planned allocation ratio, the sequence generation or concealment procedures, or the details of any interim analyses. Those features should not be inferred from the results or from the planning illustrations.

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: Some of the enrolled subjects were allocated subject numbers, but were not administered all doses of study vaccine, hence they did not start the study.

Study Start - Month 13

Reasons for not completing this period:

Month 13 - Month 25

Reasons for not completing this period:

After Month 25

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The analysis was performed on the modified Total Vaccinated Cohort, which excluded subjects from the Total Vaccinated Cohort analysis who were not administered with the second vaccination or who received vaccine doses/or replacement not on the same group or who developed a confirmed case of HZ prior to 1 month after the second vaccination.

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 Subjects With Confirmed Herpes Zoster (HZ) Episode, measured over From Month 0 until the cut-off date for final analysis (median follow-up was of 21 months).

The primary endpoint counts participants with a confirmed herpes zoster episode. It is a binary outcome at the participant level: each participant either met the confirmation definition during the stated follow-up period or did not. Suspected cases could be identified by a characteristic new rash, a vesicular rash suggestive of varicella-zoster virus infection, or a clinical presentation and laboratory findings suggestive of infection even without a characteristic rash. A suspected case was confirmed by polymerase chain reaction or by the HZ Ascertainment Committee.

Counting participants with an event is useful because it directly describes how often the event occurred in each study group over the stated observation period. It can be summarized as an event count and as a risk, and group comparisons can quantify the difference or relative contrast between those risks. The interpretation depends on the case definition, follow-up period, and analysis population: these define which events count and which participants contribute.

For this endpoint, the registry reports the modified Total Vaccinated Cohort and median follow-up of 21 months. The analysis therefore does not represent every enrolled participant automatically; its stated exclusions matter. A participant count also does not describe the duration or severity of each episode. Those are different clinical features, addressed separately among the listed secondary outcomes. The registry record does not report further endpoint adjudication details beyond the confirmation routes described.

7. Results Posted in the Registry

• GSK1437173A Group: 5.6% (49/870) versus 15.9% (135/851) on control; risk difference -10.2 percentage points (95% CI -13.1 to -7.3), risk ratio 0.36 (0.26–0.49); 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.

Number of Subjects With Confirmed HZ-associated Complications

• 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:
  • GSK1437173A Group: 0.3% (3/870) versus 1.5% (13/851) on control; risk difference -1.2 percentage points (95% CI -2.1 to -0.3), risk ratio 0.23 (0.06–0.79); unadjusted two-proportion z-test p = 0.011.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for GSK1437173A Group than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Number of Subjects With Postherpetic Neuralgia (PHN)

• 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:
  • GSK1437173A Group: 0.1% (1/870) versus 1.1% (9/851) on control; risk difference -0.9 percentage points (95% CI -1.7 to -0.2), risk ratio 0.11 (0.01–0.86); unadjusted two-proportion z-test p = 0.010.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for GSK1437173A Group than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Number of Subjects With Any Relapse

• 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:
  • GSK1437173A Group: 25.9% (239/922) versus 27.4% (253/924) on control; risk difference -1.5 percentage points (95% CI -5.5 to 2.6), risk ratio 0.95 (0.81–1.10); unadjusted two-proportion z-test p = 0.478.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between GSK1437173A Group and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover herpes zoster complications and postherpetic neuralgia, as well as pain, relapse, antibody concentrations, and safety measures. Across the secondary comparisons with posted estimates, the direction for herpes zoster-associated complications and postherpetic neuralgia aligns with the primary result, while the relapse comparison is less conclusive. This broad pattern is compatible with the primary finding, but the secondary outcomes do not independently establish the primary result or resolve its limitations.

Secondary endpoints may differ in how they are defined, which participants contribute, and how often events occur. Some are clinical events, some are symptom measures, and others concern immune response or safety. A consistent direction across related outcomes can add context, but it should not be treated as a substitute for a prespecified primary analysis. In particular, the registry record does not report a plan for controlling the chance of false-positive findings across multiple secondary comparisons.

