Study to Test the Efficacy of the Vaccine GSK 249553 in Treating Non-small-cell Lung Cancer After Tumour Removal by Surgery: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study to Test the Efficacy of the Vaccine GSK 249553 in Treating Non-small-cell Lung Cancer After Tumour Removal by Surgery (NCT00290355), a randomized phase 2 study. The primary outcome is Number of Patients Reporting Confirmed Non-small-cell Lung Cancer (NSCLC) Recurrence.

2. Clinical Question

3. Trial Design

The registry describes a randomized, double-blind, parallel-group Phase IIB study sponsored by GlaxoSmithKline. Its official title identifies the intended population as patients with MAGE-3-positive non-small-cell lung cancer in stage IB or stage II who had complete surgical resection. The condition listed is non-small-cell lung cancer. The study is marked completed, with enrollment of 182. The registry lists a start date of 2002-05-28 and primary completion date of 2011-07-19.

Participants were assigned to GSK 249553 vaccine or placebo. Both arm descriptions refer to 13 doses administered intramuscularly in the deltoid or lateral thigh regions, alternating between right and left sides. Further details of the placebo schedule are not reported in the registry record. The primary recurrence analysis used the Total Treated cohort: randomized patients who received at least one dose of the assigned study product. The registry also states that patients withdrawn before imaging were removed from this recurrence analysis.

The primary outcome was the number of patients reporting confirmed non-small-cell lung cancer recurrence over a median follow-up of 28 months post-Dose 1. A separate disease-free survival outcome counted recurrence or death over a median follow-up of 44 months post-Dose 1. The registry record does not report the original sample-size justification or the trial’s original statistical design assumptions. Those should not be inferred from the planning illustrations associated with the posted results.

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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The analyses were performed on the Total Treated cohort, which included all patients who were randomised and who received at least one dose of the randomised study product administration. This analysis was based on the confirmed NSCLC recurrence so that patients withdrawn before imaging were removed.

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 Patients Reporting Confirmed Non-small-cell Lung Cancer (NSCLC) Recurrence, measured over Over a median follow-up time of 28 months post-Dose 1.

The primary endpoint counts participants with confirmed non-small-cell lung cancer recurrence. The registry defines recurrence to include local recurrence, regional disease in the mediastinum or supraclavicular nodes, and distant recurrence, including tumours in the contralateral lung or outside the hemithorax. Second primary lung tumours are also included. Confirmation relied on appropriate imaging; the time to recurrence was defined from surgical resection to the first study assessment at which new lesions were found and confirmed.

A recurrence count is a binary outcome for each participant: the participant either met the recurrence definition during the analysis or did not. Comparing the proportions with an event gives a direct description of the observed frequency in each group. It does not, by itself, use all information about when an event occurred or account for differences in follow-up in the way a time-to-event analysis can. The sponsor’s reported Cox regression analysis addresses time to an event through a hazard ratio, a distinct statistical summary.

Recurrence is clinically relevant to an adjuvant study after surgical removal because it records whether disease was detected again during follow-up. The registry’s endpoint definition makes clear that several forms of recurrence and second primary lung tumours count. Interpretation must therefore stay tied to that composite definition and to the specified analysis population and follow-up period.

7. Results Posted in the Registry

• GSK 249553 Group: 32.5% (37/114) versus 42.1% (24/57) on control; risk difference -9.6 percentage points (95% CI -25.1 to 5.8), risk ratio 0.77 (0.51–1.15); unadjusted two-proportion z-test p = 0.214.

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 Patients Reporting Confirmed Non-small-cell Lung Cancer (NSCLC) Recurrence or Death - Disease Free Survival (DFS)

• 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:
  • GSK 249553 Group: 40.2% (49/122) versus 48.3% (29/60) on control; risk difference -8.2 percentage points (95% CI -23.5 to 7.2), risk ratio 0.83 (0.59–1.17); unadjusted two-proportion z-test p = 0.295.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between GSK 249553 Group and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists secondary outcomes spanning clinical events and immune response measures. These include disease-free survival, overall survival, recurrence, seropositivity and antibody concentrations for MAGE-A3 and protein D antigens, and a cell-mediated immunity response. That mix is useful for describing different aspects of follow-up, but the outcomes do not all answer the same question as the primary recurrence endpoint.

