A Study to Rank Different Dosages of Antigen of GlaxoSmithKline (GSK) Biologicals' Investigational Respiratory Syncytial Virus (RSV) Vaccine (GSK3003891A), Based on Their Immune Response and Safety, When Administered to Healthy Adult Women: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Rank Different Dosages of Antigen of GlaxoSmithKline (GSK) Biologicals' Investigational Respiratory Syncytial Virus (RSV) Vaccine (GSK3003891A), Based on Their Immune Response and Safety, When Administered to Healthy Adult Women (NCT02956837), a randomized phase 2 study. The primary outcome is Number of Subjects With Any Grade 2 and Grade 3 General Adverse Events (AEs) - Solicited and Unsolicited.

2. Clinical Question

3. Trial Design

This was a phase 2, randomized, parallel-group study. The registry classifies masking as triple; the official title describes the study as observer-blind. The record does not explain how those masking descriptions relate to one another or give further details about who was masked.

The study compared three formulations of investigational RSV vaccine with placebo in healthy women. Each vaccine formulation was administered as a single injection on Day 0, at the listed dose of 30 micrograms (µg), 60µg, or 120µg. The control group received a single placebo injection. The registry reports enrollment of 406. The number assigned to each group is available in the endpoint results, but the allocation ratio and details of the randomization process are not reported in the registry record.

The stated purpose was to rank formulations using safety/reactogenicity and immune-response data. Formulations considered promising were intended for further evaluation, including in studies vaccinating pregnant women. That aim describes the study’s development context; it does not establish what later studies found.

The primary safety outcome covered the 7-day (Days 0-6) post-vaccination period. Selected secondary outcomes include unsolicited adverse events during the 30-day (Days 0-29) period, serious adverse events from Day 0 up to study end at Day 360, and medically attended RSV-associated respiratory tract infections over the same longer follow-up. The registry record marks the study as completed. More detailed protocol procedures, including visit schedules beyond the listed timeframes, 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: Out of 406 subjects initially enrolled in the study, 6 subjects had numbers allocated but did not receive any study vaccine dose, hence only 400 subjects were included in the Total Vaccinated Cohort.

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.

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 Any Grade 2 and Grade 3 General Adverse Events (AEs) - Solicited and Unsolicited, measured over During the 7-day (Days 0-6) post-vaccination period.

The primary endpoint counts participants with any Grade 2 or Grade 3 general adverse event, whether solicited or unsolicited, during the 7-day (Days 0-6) post-vaccination period. Solicited general symptoms included fatigue, gastrointestinal symptoms—nausea, vomiting, diarrhea and/or abdominal pain—fever, and headache. The record also defines an unsolicited adverse event as an untoward medical occurrence temporally associated with use of a medicinal product, whether or not considered related, reported in addition to solicited symptoms or arising outside the specified solicited-symptom follow-up period.

Severity is central to this endpoint. Grade 2 indicates symptoms sufficiently uncomfortable to interfere with daily activities, while Grade 3 indicates symptoms that prevented normal activities. The measure therefore counts people who experienced an event at either of these severity levels; it does not count the total number of events or describe their duration. The posted arm values are participant counts, with the number assessed in each arm given alongside them.

This kind of endpoint is used to characterize short-term reactogenicity and safety at a level of severity intended to capture meaningful interference with ordinary activity. It does not measure immune response, and it is not an efficacy endpoint. The primary objective description says the focus was on the three vaccine groups. The registry record does not report a separate prespecified statistical comparison or a decision rule for ranking formulations based on this endpoint.

