A Phase III Study of a 2-dose Regimen of a Multivalent Human Papillomavirus (HPV) Vaccine (V503), Administered to 9 to 14 Year-olds and Compared to Young Women, 16 to 26 Years Old (V503-010): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase III Study of a 2-dose Regimen of a Multivalent Human Papillomavirus (HPV) Vaccine (V503), Administered to 9 to 14 Year-olds and Compared to Young Women, 16 to 26 Years Old (V503-010) (NCT01984697), a randomized phase 3 study. The primary outcome is Geometric Mean Titers to Human Papillomavirus (HPV) Type 6 After the Last Dose of V503 in the Planned Regimen.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group phase III clinical trial of V503, a multivalent HPV vaccine. The registry lists enrollment of 1518 and describes the study as a 37-month safety and immunogenicity study. It is marked completed. The listed study dates are 2013-12-12 for the start and 2015-06-19 for primary completion.

The groups included girls aged 9 to 14 years receiving V503 at Months 0 and 6; boys aged 9 to 14 years receiving the same schedule; girls and boys aged 9 to 14 years receiving doses at Months 0 and 12; girls aged 9 to 14 years receiving doses at Months 0, 2, and 6; and young women aged 16 to 26 years receiving doses at Months 0, 2, and 6. The descriptions specify 0.5 mL intramuscular injections. They also describe an additional Month 36 dose for the two-dose groups and for a subset of participants in the three-dose groups.

The registry records no masking. It does not report the randomization procedure, allocation ratio, stratification factors, or the basis for the enrollment target. These details are important for assessing how groups were formed and what design assumptions were made, but they are not reported in the registry record. The record likewise does not give a fuller account of study sites or the procedures used to maintain follow-up.

The study compares groups defined by both age and vaccine schedule. The young women receiving three doses serve as the principal reference for the stated immunogenicity question, but comparisons with that group do not isolate schedule from age. The girls receiving three doses provide an additional group with a different age composition but the same planned schedule as the young-women comparator. The registry’s descriptions identify these groups; they do not establish that every possible cross-group contrast was intended to answer the same scientific question.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: A total of 1536 participants were screened and 1518 were randomized into the study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All participants who 1) received all vaccinations in the planned regimen, 2) had a serum sample collected 4 weeks after the last vaccination in the planned regimen, 3) were seronegative at Day 1 for the relevant HPV type, and 4) had no protocol violations that would interfere with evaluation of the immune response.

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 Geometric Mean Titers to Human Papillomavirus (HPV) Type 6 After the Last Dose of V503 in the Planned Regimen, measured over 4 weeks after the last dose of V503 in the planned regimen (Month 7 or Month 13).

The primary endpoint is the geometric mean titer of antibodies to HPV type 6, measured four weeks after the last dose in the planned regimen, at Month 7 or Month 13. Titers were measured with a competitive Luminex immunoassay and expressed in mMU/mL. The primary analysis population was defined using vaccination completion, availability of the specified serum sample, baseline seronegativity for the HPV type, and absence of protocol violations that would interfere with evaluation. The registry record’s description of the last condition is incomplete.

A geometric mean summarizes antibody titers on a multiplicative scale. In general, this measure can be useful when values vary over a wide range and relative rather than arithmetic differences are meaningful. It describes the central level of measured antibody response in a group; it is not a count of participants with a response and does not itself describe the spread of individual titers. The registry reports 95% confidence intervals around the group estimates.

The endpoint was selected to address immunogenicity: the antibody response after the planned vaccination schedule. Comparing geometric mean titers between groups can inform whether measured responses are similar or differ in magnitude. It does not directly measure clinical HPV infection or disease. The registry record does not report a clinical efficacy endpoint as the primary outcome, nor does it provide information establishing a relationship between a particular titer and clinical protection.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Geometric means and ratios

The primary outcome is a geometric mean titer, a measure that summarizes values on a multiplicative scale. In general, a geometric mean is useful when measurements are positive and differences are more naturally understood as proportional than as fixed additive changes. It is distinct from an arithmetic average. A ratio compares two group-level values: a ratio of 1 corresponds to equal estimates, while a ratio above or below 1 indicates which group has the larger estimate. The ratio itself does not explain why the groups differ.

The registry reports group geometric mean titers with 95% confidence intervals and sponsor analyses of ratios using ANOVA. ANOVA is a framework for comparing group means. The registry record does not describe the exact model specification, transformations, covariates, or handling of departures from model assumptions. Those details should not be inferred from the method label alone.