The posted disease-free survival outcome counts confirmed recurrence or death over a median follow-up of 44 months post-Dose 1. Its recalculated risk difference is -0.082, with a 95% confidence interval from -0.235 to 0.072 and an unadjusted p-value of 0.295. Because it includes death and uses a different follow-up period, it should be interpreted as a related but distinct outcome, not as a direct replication of the primary endpoint.

The registry identifies 18 secondary outcomes as posted, but detailed numerical results for every listed measure are not reported in the registry record. As a result, the overall direction and consistency across all secondary endpoints cannot be evaluated from the information available in the record. The listed immune measures may inform biological response, but they do not replace evidence about recurrence or survival.

Unadjusted secondary analyses carry less weight when many outcomes are examined because the chance of at least one apparently notable result can increase as the number of tests grows. Whether and how multiplicity was handled is not reported in the registry record. Secondary findings should therefore be viewed as supportive or exploratory unless a pre-specified analysis plan and appropriate adjustment are documented. They cannot independently settle the primary question without full results, definitions, and analysis details.

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

• Percentage of Patients With Disease Recurrence
• Number of Patients Reporting Confirmed Non-small-cell Lung Cancer (NSCLC) Recurrence or Death - Disease Free Survival (DFS)
• Number of Participants Who Died - Overall Survival (OS)
• Number of Subjects Seropositive Against MAGE-A3
• Anti- MAGE-A3 Antibody Concentrations
• Number of Subjects Seropositive Against Protein D (PD) Antigens
• Anti-protein D (Anti-PD) Antibody Concentrations
• Number of Subjects With Cell-mediated Immunity (CMI) Cluster of Differentiation (CD) 4+ Response

9. Statistical Interpretation

10. Statistical Methods Explained

What does the unadjusted risk difference describe?

For a binary endpoint, the risk is the proportion of participants who experienced the defined event during the analysis. The risk difference subtracts the comparator-group risk from the intervention-group risk. Here, the reported estimate is -0.097 for GSK 249553 versus placebo. Its negative sign indicates that the observed proportion was lower in the vaccine group. The estimate is unadjusted: it compares the posted group proportions without adjustment for other participant characteristics or time-to-event information.

The risk difference is an absolute measure, expressed on the proportion scale. It is distinct from a relative measure such as a risk ratio or odds ratio. The registry’s posted comparison also includes relative measures, but interpretation should respect which measure is being discussed. A negative absolute difference does not itself establish causation, and uncertainty around the estimate matters.

What does a confidence interval communicate?

A confidence interval expresses the precision of an estimate under the method used to construct it. A narrower interval indicates greater precision than a wider interval, all else equal. For the primary unadjusted risk difference, the 95% interval is -0.251 to 0.058. Because it spans zero, the posted summary is compatible with no risk difference as well as with values on either side of zero. The interval does not assign a probability to the true effect taking a particular value; it describes the range of estimates compatible with the data and procedure at the stated confidence level.

Why is the Cox result a different analysis?

The sponsor reports a Cox regression analysis with a hazard ratio. Unlike a simple event proportion, a time-to-event method uses information about event timing and can account for participants having different observed follow-up, subject to the model and data assumptions. A hazard ratio compares event hazards between groups over time. It is not the same quantity as a risk difference, and it should not be translated directly into an absolute change in recurrence proportion.

The reported hazard ratio is 0.74, with a 95% confidence interval from 0.44 to 1.24 and p = 0.256. The estimate is below 1, while the interval includes 1. The registry names Cox regression but does not report details of the model specification, covariates, or diagnostics. Those details are not reported in the registry record, so the reported estimate should not be supplemented with assumptions about adjustment or model checking.

Why do analysis populations matter?

The primary recurrence analysis used the Total Treated cohort, defined as randomized participants who received at least one dose of assigned study product. The registry states that patients withdrawn before imaging were removed from the recurrence analysis. This means the analysis is not described as including every randomized participant regardless of treatment or imaging availability. Removing participants without imaging can affect which outcomes are observed; the direction or size of any resulting bias cannot be determined from the registry information alone.

The group sizes used for the primary recurrence counts differ from the group sizes listed for the disease-free survival outcome. These are separate endpoint analyses with different definitions and follow-up periods. They should not be combined or treated as though they use an identical analysis population.

Why does over-dispersion matter?