7. Results Posted in the Registry

• GSK3003891A Vaccine Formulation 1 Group: 32.0% (32/100) versus 26.5% (27/102) on control; risk difference 5.5 percentage points (95% CI -7.0 to 18.1), risk ratio 1.21 (0.79–1.86); unadjusted two-proportion z-test p = 0.388.
• GSK3003891A Vaccine Formulation 2 Group: 30.3% (30/99) versus 26.5% (27/102) on control; risk difference 3.8 percentage points (95% CI -8.6 to 16.3), risk ratio 1.14 (0.74–1.78); unadjusted two-proportion z-test p = 0.547.
• GSK3003891A Vaccine Formulation 3 Group: 24.2% (24/99) versus 26.5% (27/102) on control; risk difference -2.2 percentage points (95% CI -14.3 to 9.8), risk ratio 0.92 (0.57–1.47); unadjusted two-proportion z-test p = 0.717.

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 Any Unsolicited AEs

• 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:
  • GSK3003891A Vaccine Formulation 1 Group: 55.0% (55/100) versus 48.0% (49/102) on control; risk difference 7.0 percentage points (95% CI -6.8 to 20.7), risk ratio 1.14 (0.88–1.50); unadjusted two-proportion z-test p = 0.322.
  • GSK3003891A Vaccine Formulation 2 Group: 52.5% (52/99) versus 48.0% (49/102) on control; risk difference 4.5 percentage points (95% CI -9.3 to 18.3), risk ratio 1.09 (0.83–1.44); unadjusted two-proportion z-test p = 0.525.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between GSK3003891A Vaccine Formulation 1 Group and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between GSK3003891A Vaccine Formulation 2 Group and the control. The interval gives the range of effects that remain compatible with the data.

Number of Subjects With Any SAEs

• 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:
  • GSK3003891A Vaccine Formulation 1 Group: 1.0% (1/100) versus 2.0% (2/102) on control; risk difference -1.0 percentage points (95% CI -4.3 to 2.4), risk ratio 0.51 (0.05–5.54); unadjusted two-proportion z-test p = 0.572.
  • GSK3003891A Vaccine Formulation 2 Group: 3.0% (3/99) versus 2.0% (2/102) on control; risk difference 1.1 percentage points (95% CI -3.2 to 5.4), risk ratio 1.55 (0.26–9.05); unadjusted two-proportion z-test p = 0.626.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between GSK3003891A Vaccine Formulation 1 Group and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between GSK3003891A Vaccine Formulation 2 Group and the control. The interval gives the range of effects that remain compatible with the data.

Number of Subjects With Any Medically Attended (MA) Respiratory Tract Infections (RTIs) Associated With RSV

• 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:
  • GSK3003891A Vaccine Formulation 1 Group: 21.0% (21/100) versus 8.8% (9/102) on control; risk difference 12.2 percentage points (95% CI 2.5 to 21.9), risk ratio 2.38 (1.15–4.94); unadjusted two-proportion z-test p = 0.015.
  • GSK3003891A Vaccine Formulation 2 Group: 11.1% (11/99) versus 8.8% (9/102) on control; risk difference 2.3 percentage points (95% CI -6.0 to 10.6), risk ratio 1.26 (0.55–2.91); unadjusted two-proportion z-test p = 0.588.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for GSK3003891A Vaccine Formulation 1 Group than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between GSK3003891A Vaccine Formulation 2 Group and the control. The interval gives the range of effects that remain compatible with the data.

The selected secondary outcomes span short-term unsolicited adverse events, serious adverse events followed through study end, and medically attended RSV-associated respiratory tract infections. These are distinct outcomes, with different definitions and observation windows. Their observed patterns are not uniform: comparisons vary across formulation groups and outcome types rather than forming a consistent gradient across the three vaccine formulations.

This variation does not contradict the primary result, because the primary endpoint addresses a narrower question: Grade 2 or Grade 3 general adverse events in the immediate post-vaccination period. The secondary outcomes capture other events and longer follow-up, so they may differ in frequency and in the kinds of uncertainty involved. They provide context about the broader set of outcomes recorded, but they do not change what the primary endpoint measures.

The registry lists 14 posted secondary outcomes, while only a selection of secondary results is described in the registry record for these comparisons. The registry reports no analyses for the secondary comparisons. Accordingly, the posted secondary numbers are best read descriptively, not as a complete inferential account of every secondary objective.

