Safety and Immunogenicity Study of a Booster Dose of GSK Biologicals' 10-valent Pneumococcal Conjugate Vaccine.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Immunogenicity Study of a Booster Dose of GSK Biologicals' 10-valent Pneumococcal Conjugate Vaccine. (NCT00370396), a randomized phase 3 study. The primary outcome is Number of Subjects With Rectal Temperature Above (>) 39.0 Degrees Celsius (°C) Post Booster Between the Synflorix-Synflorix and Prevenar-Prevenar Groups.

2. Clinical Question

The question is about the frequency of a defined post-booster fever event, not about average temperature or antibody response. The registry also lists safety, reactogenicity, antibody persistence, seroprotection and functional antibody measures among the study outcomes. Those outcomes broaden the study’s scope, but the primary result is the fever count in the two named groups. The record does not give a primary-result count for the Prevenar-primed, Synflorix-boosted group.

3. Trial Design

This was a randomized, parallel-group phase 3 study sponsored by GlaxoSmithKline and marked completed in the registry. The recorded enrollment was 1200. The study began on 2006-09-25, with primary completion on 2007-06-04. Children aged 12–18 months had previously received a 3-dose primary vaccination course in study 105553. The trial assessed a subsequent booster dose and antibody persistence after the earlier course.

Three groups were described. Children previously primed with Synflorix received Synflorix as their booster. Children previously primed with Prevenar received either Prevenar or Synflorix as their booster. The primary comparison uses the first two groups: Synflorix followed by Synflorix versus Prevenar followed by Prevenar. The Prevenar-primed, Synflorix-boosted group is part of the study design but is not included in that primary comparison. All children were also scheduled to receive a concomitant booster dose of DTPa-HBV-IPV/Hib.

The endpoint counts show 735 participants in the Synflorix-Synflorix group and 91 in the Prevenar-Prevenar group. These are the denominators posted for the primary result, not the total randomized enrollment. The registry record does not report the original allocation targets for each group or explain the difference between the recorded enrollment and these endpoint denominators. It also does not report the allocation sequence or concealment procedures.

The study included outcomes covering solicited symptoms, unsolicited adverse events, serious adverse events and immune responses. The listed immune measures include antibody concentrations, seroprotection and opsonophagocytic activity for pneumococcal serotypes. A full protocol, detailed statistical analysis plan and complete explanation of how participants entered each analysis population 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: During the screening the following was performed: informed consent was obtained and signed from parents or guardians of subjects, check for inclusion/exclusion criteria and contraindications/precautions was performed, and medical history of subjects was collected. Prior to vaccination, subjects' pre-vaccination body temperature was evaluated.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The analysis was performed on the Total vaccinated cohort, which included all vaccinated subjects, who completed their symptom sheets.

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 Rectal Temperature Above (>) 39.0 Degrees Celsius (°C) Post Booster Between the Synflorix-Synflorix and Prevenar-Prevenar Groups, measured over Within 4 days (Days 0-3) after the booster vaccination at Month 0 in this study 10PN-PD-DIT-007.

The primary endpoint counts participants whose rectal temperature was above 39.0 °C within 4 days, Days 0–3, after the booster vaccination at Month 0. The result is a count of participants with the event, rather than a continuous temperature summary. The registry describes the analysis population as vaccinated participants with results available and also says the total vaccinated cohort included participants who completed their symptom sheets.

A threshold-based fever outcome gives a clinically interpretable way to describe one aspect of reactogenicity: whether a child crossed a prespecified temperature level during the stated observation window. It does not describe how high temperatures were among participants who had fever, how long fever lasted, or symptoms outside that window. Those details are not reported as part of this primary outcome.

The posted count was 24 among 735 participants in the Synflorix-Synflorix group and 7 among 91 in the Prevenar-Prevenar group. Counts alone do not convey how uncertain the comparison is, especially when the group sizes differ. The statistical comparison and its confidence interval add information about the range of effect estimates compatible with the posted data, subject to the assumptions of the method.