What does a 95% confidence interval express?

In general, a confidence interval describes the uncertainty associated with an estimate under the assumptions of the analysis. A narrower interval indicates greater precision than a wider interval, but an interval does not express the probability that the true value lies within that particular calculated interval. The intervals reported for the primary outcome quantify uncertainty in group titers or ratios, not uncertainty about clinical effectiveness.

For a ratio, the value 1 is the no-difference reference. The sponsor-reported ratio intervals in the primary analysis lie above that reference. This helps describe the direction and precision of the estimated ratio. It does not determine noninferiority without a margin. A noninferiority analysis asks whether the data rule out an unacceptable loss relative to a prespecified boundary. The boundary and its justification are not reported in the registry record, so readers cannot reconstruct that decision rule from the ratio and interval alone.

How should the reported p-values be read?

A p-value describes how compatible the data are with a specified statistical null model, given the analysis assumptions. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect. The registry reports p-values of <0.001 alongside analyses labeled noninferiority or equivalence. Without the test’s hypotheses, margin, and decision criteria, the p-value cannot be interpreted as a complete statement that a noninferiority condition was met. The registry does not report those details.

For selected secondary outcomes, the registry also includes sponsor analyses using the Miettinen and Nurminen method for seroconversion percentage differences. In general, this method is used to form confidence intervals and tests for differences in proportions. The registry reports sponsor estimates and intervals for selected comparisons. Because seroconversion percentages are close to the upper bound of 100% in these groups, a difference in percentage points has a different interpretation from a ratio of geometric mean titers. Neither measure by itself establishes a clinical outcome.

What is the role of the unadjusted recalculations?

Some posted secondary comparisons are simple unadjusted recalculations from the displayed group counts. A risk difference is the difference between the observed proportions in two groups. Its confidence interval describes uncertainty in that unadjusted contrast. These calculations do not replace the sponsor’s prespecified analysis, and they may not account for design features, analysis populations, or multiplicity. They are best treated as descriptive checks of the posted summary data. The registry does not identify these recalculations as the sponsor’s analysis.

The primary endpoint has sponsor-reported analyses, and the discrepancy indicator is false. Consequently, there is no recorded discrepancy that would make a recalculated result take precedence. More generally, when a recalculated result differs from a sponsor analysis, the sponsor’s own analysis takes precedence because it may reflect the trial’s prespecified methods and full analysis data, while a summary-statistic recalculation is necessarily limited.

Why do the analysis population and missingness matter?

The primary outcome is assessed in participants meeting specified vaccination, sample-collection, baseline-serostatus, and protocol-violation criteria. In general, limiting an analysis to participants with observed measurements can affect how representative the estimate is of everyone enrolled, particularly if the availability of measurements is related to participant characteristics or outcomes. The registry record does not report the number excluded from the primary analysis, the reasons for missing measurements, or methods for addressing missing data. It also does not report whether sensitivity analyses assessed the effect of those omissions.

Sample-size illustrations, when used, describe planning under stated assumptions and are not observed power. The trial’s sample-size justification and assumptions are not reported in the registry record. The registry’s listed enrollment should not be mistaken for the size of the primary analysis population, whose group denominators are separately reported.

10. Analyses Reported by the Sponsor

The sponsor-reported primary analyses use ANOVA to compare HPV type 6 geometric mean titers in each two-dose group with the young-women group receiving three doses. The registry gives ratios of 2.15, 2.02, and 3.47, with 95% confidence intervals of 1.83 to 2.53, 1.73 to 2.36, and 2.93 to 4.11, respectively. The sponsor-reported p-values are each <0.001. These are the primary comparative results recorded for the listed groups.

The discrepancy indicator is false, so no conflict is flagged between the sponsor’s result and a recalculation. For the primary endpoint, the listed comparisons should therefore be read as the sponsor’s analysis, not replaced by a summary-statistic calculation. The selected secondary outcomes include both sponsor-reported analyses and unadjusted recalculations, which answer related but not necessarily identical questions.

The sponsor labels the primary comparisons as noninferiority or equivalence analyses. The reported ratios and intervals communicate the estimated relative antibody titers and their precision. However, the registry record does not report the noninferiority margin or the complete statistical decision rule. The sponsor’s reported label and p-values can be stated, but independent verification of the noninferiority criterion is not possible from the information recorded. The comparisons also involve groups that differ in both age and regimen, so they do not isolate the effect of schedule alone.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and many additional outcome measures, including seroconversion outcomes for multiple HPV types. It records 22 outcome measures and 54 analyses. Those counts indicate a broad set of recorded endpoints and analyses, but do not reveal how the analyses were prioritized or whether statistical adjustments were made for multiple comparisons.