Unadjusted secondary results carry less weight because multiple outcomes and comparisons create more opportunities for apparently unusual results to arise by chance. The registry record does not report a multiplicity procedure or identify a hierarchy of confirmatory secondary tests. An isolated small p-value should therefore be interpreted cautiously, especially when other comparisons do not show a similar pattern. Secondary outcomes can motivate questions for further research, but the available record does not establish a definitive formulation ranking or a causal conclusion for these outcomes.

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

• Number of Subjects With Any, Grade 2, Grade 3 and Medically Attended Solicited Local AEs
• Number of Subjects With Any, Grade 2, Grade 3, Related and Medically Attended Solicited General AEs
• Number of Subjects With Any Unsolicited AEs
• Number of Subjects With Any SAEs
• Number of Subjects With Any Biochemical and Hematological Laboratory Abnormalities
• Number of Subjects With Any Biochemical and Hematological Laboratory Abnormalities
• Number of Subjects With Any Biochemical and Hematological Laboratory Abnormalities
• Number of Subjects With Any Biochemical and Hematological Laboratory Abnormalities

9. Statistical Interpretation

10. Statistical Methods Explained

For a binary endpoint such as whether a participant experienced a qualifying event, the basic descriptive result is the number of participants with the event alongside the number assessed. That makes the observed burden visible in each group without implying that every participant experienced the same course or severity. Here, the primary outcome counts participants with any qualifying Grade 2 or Grade 3 general adverse event during the specified follow-up period.

Risk differences and comparisons

A risk difference contrasts the event risk in one group with the event risk in a reference group. A positive difference means the observed risk is higher in the first group; a negative difference means it is lower. A value near zero indicates little observed separation on this scale. Risk differences are expressed as differences in proportions, not as the number of additional participants in a future trial. The posted primary comparisons use the control group as the reference.

The estimate is accompanied by an interval describing uncertainty around the estimated difference. An interval that crosses zero includes values consistent with no difference, while also allowing for differences in either direction. An interval entirely on one side of zero indicates that the range of values represented by that interval does not include no difference. Such an interval does not, by itself, settle whether a result is clinically important, whether the statistical analysis was prespecified, or whether it remains persuasive after accounting for multiple comparisons.

Interpreting p-values

A p-value describes how compatible the observed data are with a specified null hypothesis under the assumptions of the statistical test. A smaller p-value indicates greater incompatibility with that null model; it is not the probability that the null hypothesis is true, nor a measure of the size or importance of an effect. The posted p-values are explicitly unadjusted. They should therefore be considered with the effect estimate, its interval, the endpoint’s role, and the number of comparisons being examined.

The registry record does not report the statistical test used for the posted comparisons, whether the comparisons were prespecified, or an adjustment for multiplicity. Those details matter because different methods can yield different uncertainty estimates, and repeated testing can increase the chance of at least one apparently notable result when no true differences exist. The analyses described in the record should not be assumed to include an adjustment that is not reported.

Counts, risks, and follow-up

Participant counts are not interchangeable with event counts. A participant with more than one event may still be counted once in an outcome defined as the number of subjects with any event. The registry defines the listed outcomes as counts of participants, but does not report the underlying event-level data in the registry record. Counts also need their corresponding group sizes and time windows for context. The primary endpoint concerns a short post-vaccination period; the unsolicited adverse-event and longer-term outcomes use different follow-up periods.

Different outcomes can also capture different aspects of safety or illness. Solicited symptoms are predefined symptoms actively assessed within a specified window; unsolicited adverse events include other occurrences or symptoms outside the solicited period. Serious adverse events are a separate outcome category. Medically attended RSV-associated respiratory tract infections concern a particular clinical event and follow-up. Their counts should not be combined or treated as though they measure the same underlying construct.

What the record leaves unspecified

The primary analysis population is the Total Vaccinated Cohort, described as all vaccinated subjects. The registry record does not report the method used to handle missing outcome data, whether any participants were excluded after vaccination, or whether sensitivity analyses assessed alternative assumptions about missingness. It also does not report the statistical analysis plan, the model used for the endpoint comparisons, or whether analyses were adjusted for baseline characteristics.

