Study of Purified Vero Rabies Vaccine Compared With Two Reference Rabies Vaccines in a Simulated Post-Exposure Regimen in Adults: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Purified Vero Rabies Vaccine Compared With Two Reference Rabies Vaccines in a Simulated Post-Exposure Regimen in Adults (NCT03965962), a randomized phase 3 study. The primary outcome is Percentage of Participants With Rabies Virus Neutralizing Antibody (RVNA) Titers Greater Than or Equal to (>=) 0.5 International Units Per Milliliter (IU/mL)-Non-Inferiority Analysis.

2. Clinical Question

The primary endpoint is an antibody-response threshold, rather than a measure of clinical rabies or protection against exposure. The objective also included describing safety and immune response at other visits. The study was designed around a simulated post-exposure regimen in healthy adults; the registry record does not establish how these findings apply to people who have experienced a rabies exposure.

3. Trial Design

This was a randomized, quadruple-masked, parallel-group Phase 3 trial. The registry lists enrollment of 640 and a France setting. The study status is completed. The record lists a start date of 2019-07-01 and primary completion of 2020-12-22. Those dates describe the study record and do not add information about the conduct of individual visits or analyses.

Four groups are described. Group 1 received 0.5 mL of VRVg-2 by intramuscular injection with HRIG at Day 0. Group 2 received 0.5 mL of Verorab with HRIG, and Group 3 received 1 mL of Imovax Rabies with HRIG. Group 4 received VRVg-2 without HRIG. The intervention descriptions list vaccine injections on Days 0, 3, 7, 14 and 28, while HRIG was given at Day 0 in the groups assigned to receive it.

The principal comparison focused on the three groups receiving HRIG: VRVg-2 plus HRIG versus Verorab plus HRIG, and VRVg-2 plus HRIG versus Imovax Rabies plus HRIG. The separate VRVg-2 group allows the registry to describe responses and safety without HRIG, but it is not listed as a comparator in the primary non-inferiority analyses.

The registry reports primary endpoint results for 239 participants in the VRVg-2 plus HRIG group, 77 in the Verorab plus HRIG group, and 78 in the Imovax Rabies plus HRIG group. These are endpoint analysis counts, not a breakdown of the full enrollment. The record does not report a sample-size justification or enough detail to explain the difference between enrollment and these analysis counts.

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 640 participants were enrolled in the study. Data presented in the disposition table (all milestones) is based on randomized group, except for safety analysis set (SafAS) milestone which is presented based on the actual vaccination group.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Analysis was performed on the per-protocol analysis set (PPAS) that included all participants who received at least 1 dose of the study vaccines. The participants who presented protocol deviations, met PPAS exclusion criteria were excluded from PPAS. Data for this outcome measure was planned to be collected and analyzed only for Groups 1, 2, and 3.

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.

Population: Analysis was performed on all randomized participants.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With Rabies Virus Neutralizing Antibody (RVNA) Titers Greater Than or Equal to (>=) 0.5 International Units Per Milliliter (IU/mL)-Non-Inferiority Analysis, measured over Day 28.

The primary endpoint is the percentage of participants whose RVNA titer against rabies virus was greater than or equal to 0.5 IU/mL at Day 28. Antibody titers were assessed using the Rapid Fluorescent Focus Inhibition test (RFFIT) assay. The registry defines the assessment time as Day 28, or 14 days after the fourth vaccine injection.

This threshold converts a measured antibody level into a binary outcome: each participant either meets the threshold or does not. The resulting percentage is the proportion who meet it in each analysis group. A confidence interval describes uncertainty around an estimated percentage or contrast; it does not indicate that every participant’s antibody level is known with equal precision.

The endpoint was selected for the stated non-inferiority question: whether the proportion reaching the threshold with VRVg-2 plus HRIG was not unacceptably lower than the proportion with either reference vaccine plus HRIG. For a non-inferiority conclusion, the estimated contrast must be considered against a prespecified margin and the analysis method. The registry record labels the sponsor analyses as non-inferiority analyses, but the margin and method are not reported in the registry record.

The antibody threshold is an immunogenicity outcome. It is not a direct count of rabies cases, and these results alone do not answer questions about clinical outcomes after an actual exposure. The registry also describes secondary antibody assessments at other visits and safety outcomes.