Multiplicity matters because repeated testing across outcomes, groups, or time points can increase the chance of obtaining apparently strong evidence by chance. In general, trial protocols may define one primary analysis and specify how secondary analyses are treated, including whether p-values are adjusted or interpreted as exploratory. The registry record does not report a multiplicity adjustment, a testing hierarchy, or a complete analysis plan. The sponsor-reported p-values should therefore be interpreted in the context of their stated analyses, without assuming a family-wise error strategy or a particular adjustment.

The registry defines the primary analysis population using receipt of planned vaccinations, collection of a serum sample four weeks after the last planned dose, baseline seronegativity for the relevant HPV type, and the absence of protocol violations that would interfere with evaluation. The description of the final criterion ends mid-sentence, so its complete definition is not reported in the registry record. The available record does not provide the number of participants meeting each criterion or explain how exclusions affected the group estimates.

Missing-data handling is not reported in the registry record. In particular, the record does not describe imputation, assumptions about missing samples, sensitivity analyses, or whether missingness differed among groups. The reported primary endpoint therefore describes the defined analysis population with evaluable measurements; it should not automatically be taken as an estimate for every enrolled participant.

The record also does not report an interim-analysis plan, stopping rules, or a sample-size justification. The enrollment figure alone does not establish the precision intended for every endpoint. These omissions limit independent assessment of the analysis strategy and of how secondary findings should be weighted relative to the primary endpoint.

12. 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: abortion induced (1.3%), abdominal pain (0.7%), appendicitis (0.3%), forearm fracture (0.3%), dermatitis atopic (0.3%).

The safety data list the number at risk and the number with serious adverse events for each group. The recorded serious adverse event counts are 6 among 294 girls aged 9 to 14 years receiving doses at Months 0 and 6; 9 among 296 boys receiving that schedule; 6 among 293 girls and boys receiving doses at Months 0 and 12; 6 among 300 girls receiving three doses; and 11 among 313 young women receiving three doses. The registry also records one death in the three-dose girls’ group and zero deaths in each of the other listed groups.

These are descriptive counts. They do not establish that an event was caused by vaccination, and the registry information presented does not give the timing, relatedness assessments, clinical details, or comparative statistical analysis for these events. The groups differ in age and regimen, and the displayed counts should not be read as a controlled causal comparison.

The record lists serious-event terms including “Abortion induced” at 1.3%, “Abdominal pain” at 0.7%, and “Appendicitis,” “Forearm fracture,” “Dermatitis atopic,” and “Gastritis” at 0.3% each. These terms are not assigned to specific arms in the recorded information, so they cannot be used to compare groups. The listed serious adverse event counts and terms describe reported safety information; they do not, on their own, characterize the full safety profile.

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

The primary findings describe HPV type 6 antibody titers measured four weeks after participants completed their planned vaccination regimen. The group geometric mean estimates and sponsor-reported ratios show the measured antibody response in the specified immunogenicity population. Their confidence intervals convey uncertainty around those estimates. The results are evidence about an immunologic measurement at the specified assessment time, not direct evidence about HPV infection, disease, or long-term protection.

The registry’s broader stated objective was to assess whether antibody responses after two-dose schedules in participants aged 9 to 14 years were not inferior to responses after the three-dose schedule in young women aged 16 to 26 years. The reported ratios are above 1, and the sponsor labels the analyses as noninferiority or equivalence. However, the noninferiority margin and complete decision rule are not reported in the registry record. The ratios and p-values alone therefore do not allow an independent reader to verify the formal noninferiority criterion.

The comparison also combines differences in age and vaccination schedule. A higher estimated titer in one group cannot be attributed solely to receiving a different number or timing of doses when the groups differ in age as well. The three-dose girls’ group offers a further reference, but the recorded results do not provide a design-based separation of schedule and age effects.

The secondary seroconversion results add information about the proportion of participants meeting response criteria for selected HPV types. They complement, but do not replace, the primary titer outcome. The outcomes with numerical results in the registry show a generally high seroconversion pattern, while the numerical results for the other listed types are not reported in the registry record. The available information also does not establish how multiplicity was handled.