Over-dispersion is a general statistical issue that arises when observed variability is greater than a model assumes. It can make standard errors too small and confidence intervals too narrow if it is ignored. The primary outcome is a participant count for a binary event, and the registry does not report a dispersion parameter or a model that accounts for over-dispersion. The re-calculated comparison is based on posted summary counts and should not be described as an adjusted model with a dispersion correction.

More generally, model-based results depend on assumptions about the outcome distribution, independence, censoring, and the functional form of any covariates. Some of these assumptions are relevant to time-to-event models. The record names the Cox method but does not provide enough detail to assess all such assumptions. The appropriate statistical reading is therefore to report the method and estimates as registered, while keeping the unspecified modeling details explicit.

How should p-values be used?

A p-value describes how unusual the observed data, or data more extreme, would be under a specified null model and its assumptions. It is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of an effect. The primary unadjusted p-value is 0.214, while the sponsor’s Cox analysis reports p = 0.256. These values arise from different analyses and should not be expected to match exactly.

Neither p-value replaces the estimates and confidence intervals. In particular, the interval conveys a range of plausible values under the calculation, while the p-value compresses evidence relative to a null hypothesis into a single measure. Registry details about a complete pre-specified statistical analysis plan, covariate handling, or multiplicity adjustment are not reported in the registry record.

11. Confidence Intervals and Precision

The primary unadjusted risk difference is -0.097, with a 95% confidence interval from -0.251 to 0.058. The interval is broad enough to include zero and values on either side. This indicates limited precision for distinguishing the direction and size of the absolute difference from these posted counts. The estimate should not be read as an exact effect; the interval communicates substantial uncertainty around it.

The sponsor’s Cox result is a hazard ratio of 0.74 with a 95% confidence interval from 0.44 to 1.24. That interval includes 1, the value representing equal hazards in the usual interpretation of a hazard ratio. It therefore does not pin down the relative hazard with high precision. The hazard-ratio interval and risk-difference interval answer different statistical questions, so their endpoints are not directly comparable.

For the disease-free survival count, the recalculated risk difference is -0.082, with a 95% confidence interval from -0.235 to 0.072. This interval also includes zero. Its endpoint includes death as well as confirmed recurrence, and its median follow-up is 44 months post-Dose 1 rather than the primary endpoint’s 28 months post-Dose 1.

Confidence intervals describe sampling uncertainty under the method used; they do not address every source of uncertainty. In this study, the registry does not provide full details of the original analysis plan, missing-data procedures, or model diagnostics. Those limitations affect how completely the precision and robustness of the findings can be evaluated.

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 these participant-count endpoints. They are based on a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set at the stated fraction of the observed effect. These are planning assumptions for the illustrations, not reported assumptions from the trial’s original design; the original sample-size justification is not reported in the registry record.

The 80% power column describes the number of participants per arm that a hypothetical study would plan for to have an 80% chance of detecting the specified effect under those assumptions. The 90% power column uses a higher target chance of detection, and therefore requires more participants under the listed scenarios. For the primary recurrence endpoint, planning for 100% of the observed effect gives 394 participants per arm for 80% power and 526 for 90% power. Planning for 75% of the observed effect gives 709 and 948, respectively; planning for 50% gives 1,614 and 2,160.

Smaller effects are harder to distinguish from random variation, so the planning illustrations require more participants when the assumed effect is a smaller fraction of the observed effect. These figures are not observed power and do not describe the probability that the completed trial’s result is correct. They are hypothetical planning calculations based on the stated assumptions, not evidence that the trial was designed or conducted using those values. The curve should be read as the same planning relationship between effect size and required enrollment, rather than as a second estimate of treatment effect.

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 = 0.421; treatment proportion p₁ = 0.325; average p̄ = 0.373.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6838 = 1.3403.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6804 = 0.5727.
4. n = (1.3403 + 0.5727)² / (-0.0965)² = 393.0, rounded up to 394 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 <- 0.421052631578947                          # control proportion
p1 <- 0.324561403508772                          # 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(290355)
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 = 394 (80% power), 526 (90% power)
 75% of observed effect: n per arm = 709 (80% power), 948 (90% power)
 50% of observed effect: n per arm = 1,614 (80% power), 2,160 (90% power)
power.prop.test: 393 
Simulated power at n = 394 per arm: 0.802 (20000 replicates)

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

14. Analyses Reported by the Sponsor

The sponsor’s registered analysis uses Cox regression and reports a hazard ratio of 0.74, with a 95% confidence interval from 0.44 to 1.24 and p = 0.256. The unadjusted re-calculation from the posted recurrence counts gives a risk difference of -0.097, with a 95% confidence interval from -0.251 to 0.058 and p = 0.214.