Unadjusted secondary p-values and intervals carry less weight when many outcomes are examined because each additional comparison creates another opportunity for an apparently notable result by chance. This is a general statistical concern, not evidence that any particular secondary result is false. The record lists 11 posted secondary outcomes and reports one primary analysis. The registry record does not explain whether the secondary outcomes were formally prioritized or whether a multiplicity strategy was used. For that reason, secondary findings are best treated as supportive or exploratory context rather than as equally definitive evidence.

The endpoint-specific counts and estimates should be interpreted on their own terms, particularly where event counts are small or intervals are broad. The registry record does not report sponsor analyses for the listed secondary comparisons, so the unadjusted calculations should not be mistaken for the sponsor's own inferential conclusions.

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

• Duration of 'Worst' HZ-associated Pain
• Number of Subjects With Confirmed HZ-associated Complications
• Number of Subjects With Postherpetic Neuralgia (PHN)
• Antigen-glycoprotein E (gE) Antibody Concentrations in a Sub-cohort of Subjects
• Number of Subjects With Any and Grade 3 Solicited Local Symptoms
• Number of Subjects With Any, Grade 3 and Related Solicited General Symptoms
• Number of Subjects With Any, Grade 3 and Related Unsolicited Adverse Events (AEs)
• Number of Subjects With Any and Related Potential Immune Mediated Diseases (pIMDs)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is a binary participant-level outcome: a confirmed herpes zoster episode was either recorded or was not. For this kind of endpoint, the group event counts can be converted to risks using the number of participants in each group as the denominator. A risk difference expresses the absolute contrast between those risks. The posted unadjusted estimate is -0.102, with a 95% confidence interval from -0.131 to -0.073. Negative values indicate a lower observed risk in the vaccine group under the comparison's ordering. These calculations do not adjust for other characteristics.

A confidence interval describes a range of values compatible with the observed data under the statistical method used. It reflects uncertainty in an estimate; it is not a range containing a fixed proportion of individual participant outcomes. For a risk difference, zero corresponds to no difference in risks. The posted interval excludes zero. A p-value addresses how unusual data at least as discrepant as those observed would be under a specified null model. The unadjusted p-value for the recalculated risk comparison was <0.001. It does not measure the probability that the null hypothesis is true, and it does not describe the size or importance of an effect.

The registry also reports a sponsor analysis using a Poisson method, expressed as vaccine efficacy. Its estimate was 68.17, with a 95% confidence interval from 55.56 to 77.53 and p < 0.001. Vaccine efficacy and risk difference use different scales and methods. The registry record identifies the method and result but does not report enough detail to reconstruct the full model, such as additional covariates or any further assumptions. The sponsor's result should therefore be presented as reported, not reverse-engineered from the unadjusted risk comparison.

Randomization is intended to create groups that are comparable on average at the start of a trial, including with respect to factors that may not be measured. Masking can reduce the risk that expectations influence participant care, event assessment, or reporting. These are general purposes of the design features. The record labels this study randomized and triple-masked but does not describe the implementation details, so the effectiveness of particular procedures cannot be assessed from the registry entry alone.

What does the risk difference communicate?

As a general statistical concept, an absolute difference communicates how much the observed event risk differs between groups on the risk scale. It is often readily interpretable because it retains the same scale as the event probability. The estimate must still be read with its confidence interval, study population, endpoint definition, and follow-up period. It does not by itself explain why a difference occurred or whether it applies to another clinical setting.

Why might methods for binary outcomes differ?

General statistical methods can model binary outcomes in different ways and can target different effect measures. A direct comparison of risks yields an absolute risk difference, while other approaches can yield relative measures or model-based estimates. These estimates need not match numerically because they answer different questions and may rely on different modeling assumptions. The sponsor's Poisson-method vaccine-efficacy estimate and the unadjusted risk difference should therefore be kept distinct. The registry record does not report enough methodological detail to characterize every assumption behind the sponsor analysis.

What does statistical significance establish?

In general, a small p-value indicates that the observed data are not readily explained by a particular null model, given the analysis assumptions. It does not establish that a result is free of bias, clinically important, or applicable to other populations. It also does not incorporate all uncertainties arising from missing information, endpoint ascertainment, or multiple comparisons unless the analysis explicitly accounts for them. The registry record does not report a multiplicity adjustment plan for the secondary outcomes.

How should the analysis population be considered?