7. Results Posted in the Registry

• Prevenar-Prevenar Group: 7.7% (7/91) versus 3.3% (24/735) on control; risk difference 4.4 percentage points (95% CI -1.2 to 10.1), risk ratio 2.36 (1.04–5.31); unadjusted two-proportion z-test p = 0.036.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary at the participant level: each participant either met the stated fever threshold during the specified observation window or did not. For this kind of outcome, a count can be expressed relative to the number of participants assessed. Comparing those group proportions describes the difference in event frequency, while a confidence interval conveys uncertainty around the estimated contrast. These are general statistical concepts; the registry’s analysis method is not fully specified.

What does the risk difference describe?

The recalculation reports the risk difference as Prevenar-Prevenar minus Synflorix-Synflorix. In general, a risk difference is the difference between the two group event proportions, expressed on the probability scale. A positive value under this ordering means the estimated event proportion is higher in the first named group; a negative value means it is lower. The posted estimate is 0.044. The confidence interval from -0.012 to 0.101 includes zero, which represents equal event proportions under this contrast. The interval also includes values in either direction, so the estimate is not precise enough to rule out all of those possibilities.

The registry-reported analysis instead states a difference in percentage of -4.43 for the groups listed as Synflorix-Synflorix and Prevenar-Prevenar. This is the reverse group ordering from the recalculation. The reported confidence interval is -11.85 to -0.21 at 95%. The registry labels the analysis non-inferiority or equivalence, but gives no method, margin or p-value. Those omissions prevent a full interpretation of the test as a non-inferiority or equivalence procedure.

How should the p-value be read?

The recalculated unadjusted p-value is 0.036. In general, a p-value measures how compatible data at least as discrepant as those observed would be with a specified null model, under the assumptions of the test. It is not the probability that the null hypothesis is true, nor does it measure the size or importance of a difference. Its interpretation depends on the analysis and on how the outcome and comparison were selected. The registry record does not report a p-value for its own primary analysis.

The reported confidence interval and the recalculated interval come from different sources and are attached to different contrasts. The recalculated interval is based on the posted counts using a simple unadjusted calculation. The sponsor’s method is not given, so the difference between these outputs cannot be attributed to a particular statistical choice. A direct comparison requires the original analysis details.

Why does group size matter?

Precision generally depends on the number of observed participants and events in each group. When group sizes differ substantially, the smaller group often contributes more uncertainty to the contrast. Here, the endpoint denominators are 735 and 91, and the event counts are 24 and 7, respectively. A count-based comparison also depends on having the same event definition and observation window applied consistently across groups. The registry describes a common threshold and follow-up window, but does not provide further details about assessment procedures or missing symptom sheets.

For count outcomes, methods may use different approximations or interval constructions. General methods can yield different confidence intervals or p-values, especially when event counts are limited or group sizes are imbalanced. The method used for the recalculation is explicitly unadjusted; the method underlying the registry analysis is not reported. The record does not report whether covariates, stratification or other adjustments were used in the sponsor analysis.

What does non-inferiority or equivalence require?

In general, a non-inferiority analysis evaluates whether an estimate rules out an unacceptable loss relative to a prespecified margin. Equivalence asks whether an interval lies within prespecified bounds in both directions. These are not established solely by a difference estimate or by a conventional test of no difference. The registry labels its reported analysis as non-inferiority or equivalence, but does not report the margin, detailed hypothesis, analysis method or decision rule. The label therefore identifies the analysis category but does not provide enough information to independently assess its criteria.

Randomization is intended, in general, to support comparable groups at baseline, but the public record does not describe how randomization was generated or implemented. Because children entered this study after earlier vaccination in another study, the three groups also reflect prior vaccination history. The record’s brief descriptions identify these histories, but do not report detailed baseline characteristics or adjustments for them. Statistical results must be interpreted within the limits of the described comparison and available analysis details.