These estimates need not be identical because they describe different quantities and use different methods. The risk difference compares the observed proportions with recurrence, whereas the Cox hazard ratio summarizes relative event hazards over time. The Cox analysis can use event timing and follow-up information; the unadjusted count comparison does not. The registry does not report the full Cox model specification or covariate details, so no additional adjustment should be presumed.

The sponsor’s analysis is the trial’s reported time-to-event analysis, while the re-calculation is a simple unadjusted comparison of the posted summaries. Neither should be substituted for the other. Both point estimates are below their respective null values, but both confidence intervals include their corresponding null values. The difference between their p-values does not indicate a contradiction; the analyses are not estimating the same effect in the same way.

15. Multiplicity, Missing Data and Analysis Populations

The record identifies a primary recurrence outcome and a broad set of secondary outcomes, but it does not report a full statistical analysis plan describing the order of testing or any multiplicity procedure. Multiplicity concerns arise when several outcomes or comparisons are tested: even if each test uses a conventional threshold, the chance of at least one chance finding can increase across a family of tests. Without a stated plan, the evidential status of secondary p-values cannot be fully assessed.

The registry includes 19 outcome measures and reports 18 secondary outcomes. The primary endpoint is confirmed recurrence; secondary measures include recurrence or death, overall survival, recurrence, and immunologic outcomes. The record does not describe whether secondary tests were intended to be confirmatory, how their testing was sequenced, or whether any adjustment was used. Those details are not reported in the registry record, so unadjusted secondary results should not be treated as independent confirmation of the primary result.

The primary recurrence population is defined as randomized participants who received at least one dose of assigned study product. The record further states that participants withdrawn before imaging were removed from this recurrence analysis. It does not provide a complete account of missing observations, reasons for withdrawal, imputation, censoring, or sensitivity analyses. Consequently, the influence of missing imaging information on the estimate cannot be quantified from the registry record.

The sponsor reports a Cox regression analysis, but details such as covariate specification, model diagnostics, and handling of censoring are not reported in the registry record. The original sample-size justification and trial-design assumptions are also not reported. These omissions do not invalidate the posted estimates, but they limit independent assessment of pre-specification, robustness, and how fully the analysis followed the intended design.

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: chronic obstructive pulmonary disease (3.3%), acute myocardial infarction (3.3%), pneumonia (3.3%), metastases to central nervous system (1.7%), myocardial infarction (1.7%).

The registry reports serious adverse events for the GSK 249553 group with 122 participants at risk and 41 affected, and for the placebo group with 60 at risk and 21 affected. For other adverse events, it reports 120 affected among 122 at risk in the vaccine group and 53 among 60 at risk in the placebo group. The listed deaths are 54 and 26, respectively. These figures are descriptive and should be read in the context of the registry’s safety populations and reporting definitions.

The serious-event terms listed with percentages include chronic obstructive pulmonary disease, acute myocardial infarction, and pneumonia at 3.3% each, as well as metastases to central nervous system, myocardial infarction, and cardiac failure at 1.7% each. These are reported terms, not a complete account of every event or a causal comparison. The registry information does not establish whether an event was caused by study product. The serious adverse event figures do not, on their own, determine comparative safety; definitions, exposure, follow-up, and detailed event-level context matter.

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

The posted primary counts show confirmed recurrence in 37 of 114 participants assigned to GSK 249553 and 24 of 57 assigned to placebo. The unadjusted risk difference is -0.097, with a 95% confidence interval from -0.251 to 0.058. The sponsor’s Cox analysis reports a hazard ratio of 0.74, with a 95% confidence interval from 0.44 to 1.24. Both estimates point below their respective null values, while both intervals include the null value.

That combination means the data do not precisely establish the direction or magnitude of a group difference. A point estimate is the value calculated from the observed data, not a guarantee of the underlying effect. The confidence intervals show that the estimates remain compatible with a range of possibilities. The unadjusted p-values, 0.214 for the risk difference and 0.256 for the Cox analysis, should be interpreted alongside those estimates and intervals.

The results do not prove that the vaccine reduces recurrence, and they do not prove that the groups are equivalent. Failure to establish a clear difference is not the same as demonstrating no difference. The study’s specific population, treatment regimen, recurrence definition, analysis population, and follow-up period constrain what can be concluded.