The primary analysis used a modified Total Vaccinated Cohort. The registry says this population excluded participants who did not receive the second vaccination, received vaccine doses or replacement not on the same group, or developed a confirmed herpes zoster case before the specified analysis point. Excluding participants based on events or treatment received can affect the population represented by the estimate. The registry record does not give enough detail to evaluate the consequences of each exclusion or how missing endpoint information was handled. The result should be understood as applying to the stated analysis population, rather than automatically to all enrolled participants.

Why can over-dispersion matter?

As a general concept, over-dispersion occurs when the variability in observed count data is greater than a statistical model expects. It can make uncertainty estimates too narrow if the model does not allow for that extra variability. This endpoint is reported as a count of participants with a binary event, and no dispersion estimate is listed for it. The registry record does not report an over-dispersion assessment or identify over-dispersion as a concern in the primary analysis. This general issue should not be taken as evidence that the study's analysis had such a problem.

11. Confidence Intervals and Precision

The unadjusted risk difference for confirmed herpes zoster was -0.102, with a 95% confidence interval from -0.131 to -0.073. This interval represents uncertainty around the estimated group contrast under the unadjusted calculation. It excludes zero, the value corresponding to no risk difference, so the observed data are not consistent with zero difference under this interval method at the stated confidence level.

The sponsor's Poisson-method vaccine-efficacy estimate was 68.17, with a 95% confidence interval from 55.56 to 77.53. That interval describes precision for the sponsor's effect measure and method, not the unadjusted absolute risk difference. The two intervals should not be directly compared as if they shared a scale. The sponsor analysis reported p < 0.001, while the unadjusted risk comparison reported p <0.001.

In general, a confidence interval reflects both the size of the observed estimate and its statistical uncertainty. It does not capture every possible source of uncertainty, such as limitations in the analysis population, outcome ascertainment, or unreported design and analysis details. Nor does it mean that individual outcomes fall within the interval. The registry record reports the primary results but does not report additional sensitivity analyses that would show how estimates change under alternative assumptions.

Precision is also endpoint-specific. The secondary outcomes have their own event counts and intervals, and should not inherit the primary endpoint's degree of precision. Where fewer participants have an event, estimates can be more uncertain; interpretation should rely on each endpoint's own interval rather than the apparent size of its point estimate alone.

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 illustrations give participants per arm for this binary participant-count outcome. They use two-sided significance level 0.05, equal allocation between the two arms, and a normal approximation. The effect targeted in each illustration is the stated fraction of the observed effect. These are planning calculations based on the observed effect, not the trial's original design justification. The trial's own sample-size assumptions are not reported in the registry record.

The 80% power column gives the estimated number of participants per arm under those assumptions for a design with an 80% chance of detecting the specified effect if that effect were the true one. The 90% power column gives the corresponding planning number for a 90% chance. Power is a design property under assumed conditions; it does not indicate the probability that the observed result is true. The illustrations are not observed power.

Rows describing 100%, 75%, and 50% of the observed effect show how planning requirements change when the target is smaller than the observed estimate. Smaller effects are harder to distinguish from random variation, so the illustrations require more participants per arm as the target effect becomes smaller. The actual numbers vary by endpoint because event frequencies and estimated effects differ. These illustrations should be read as conditional planning examples, not as proof that a future study will reproduce the observed result.

The registry record does not report whether the original trial design used these assumptions, whether it included allowance for attrition, or what target effect was used at design. The planning figures therefore cannot be used to reconstruct the actual rationale for the trial's enrollment.

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 = 0.159; treatment proportion p₁ = 0.056; average p̄ = 0.107.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4380 = 0.8585.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4320 = 0.3636.
4. n = (0.8585 + 0.3636)² / (-0.1023)² = 142.7, rounded up to 143 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 <- 0.158636897767333                          # control proportion
p1 <- 0.0563218390804598                          # 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(1610414)
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 = 143 (80% power), 191 (90% power)
 75% of observed effect: n per arm = 281 (80% power), 376 (90% power)
 50% of observed effect: n per arm = 691 (80% power), 925 (90% power)
power.prop.test: 142.7 
Simulated power at n = 143 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 sponsor's own primary analysis used a Poisson method and reported vaccine efficacy of 68.17, with a 95% confidence interval from 55.56 to 77.53 and p < 0.001. The separate unadjusted calculation from the posted group counts expresses the comparison as a risk difference of -0.102, with a 95% confidence interval from -0.131 to -0.073 and p <0.001.