10. Confidence Intervals and Precision

The recalculated risk difference is 0.044, with a confidence interval from -0.012 to 0.101. In general, a confidence interval describes a range of values compatible with the data under the method used. It is not a guarantee that the true effect lies in that particular interval, and its coverage depends on the method’s assumptions. The interval includes zero and spans estimates on both sides of it. Thus, the recalculation does not pin down the direction of the risk difference with high precision.

The registry-reported analysis gives a 95% confidence interval from -11.85 to -0.21 for the difference in percentage in the Synflorix-Synflorix minus Prevenar-Prevenar order. This interval does not include zero. It should not be treated as interchangeable with the recalculated interval, which uses the reverse contrast and an unadjusted calculation. The registry does not report the method behind its interval, so the statistical assumptions and reason for the differing interval cannot be assessed from the record alone.

Precision is affected by the number of participants and events, and the primary comparison has unequal group denominators: 735 and 91. The smaller denominator limits the information contributed by that group. The registry does not report a confidence interval for each group’s event proportion or additional precision measures for the counts. The reported intervals quantify uncertainty in their respective contrasts; they do not address missing outcome data, unmeasured differences between groups or multiplicity across outcomes.

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 sample-size display is a planning illustration based on the observed effect, not the trial’s original design calculation. For this binary endpoint, the entries are participants per arm, not total events. The calculations assume a two-sided significance level of 0.05, equal allocation between the two arms and a normal approximation. They use an effect equal to the stated fraction of the observed effect. The registry’s own original sample-size assumptions are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm for a design intended to have an 80% chance of detecting the selected effect, under the listed assumptions. The 90% power column describes the corresponding planning target for a 90% chance. These are design calculations for hypothetical studies, not evidence about the probability that the observed result is true or reproducible. This is not observed power.

The rows consider planning for 100% of the observed effect, 75% of the observed effect and 50% of the observed effect. The displayed participant requirements rise as the target effect becomes smaller: detecting a less pronounced difference generally requires more information and therefore more participants under the same assumptions. These figures should be read as conditional illustrations, not as a recommended enrollment target or a reconstruction of the trial’s actual planning process.

The planning illustration assumes equal allocation, whereas the posted primary endpoint denominators are 735 and 91. The calculation also uses a normal approximation, and the registry does not report whether the original design used that approach. The assumptions make the entries easier to interpret, but they do not establish that the actual study was designed or analyzed in this way.

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 = 0.033; treatment proportion p₁ = 0.077; average p̄ = 0.055.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.3218 = 0.6308.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.3203 = 0.2696.
4. n = (0.6308 + 0.2696)² / (0.0443)² = 413.6, rounded up to 414 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 <- 0.0326530612244898                          # control proportion
p1 <- 0.0769230769230769                          # 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(370396)
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 = 414 (80% power), 554 (90% power)
 75% of observed effect: n per arm = 666 (80% power), 891 (90% power)
 50% of observed effect: n per arm = 1,339 (80% power), 1,792 (90% power)
power.prop.test: 413.6 
Simulated power at n = 414 per arm: 0.809 (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. Analyses Reported by the Sponsor

The registry reports a sponsor analysis as a difference in percentage of -4.43, with a 95% confidence interval from -11.85 to -0.21. Its group order is Synflorix-Synflorix minus Prevenar-Prevenar, and the analysis is labelled non-inferiority or equivalence. The method, margin, test details and p-value are not reported in the registry record.

The simple recalculation uses the posted counts and expresses the contrast in the opposite direction, Prevenar-Prevenar minus Synflorix-Synflorix. Its risk difference is 0.044, with a confidence interval from -0.012 to 0.101 and an unadjusted p-value of 0.036. Reversing the group order reverses the sign of a difference. However, the confidence intervals are not simply sign reversals of each other, so the registry analysis cannot be reproduced by changing the sign of the recalculated result. They are distinct analyses, and the sponsor’s underlying method is not documented in the available record.