The findings also do not establish what would happen in other stages of disease, other treatment contexts, or populations not described in the registry. The registered outcome includes several types of recurrence and second primary lung tumours. The disease-free survival outcome is broader because it includes death and uses a different follow-up period. Those distinctions matter when relating the posted statistics to clinical questions.

Finally, the record does not supply enough detail to assess all aspects of the original statistical plan, missing-data handling, or secondary-outcome multiplicity. The most defensible reading is descriptive: the registry reports the event counts and model estimate with uncertainty, while the available information does not support a definitive causal conclusion.

18. Why This Trial Matters Statistically

This trial is statistically instructive because it presents the same clinical question through both an absolute event-proportion comparison and a time-to-event analysis. The unadjusted risk difference and the Cox hazard ratio are related in purpose but are not the same estimand. Reading them side by side demonstrates why an estimate from posted counts should not be expected to reproduce a sponsor’s time-to-event result exactly.

The confidence intervals are central to interpretation. Both the primary risk-difference interval and the sponsor’s hazard-ratio interval include their respective null values. The point estimates alone suggest a direction, but the intervals show that the estimates are not precise enough to settle the magnitude or direction. This is a practical example of why statistical interpretation should consider estimates, uncertainty, outcome definitions, and analysis methods together.

The case also highlights how analysis populations and outcome definitions shape evidence. The primary recurrence analysis excludes patients withdrawn before imaging, according to the registry description. The secondary disease-free survival outcome includes death and uses a longer median follow-up period. Neither distinction is merely administrative: each affects which events are counted and what the comparison means.

Finally, the registry’s limited reporting of the original design assumptions, missing-data procedures, model details, and multiplicity handling illustrates the limits of registry-only interpretation. These gaps do not justify filling in details from convention or assumption. They instead define what can be assessed: the posted estimates can be explained, while claims about the full pre-specified statistical strategy remain unavailable.

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 primary recurrence analysis is not described as an all-randomized analysis. It uses the Total Treated cohort, and the registry says patients withdrawn before imaging were removed. The information provided does not specify the number or characteristics of those withdrawals, reasons for missing imaging, or how missing outcomes were otherwise handled. The consequences for the estimated group comparison cannot be determined from the registry record.

The registry gives a sponsor Cox estimate but not its full model specification, covariates, diagnostics, or censoring details. The re-calculated risk difference is unadjusted and does not account for event timing. These estimates are complementary descriptions, but neither resolves the limitations of the other method.

The endpoint includes local, regional, and distant recurrence as well as second primary lung tumours. This composite definition may combine events with different clinical features. The registry defines them, but does not report separate primary-endpoint estimates for each type in the registry record for this analysis.

The original sample-size justification and statistical design assumptions are not reported in the registry record. The planning illustrations use assumptions stated for those calculations and should not be taken as evidence of the original design. Multiplicity procedures for secondary analyses and detailed results for every listed secondary measure are also not reported in the registry record. These omissions limit assessment of confirmatory status and robustness.

Finally, the results apply to the described population and treatment setting: patients with the specified MAGE-3-positive stages after complete resection, assigned to vaccine or placebo. Extrapolation beyond those characteristics is not supported by the registry results alone.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Study to Test the Efficacy of the Vaccine GSK 249553 in Treating Non-small-cell Lung Cancer After Tumour Removal by Surgery — NCT00290355.
• Registry record last updated 2020-01-02; status last verified 2019-12.
• Results first posted 2019-02-25.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

NCT00290355 is a completed randomized, double-blind, parallel-group Phase IIB study sponsored by GlaxoSmithKline. It compared GSK 249553 vaccine with placebo as adjuvant therapy for specified MAGE-3-positive non-small-cell lung cancer after complete surgical resection. Registry enrollment was 182.

The primary outcome was confirmed recurrence over a median follow-up of 28 months post-Dose 1. The unadjusted risk difference was -0.097, with a 95% confidence interval from -0.251 to 0.058 and p = 0.214. The sponsor reported a Cox hazard ratio of 0.74, with a 95% confidence interval from 0.44 to 1.24 and p = 0.256. Both analyses have confidence intervals that include their respective null values.

The registry also reports a disease-free survival outcome, safety counts, and secondary immune and clinical outcomes. Important details—including the original sample-size justification, complete missing-data procedures, and multiplicity handling—are not reported in the registry record.