These analyses use different effect measures and methods, so their numerical values should not be expected to match. The sponsor result is the registry's reported analysis; the risk difference is a simple unadjusted recalculation from the posted counts. The recalculation does not replace the sponsor's analysis or establish that it uses the same model assumptions. The registry record does not report enough detail to reproduce the sponsor's full statistical procedure.

There is no reported discrepancy between the recalculation and sponsor result. The results are directionally consistent: both indicate fewer primary endpoint events in the vaccine group than in the placebo group. The sponsor's analysis should be cited as the trial's own reported analysis, while the unadjusted estimate can help describe the absolute difference in the posted counts.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several statistical comparisons are made. As a general principle, testing more outcomes can increase the chance that at least one appears statistically notable even when there is no underlying effect. Researchers may address this by prespecifying a hierarchy, adjusting significance thresholds, or describing secondary findings as exploratory. The registry record lists multiple secondary outcomes but does not report a multiplicity adjustment plan, a testing hierarchy, or how the interpretation of secondary p-values was intended to account for multiple comparisons.

The primary endpoint is distinct from the collection of secondary outcomes. The record reports one primary analysis, using the modified Total Vaccinated Cohort. The existence of a primary endpoint does not itself show how secondary testing was handled, and no unreported procedure should be assumed. Secondary findings should therefore be interpreted with caution and in the context of their endpoint definitions and precision.

Missing-data handling is another important part of an analysis plan. The registry record does not report a detailed method for handling missing primary outcome assessments, loss to follow-up, or missing measurements for secondary outcomes. It also does not report sensitivity analyses for alternative missing-data assumptions. The described modified Total Vaccinated Cohort exclusions are not a complete account of missing-data management; eligibility exclusions and handling of unavailable outcomes are different issues.

The record describes exclusions from the modified Total Vaccinated Cohort, including participants who did not receive the second vaccination, received vaccine doses or replacement not on the same group, or developed a confirmed herpes zoster case before the stated analysis point. It does not report enough detail to assess how those exclusions affected the estimate, whether all randomized participants were analyzed in other populations, or whether an intention-to-treat analysis was performed. The original statistical analysis plan, including interim analyses and any prespecified missing-data procedures, is not reported in the registry record.

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: plasma cell myeloma (9.8%), pneumonia (6.0%), sepsis (1.6%), febrile neutropenia (1.5%), diffuse large b-cell lymphoma (1.4%).

The safety record gives serious adverse event figures for groups of 922 participants in the GSK1437173A group and 924 in the placebo group. Serious adverse events affected 329 participants in the vaccine group and 310 in the placebo group. The record also lists other adverse events affecting 813 and 485 participants, respectively, and deaths of 118 and 124. These are posted counts; they should be read with the relevant safety populations and follow-up, not as proof of causation.

The listed serious-event terms include Plasma cell myeloma at 9.8%, Pneumonia at 6.0%, Sepsis at 1.6%, Febrile neutropenia at 1.5%, Diffuse large b-cell lymphoma at 1.4%, and Lymphoma at 1.4%. The record does not provide the denominators or additional context needed to interpret these percentages independently, and they should not be converted into group-specific risks without that information.

Counts of serious events and deaths do not establish whether an event was caused by the vaccine, placebo, transplant, underlying illness, or another factor. The registry record does not report a causal assessment or detailed comparative analysis for these events. These figures describe what was recorded and should not be used alone to draw conclusions about comparative safety.

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

The primary result describes fewer participants with confirmed herpes zoster in the GSK1437173A group than in the placebo group in the modified Total Vaccinated Cohort. The sponsor's Poisson-method analysis reported vaccine efficacy of 68.17, with a 95% confidence interval from 55.56 to 77.53. The unadjusted risk difference was -0.102, with a 95% confidence interval from -0.131 to -0.073. These are statistical summaries of this study's data for its specified endpoint and follow-up.

The result does not mean that every person receiving the vaccine would avoid herpes zoster, or that an individual outcome can be predicted from a group comparison. It does not establish how the effect would compare in people outside the enrolled transplant-recipient population, under a different vaccination schedule, or over a different follow-up period. The study assessed confirmed episodes under its stated case definition, not every possible herpes zoster-related outcome.