Thus, the study record supports a statistical description of antibody measurements and reported safety events in the listed groups. It does not by itself establish clinical efficacy, protection duration, or the causal safety effect of the vaccine. Those conclusions would require evidence and analysis beyond the registry results described.

14. Why This Trial Matters Statistically

This trial illustrates an important feature of immunogenicity studies: the main comparison may concern a measured biological response rather than a clinical event. Here, the primary outcome is a geometric mean antibody titer, while selected secondary outcomes measure seroconversion percentages. Understanding which quantity is being compared prevents a statistical result about antibody levels from being misread as a direct estimate of infection or disease prevention.

The study also highlights the distinction between a noninferiority objective and a conventional test of whether two groups differ. A noninferiority analysis relies on a clinically or scientifically justified margin and a prespecified decision rule. The registry reports ratios, confidence intervals, p-values, and a noninferiority or equivalence label, but not the margin. That gap matters because the conclusion depends on the boundary, not solely on whether a ratio is above or below 1 or whether a p-value is small.

Group composition is another central statistical issue. The two-dose groups are participants aged 9 to 14 years, while the key three-dose reference consists of young women aged 16 to 26 years. Age and regimen vary together in that comparison, limiting interpretation as a pure schedule contrast. This is a reminder that a numerical comparison can be precise while still not separating the possible reasons for a difference.

Finally, a trial with numerous outcomes requires clarity about endpoint priority, multiplicity, and missing-data handling. The registry’s primary endpoint is identifiable, but its complete analysis plan, adjustment strategy, and missing-data methods are not reported. These omissions do not erase the estimates, but they constrain how confidently readers can interpret secondary p-values or independently evaluate the sponsor’s noninferiority claim.

15. 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.

Several trial-specific limitations affect interpretation. First, the principal comparison contrasts participants aged 9 to 14 years receiving two doses with young women aged 16 to 26 years receiving three doses. Age and regimen differ together, so the comparison cannot be interpreted as the isolated effect of dose schedule. The registry includes girls aged 9 to 14 years receiving three doses, but the recorded primary comparisons are against the young-women group.

Second, the primary analysis is restricted to a defined population requiring completion of planned vaccinations and an evaluable serum sample, among other criteria. This is relevant to an immunogenicity endpoint, but it means the primary estimate does not necessarily represent all enrolled participants. The registry’s definition of the final protocol-violation criterion is truncated, and the number of participants excluded for each reason is not reported in the registry record.

Third, the noninferiority or equivalence margin, its justification, and the complete decision rule are not reported. This prevents an independent assessment of whether the reported intervals meet the intended criterion. The statistical model details beyond the ANOVA label, the multiplicity strategy, and missing-data procedures are also not reported in the registry record.

Fourth, numerical secondary results are included for selected HPV types, but the registry’s listed secondary outcomes cover additional types without numerical results in this record. The breadth of outcome measures makes the absence of a reported multiplicity strategy particularly relevant to interpreting secondary p-values.

Finally, the safety information consists of group-level serious adverse event counts, deaths, and a list of serious-event terms without arm attribution for those terms. The record does not provide enough detail to infer causality or compare event risk after accounting for differences in age and regimen. These limitations apply to interpretation of this registry record and do not, by themselves, establish that the study was conducted improperly.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Phase III Study of a 2-dose Regimen of a Multivalent Human Papillomavirus (HPV) Vaccine (V503), Administered to 9 to 14 Year-olds and Compared to Young Women, 16 to 26 Years Old (V503-010) — NCT01984697.
• Registry record last updated 2018-08-08; status last verified 2018-07.
• Results first posted 2016-04-26.
• 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.

19. Record Summary

This completed randomized, unmasked, parallel-group phase III study evaluated V503 schedules in boys and girls aged 9 to 14 years and young women aged 16 to 26 years. Its primary endpoint was HPV type 6 geometric mean antibody titer four weeks after the last planned dose. Sponsor-reported ratios comparing the two-dose groups with the young-women three-dose group were 2.15, 2.02, and 3.47, with 95% confidence intervals above 1. The registry labels these analyses noninferiority or equivalence, but the margin and full decision rule are not reported.

Selected secondary seroconversion results show a broadly high response pattern for the HPV types with numerical results recorded. Group-level serious adverse event counts and deaths are also listed. The results address antibody responses in the specified analysis population; they do not directly establish clinical efficacy or isolate schedule effects from age differences. Multiplicity procedures and missing-data methods are not reported in the registry record.