These omissions limit how fully the results can be reconstructed or assessed. A randomized design supports comparisons between assigned groups, but the details of implementation and analysis still matter. The available results support a descriptive reading of the posted counts and unadjusted comparisons; they do not justify assuming an unreported model, analysis population, or correction procedure.

11. Confidence Intervals and Precision

Precision concerns how narrowly the results locate the underlying group difference. The primary endpoint counts are observed in groups of about the same size, but the event counts are still limited. As a result, the estimated differences have uncertainty, and the intervals for all three primary comparisons include zero. The posted results therefore do not precisely distinguish a difference from no difference for these comparisons.

An interval is more informative than a point estimate alone because it displays a range of values compatible with the data under the analysis used. The width reflects uncertainty; a wide interval leaves more possible values open than a narrow one. An interval that includes zero does not prove that the groups are identical. It indicates that the data do not rule out no difference at the interval’s stated level, while retaining the other values within that interval as plausible under the method.

The registry record does not state the confidence level associated with the displayed intervals in the registry record, so no confidence percentage should be assumed. The interpretation also depends on the analysis method, which is not reported in the registry record. The unadjusted p-values and intervals do not account for multiplicity unless such an adjustment is explicitly described; none is reported for these comparisons.

Precision is especially important when comparing formulations. A numerical ordering of observed counts does not itself show that the underlying risks differ, and a lack of a statistically notable comparison does not establish equivalence. Demonstrating equivalence or non-inferiority would require a prespecified margin and an analysis designed for that question; these are not reported in the registry record. The results are best read as estimates with uncertainty rather than as definitive rankings.

The secondary outcomes also have different event frequencies and follow-up periods, so their precision is not uniform. Rare events, in particular, can yield estimates with substantial uncertainty even when the observed counts differ. That makes the serious adverse-event comparisons descriptive rather than a strong basis for concluding that risks are the same or different.

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 translate an observed effect into hypothetical group sizes for future two-arm comparisons. For these binary outcomes, the figures are participants per arm, not a total across arms and not total events. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is set to the stated fraction of the observed effect. These calculations are planning illustrations based on observed results, not the original trial’s reported design assumptions.

The 80% power column means the illustrated number of participants per arm intended to give an 80% chance of detecting the specified effect under the stated assumptions, if that effect and the assumptions hold. The 90% power column applies the same interpretation at a higher target chance of detection. These are design targets for a hypothetical future study; they are not probabilities that the observed result is correct, and they are not estimates of the power achieved in this completed study.

Rows using 100% of the observed effect use the observed difference as the planning target. The 75% and 50% rows use progressively smaller targets. Smaller effects are harder to distinguish from random variation, so the planning illustration requires more participants per arm as the target effect gets smaller. For example, planning for 50% of the observed effect produces larger required group sizes than planning for the full observed effect. The same principle applies when aiming for 90% rather than 80% power: the higher target generally calls for more participants.

The original trial’s sample-size justification, design effect, and planned power are not reported in the registry record. The displayed calculations should not be read as evidence that the completed study was designed to detect the observed effects, nor as a recommendation for a future study. Their usefulness depends on whether the observed effect is a reasonable planning target, an assumption that can be uncertain when estimates come from a modest study.

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 = 0.265; treatment proportion p₁ = 0.320; average p̄ = 0.292.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6432 = 1.2607.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6421 = 0.5404.
4. n = (1.2607 + 0.5404)² / (0.0553)² = 1,061.0, rounded up to 1,062 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 <- 0.264705882352941                          # control proportion
p1 <- 0.32                          # 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(2956837)
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 = 1,062 (80% power), 1,420 (90% power)
 75% of observed effect: n per arm = 1,861 (80% power), 2,491 (90% power)
 50% of observed effect: n per arm = 4,126 (80% power), 5,523 (90% power)
power.prop.test: 1061 
Simulated power at n = 1,062 per arm: 0.798 (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. Multiplicity, Missing Data and Analysis Populations

The registry record includes a range of outcomes: the primary general adverse-event endpoint, multiple secondary outcomes, and immune-response aims described in the study purpose. It reports 21 outcome measures and 14 posted secondary outcomes. Examining several outcomes and comparing multiple vaccine groups with control increases the number of statistical questions. If each is assessed separately without accounting for the collection of tests, the chance of at least one apparently notable result can increase.