The primary analysis also does not answer every question about duration of pain, complications, postherpetic neuralgia, immune response, relapse, or safety. These were addressed as separate outcomes, with their own limitations. The secondary comparisons should be interpreted cautiously because the registry record does not report whether multiplicity was controlled.

Finally, statistical evidence of a group difference is not the same as a complete account of clinical importance or safety. The registry record does not report detailed information on missing-data methods, the full analysis plan, or causal assessments of adverse events. Those absent details limit what can be concluded from the registry entry alone. No treatment decision follows from these statistical results without broader clinical judgment and evidence.

18. Why This Trial Matters Statistically

This trial illustrates why the analysis population and endpoint definition matter as much as the headline effect estimate. The primary outcome was not simply any reported rash: suspected cases had defined clinical or laboratory criteria, and confirmation depended on polymerase chain reaction or review by an expert committee. The analysis also used a modified Total Vaccinated Cohort rather than automatically including every enrolled participant. These choices shape the question answered by the estimate.

It also demonstrates the value of reporting an effect on more than one scale. The sponsor's analysis presented vaccine efficacy, while the unadjusted calculation presents an absolute risk difference. Each conveys a different aspect of the comparison. Confidence intervals add information about precision that a point estimate or p-value alone cannot provide.

The secondary outcomes highlight the distinction between a coherent pattern and confirmatory evidence. Related outcomes can support interpretation, but unadjusted tests across multiple endpoints require caution, especially when the multiplicity plan is not reported. Safety counts likewise need context: event counts describe recorded experience but do not by themselves attribute cause.

Finally, planning illustrations based on observed effects should not be confused with the trial's actual design rationale. Without the original sample-size assumptions or analysis plan, the registry results cannot answer every methodological question. This separation between reported evidence, recalculation, and unreported design details is central to transparent interpretation.

19. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record does not report the original sample-size justification, the planned allocation ratio, details of randomization implementation, masking procedures, or interim analyses. As a result, the design rationale and some safeguards against bias cannot be assessed from the registry entry alone.

The primary result applies to a modified Total Vaccinated Cohort with specified exclusions, not necessarily all enrolled or randomized participants. The record does not describe the effect of those exclusions, provide a full account of missing outcome data, or report sensitivity analyses. This limits assessment of how robust the result is to analysis-population and missing-data choices.

The sponsor's analysis method is identified as Poisson, but the registry record does not provide enough detail to reconstruct the model or determine whether additional covariates or other adjustments were used. The unadjusted calculation is therefore not a substitute for a complete independent reanalysis. The registry record also does not report a multiplicity strategy for the secondary outcomes, which limits how strongly secondary p-values can be interpreted.

Follow-up for the primary outcome had a median of 21 months. The result consequently pertains to the specified observation period, and the registry record does not establish longer-term effects. The primary endpoint is confirmed herpes zoster, so it does not alone characterize symptom duration, severity, or all complications. Safety counts are reported, but the record does not provide detailed causal attribution or enough context to interpret each serious-event term. These limitations do not negate the reported findings; they define the boundaries of what the registry results can support.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study to Evaluate Efficacy, Safety, and Immunogenicity of GlaxoSmithKline (GSK) Biologicals' Herpes Zoster Vaccine GSK1437173A — NCT01610414.
• Registry record last updated 2018-01-23; status last verified 2017-12.
• Results first posted 2018-01-23.
• 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

NCT01610414 is a completed Phase 3 randomized, triple-masked, parallel-group study sponsored by GlaxoSmithKline. It enrolled 1877 participants and compared two doses of GSK1437173A with two doses of placebo in adults 18 years of age and older receiving autologous haematopoietic cell transplants. The primary endpoint was the number of participants with a confirmed herpes zoster episode.

In the modified Total Vaccinated Cohort, 49 of 870 vaccine-group participants and 135 of 851 placebo-group participants had a confirmed episode. The sponsor reported vaccine efficacy of 68.17, with a 95% confidence interval from 55.56 to 77.53. Median follow-up was 21 months. The registry also reports secondary clinical, immunogenicity, and safety outcomes, with limitations in the available detail on multiplicity, missing data, and analysis procedures.