The sponsor result is the registry’s reported analysis; the recalculation is an independent, unadjusted check from summary counts. The latter does not replace the sponsor analysis or reveal how it was conducted. A full reconciliation would require the original analysis specification and data handling details, which are not reported in the registry record.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists 16 outcome measures and 1 analysis for the primary outcome. It also lists secondary outcomes covering solicited local and general symptoms, unsolicited adverse events, serious adverse events, seroprotection, antibody concentrations and opsonophagocytic activity. The registry record does not report a multiplicity plan, such as which outcomes were treated as confirmatory, whether testing was adjusted across outcomes, or how the primary comparison related to the other analyses.

In general, when many outcomes or comparisons are examined, the chance of at least one apparently unusual result can increase unless the analysis plan accounts for that multiplicity. This does not mean that every listed outcome was tested in the same way. The registry reports only one analysis for the primary outcome, and the full analysis plan is not available in the record. The unadjusted recalculated p-value should therefore be interpreted as a result for that comparison, not as a multiplicity-adjusted finding across the study.

The registry describes the primary analysis as involving the total vaccinated cohort, including vaccinated participants with available results, and says the population included participants who completed symptom sheets. It does not report the number excluded for missing symptom sheets, how missing temperature data were handled, whether any imputation was used, or whether sensitivity analyses were conducted. Consequently, the possible effect of missing outcome information cannot be assessed from the posted details.

The registry record does not report a full analysis-population definition beyond the stated vaccinated cohort description. It does not provide a detailed intention-to-treat analysis, per-protocol population, or participant flow for the primary outcome. The reported endpoint denominators are available, but reasons for differences between those denominators and recorded enrollment are not explained. These omissions limit assessment of whether the analyzed groups correspond exactly to all randomized participants.

15. 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 4% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: bronchitis (4.3%), laryngitis (2.2%), bronchitis chronic (2.2%), gastroenteritis (1.2%), diarrhoea (1.1%).

The adverse-event summary gives serious adverse event counts and numbers at risk for each of the three groups. The counts are 33 of 737 in Synflorix-Synflorix, 6 of 92 in Prevenar-Prevenar and 8 of 93 in Prevenar-Synflorix. The summary also reports zero deaths in each group. These are descriptive counts; they do not by themselves establish that an event was caused by a vaccine or allow a direct causal comparison.

The same summary gives other adverse event counts of 684 of 737, 85 of 92 and 253 of 283, respectively. The registry separately lists serious adverse events during the active phase and during the entire study as outcomes, but the summary does not provide the timing breakdown for the serious-event counts. The event categories, collection procedures and detailed follow-up definitions should therefore be interpreted according to the registry’s full outcome descriptions, where available.

The listed serious-event terms include bronchitis, laryngitis, chronic bronchitis, gastroenteritis, diarrhoea and bronchiolitis, with percentages reported for those terms. These data are descriptive safety information. Differences in group size and follow-up, and the absence of detailed event-level information, limit comparisons based only on aggregate counts.

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

The primary result describes the number of participants who crossed a specified rectal temperature threshold during a short post-booster window. It does not describe all dimensions of reactogenicity, such as symptom duration or severity beyond the threshold definition, and it does not summarize every safety outcome in the study. The counts apply to the two groups included in the primary comparison, not to the third group.

The posted data show 24 events among 735 participants in the Synflorix-Synflorix group and 7 among 91 participants in the Prevenar-Prevenar group. The registry’s analysis reports a percentage difference and confidence interval for one contrast order; a simple recalculation gives a risk difference and interval in the reverse order. The recalculated interval includes zero, while the registry-reported interval does not. Because the registry method and non-inferiority or equivalence margin are not reported, these two statistical summaries cannot be fully reconciled from the public record.

A confidence interval describes statistical uncertainty under a particular method, but it does not resolve uncertainties arising from missing data, outcome measurement, design conduct or unreported analysis choices. The unequal endpoint group sizes also mean the comparison is informed by substantially different amounts of data in the two groups. The result should therefore be understood with its group definitions, follow-up window and analysis source in view.