The comparisons described for the primary and secondary endpoints are unadjusted. A multiplicity adjustment, a prespecified outcome hierarchy, or another strategy for controlling the overall error rate is not reported in the registry record. The record also does not report which secondary comparisons, if any, were designated confirmatory before the data were examined. Consequently, p-values should be treated as individual, unadjusted results rather than as evidence that the full family of comparisons has been controlled.

The primary analysis population is reported as the Total Vaccinated Cohort, including all vaccinated subjects. Other analysis populations, such as a separate intention-to-treat or per-protocol set, are not reported in the registry record. The method for handling missing observations, withdrawals, or incomplete follow-up is not reported either. Without those details, it is not possible to assess whether missingness could have affected particular outcomes or whether sensitivity analyses were conducted.

The registry record does not report the original statistical analysis plan, interim analyses, sample-size justification, or the full method used for the comparisons. These omissions do not show that such procedures were absent; they mean that they are not reported in the registry record. Interpretation should therefore stay close to the posted counts and unadjusted estimates, without assuming an unreported analysis plan or correction.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: campylobacter gastroenteritis (1.0%), soft tissue infection (1.0%), lung adenocarcinoma (1.0%), pancreatic carcinoma (1.0%), hemiparesis (1.0%).

The registry reports serious adverse events from Day 0 up to study end at Day 360. The observed counts were 1 among 100 participants in formulation 1, 3 among 99 in formulation 2, 2 among 99 in formulation 3, and 2 among 102 in the control group. No deaths are reported in any group.

These are small counts, so the comparisons are imprecise. The posted unadjusted comparison p-values are 0.572 for formulation 1 and 0.626 for formulation 2; both intervals include zero. Formulation 3 has no posted comparison against control for this outcome. These results should not be taken to show that serious-event risks are identical, or that an observed numerical difference establishes a causal effect.

The registry lists several serious-event terms, each with a stated frequency of 1.0%, but the registry entries do not assign those terms to particular groups. Serious adverse events are also not necessarily judged related to study treatment; the relationship assessment is not reported in the registry record. The figures describe events recorded during follow-up and do not, on their own, determine causation.

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

The results describe what was recorded in a randomized phase 2 study of three investigational RSV vaccine formulations and placebo in healthy women. For the primary endpoint, the posted data count participants who experienced Grade 2 or Grade 3 solicited or unsolicited general adverse events during the 7-day (Days 0-6) post-vaccination period. The counts differ across groups, but the unadjusted comparisons have intervals that include zero. The record therefore does not establish a clear difference for this endpoint.

That is not the same as proving the formulations and placebo have equal event risks. A non-significant comparison may reflect uncertainty as well as a small underlying difference, and an equivalence conclusion requires a design and margin suited to that purpose. Neither equivalence margins nor a formal equivalence analysis are reported in the registry record.

The trial’s stated purpose also included ranking formulations by immune response and safety. The registry entries described for this record do not include numerical immune-response results, so no conclusion about relative immune responses can be drawn from the adverse-event results. Nor do the results establish prevention of RSV infection or illness. The medically attended RSV-associated respiratory tract infection outcome is a secondary outcome; its posted comparisons are not a substitute for a prespecified efficacy assessment.

Secondary findings should be interpreted cautiously because they cover different outcomes and timeframes, and the comparisons are unadjusted. The registry record does not report a multiplicity strategy or the full statistical plan. An individual result should not be treated as definitive evidence when considered apart from the rest of the outcome set.