The trial also assessed immune responses, antibody persistence and multiple safety outcomes, but the registry entries do not include their detailed results. The primary fever result cannot answer those other questions. It also cannot establish that one booster strategy is generally preferable, or support a broad safety conclusion on its own. The result is a statistical description of a particular endpoint in the posted comparison, not medical advice.

17. Why This Trial Matters Statistically

This trial raises several useful statistical issues. First, the endpoint is a count of participants meeting a threshold, so a comparison depends on both the event counts and the denominators. Reporting only the counts would obscure the differing group sizes; reporting only a contrast would make it harder to understand how many participants contributed the events. Reading the counts and contrast together gives a more complete account of the primary result.

Second, the two primary groups are unequal in size. The smaller group contributes less information to the comparison, which is relevant when interpreting the confidence interval and considering how much precision the estimate has. A statistically reported contrast should be read alongside the underlying group counts and the interval, rather than relying only on a p-value.

Third, the registry contains a sponsor-reported analysis labelled non-inferiority or equivalence, as well as a recalculation from the posted summary counts. The group ordering and methods matter: a contrast reverses sign when the groups are reversed, and an unadjusted recalculation is not a substitute for a specified analysis. The absence of a reported margin and method illustrates why a label alone may not be enough to evaluate a non-inferiority or equivalence claim.

Finally, the primary result is one part of a study with many listed outcomes and three groups. Multiplicity, analysis populations and missing symptom-sheet data can affect interpretation, but the registry does not report enough detail to evaluate those issues fully. Distinguishing what the posted data quantify from what the record leaves unspecified is central to a careful reading of clinical trial statistics.

18. 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 public record does not report the detailed randomization procedure, allocation concealment, baseline characteristics or the original group allocation targets. The endpoint denominators are 735 and 91, while recorded enrollment is 1200; the reasons for this difference are not stated. Participant flow and reasons for exclusions from the primary analysis are not reported.

The primary analysis population is described as vaccinated participants with available results and as participants who completed symptom sheets. The number with missing symptom sheets, the handling of missing temperature measurements and any sensitivity analyses are not reported. These omissions limit assessment of how missing outcome information might affect the fever comparison.

The registry labels its reported primary analysis as non-inferiority or equivalence but does not report the method, margin, detailed hypothesis or decision rule. The sponsor analysis confidence interval differs from the interval in the simple unadjusted recalculation, and the available record does not provide sufficient methods to explain that difference. The original sample-size rationale and trial-specific planning assumptions are also not reported.

The primary result covers only the Synflorix-Synflorix and Prevenar-Prevenar groups. No primary endpoint count for the Prevenar-Synflorix group is available in the stated results. Although the registry lists numerous secondary outcomes, their detailed results are not included in the registry record for this analysis. The record also does not report how multiplicity was addressed or provide full event definitions and follow-up details for every safety summary.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Safety and Immunogenicity Study of a Booster Dose of GSK Biologicals' 10-valent Pneumococcal Conjugate Vaccine. — NCT00370396.
• Registry record last updated 2019-06-10; status last verified 2019-06.
• Results first posted 2018-11-26.
• 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.

22. Record Summary

NCT00370396 was a completed, randomized, parallel-group phase 3 study sponsored by GlaxoSmithKline, with recorded enrollment of 1200. It assessed booster vaccination in children aged 12–18 months previously primed with pneumococcal vaccine in study 105553. Three groups were described: Synflorix followed by Synflorix, Prevenar followed by Prevenar, and Prevenar followed by Synflorix.

The primary outcome was the number of participants with rectal temperature above 39.0 °C within 4 days after booster vaccination. Posted counts were 24 of 735 in the Synflorix-Synflorix group and 7 of 91 in the Prevenar-Prevenar group. The registry reports a non-inferiority or equivalence analysis, while a separate unadjusted recalculation from the posted counts gives a risk difference in the reverse contrast direction. The registry does not report enough methodological detail to fully reconcile the analyses or assess all missing-data and multiplicity questions.