The record does not support medical advice or a recommendation about vaccination. It provides trial observations and statistical comparisons, with limitations in precision and reporting. Decisions about clinical use require evidence and evaluation beyond these posted results.

17. Why This Trial Matters Statistically

This trial illustrates why a clinical study’s stated purpose and its primary endpoint need to be distinguished. The stated aim was to rank vaccine formulations using both safety/reactogenicity and immune-response data. The primary endpoint, however, counts participants with selected moderate or severe general adverse events during a short follow-up period. It is a focused safety measure, not a complete ranking rule and not an immune-response endpoint.

It also shows why observed counts alone are not enough. Counts give a direct description of what happened in each group, while the comparison estimates and their intervals address uncertainty about differences between groups. Here, the primary intervals include zero. A numerical ordering of the observed counts therefore should not be mistaken for a demonstrated ordering of underlying risks.

The secondary outcomes demonstrate another important issue: outcome definitions and follow-up periods affect what an event count means. Unsolicited events, serious adverse events, and medically attended RSV-associated respiratory infections are not interchangeable measures. Their comparisons also contribute to multiplicity, particularly when several formulations and outcomes are examined without a reported adjustment strategy.

Finally, planning calculations based on observed effects are not the same as the study’s original sample-size justification. The registry record does not report that original justification. Distinguishing planning illustrations from observed evidence helps avoid treating a hypothetical future-study calculation as proof of what the completed trial could reliably detect. Together, these points support a cautious, endpoint-specific reading rather than a single broad conclusion about formulation ranking.

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 registry record does not report the full statistical analysis plan, the original sample-size justification, or the methods used to generate the posted comparisons. It also does not report the confidence level associated with the intervals in the registry record. These omissions limit independent assessment of the analysis choices and their relationship to the study’s planned objectives.

The primary endpoint captures a defined set of general adverse events at Grade 2 or Grade 3 during a short post-vaccination window. It does not represent every possible safety outcome, and it does not summarize immune response. The registry gives participant counts rather than event-level detail, so recurrent events, timing within the follow-up window, and duration cannot be assessed from these registry entries.

The secondary outcomes vary in definition and follow-up, and their comparisons are unadjusted. A multiplicity plan is not reported. The available secondary results therefore have limited confirmatory value, particularly when viewed as a collection. The record also does not report how missing data were handled, whether follow-up was complete for each endpoint, or whether results were assessed in additional analysis populations.

Although the study is randomized, the registry record does not provide details of sequence generation, allocation concealment, or the masking procedures. It classifies masking as triple while the official title describes the study as observer-blind; the relationship between these descriptions is not explained. The record also does not report the results for immune-response outcomes in the registry record, even though immune response was part of the study’s stated purpose. These limits mean the posted safety comparisons cannot be used to provide a complete assessment of the study’s ranking objective.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Study to Rank Different Dosages of Antigen of GlaxoSmithKline (GSK) Biologicals' Investigational Respiratory Syncytial Virus (RSV) Vaccine (GSK3003891A), Based on Their Immune Response and Safety, When Administered to Healthy Adult Women — NCT02956837.
• Registry record last updated 2019-07-02; status last verified 2019-06.
• Results first posted 2018-11-20.
• 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

NCT02956837 is a completed phase 2 randomized, parallel-group study in healthy women comparing three single-dose investigational RSV vaccine formulations with placebo. The registry classifies masking as triple, while the official title describes the study as observer-blind. Enrollment is reported as 406.

The primary endpoint counts participants with Grade 2 or Grade 3 solicited or unsolicited general adverse events during the 7-day (Days 0-6) post-vaccination period. The posted unadjusted comparisons against control have intervals that include zero. Selected secondary outcomes cover unsolicited adverse events, serious adverse events, and medically attended RSV-associated respiratory tract infections. The registry record does not report a multiplicity strategy, missing-data method, or original sample-size justification. The posted findings are descriptive and do not establish an immune-response ranking or a definitive efficacy conclusion.