7. Results Posted in the Registry

• Group 2 Verorab+HRIG: 100.0% (77/77) versus 99.6% (238/239) on control; risk difference 0.4 percentage points (95% CI -0.4 to 1.2), risk ratio 1.00 (1.00–1.01); unadjusted two-proportion z-test p = 0.570.
• Group 3 Imovax Rabies+HRIG: 98.7% (77/78) versus 99.6% (238/239) on control; risk difference -0.9 percentage points (95% CI -3.5 to 1.8), risk ratio 0.99 (0.97–1.02); unadjusted two-proportion z-test p = 0.403.

8. Statistical Interpretation

9. Statistical Methods Explained

What does the percentage measure?

The endpoint is binary at the participant level: the RFFIT-assessed titer either reached the stated threshold or did not. A group percentage summarizes the number meeting the threshold relative to the number analyzed. This is easy to interpret, but it discards information about how far each participant’s titer was above or below the threshold. The registry also lists continuous-style immune-response measures, including geometric mean titers, as secondary outcomes.

What is a risk difference?

General statistical explanation: a risk difference expresses the difference between two group proportions on the same probability scale. For a percentage endpoint, it can be presented in percentage points. A value near zero means the observed proportions are close; the interval around the estimate indicates which contrasts remain plausible given the data and the analysis assumptions. The unadjusted re-calculations use the posted group summary statistics and should be understood as simple contrasts, not as replacements for the sponsor’s analysis.

How does non-inferiority differ from a difference test?

General statistical explanation: a non-inferiority question asks whether an intervention is no worse than a comparator by more than a prespecified amount. That amount is the non-inferiority margin. The analysis normally evaluates an estimated contrast and its confidence interval relative to that margin. A conventional test that fails to find a difference does not answer the same question: it may reflect a small true contrast, limited precision, or both. The registry labels the sponsor’s analyses as non-inferiority analyses but does not report the margin or method, so the numerical results can be described without independently verifying the sponsor’s decision rule.

Why does the confidence interval matter?

General statistical explanation: a confidence interval represents uncertainty in an estimate under the assumptions of the method used. A narrower interval indicates greater precision for that estimate; a wider interval leaves more possible values consistent with the data. For non-inferiority, the interval’s position relative to the margin is central. Without the margin, the sponsor’s interval cannot be independently assessed against the stated criterion. The confidence intervals for the group percentages and the sponsor’s intervals for contrasts are distinct quantities and should not be conflated.

What do the re-calculated p-values say?

The unadjusted re-calculated p-values are 0.57 and 0.403 for the two active comparator contrasts. General statistical explanation: a p-value measures how unusual data at least as discrepant as those observed would be under a specified null hypothesis and model. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a contrast. These p-values are not a non-inferiority margin test and do not establish equivalence. The sponsor’s own reported non-inferiority results remain the relevant analyses for the sponsor’s stated objective.

Why can high response percentages still leave uncertainty?

When most participants meet a binary threshold, the observed percentages can cluster close to the upper bound of 100%. This can make direct comparisons harder to interpret: few participants fall below the threshold, and the outcome does not distinguish degrees of response among those above it. General statistical explanation: such a ceiling pattern can limit the information a binary endpoint provides about differences between groups. It does not make the endpoint invalid, but it makes the analysis population, interval estimates, and non-inferiority margin especially important.

What does the planning illustration represent?

The planning illustration is based on the observed effect and states assumptions of a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It presents participants per arm for 80% power and participants per arm for 90% power under planning for 100%, 75% or 50% of the observed effect. The listed values are 1,774 and 2,374 for 100% of the observed effect; 2,756 and 3,689 for 75%; and 5,305 and 7,102 for 50%. These are planning illustrations, not observed trial results or observed power. The registry record does not report that these assumptions were used to design this trial, nor does it report the trial’s own sample-size justification.

10. Confidence Intervals and Precision

The posted confidence intervals for the three group percentages are 97.7 to 100.0 for VRVg-2 plus HRIG, 95.3 to 100.0 for Verorab plus HRIG, and 93.1 to 100.0 for Imovax Rabies plus HRIG. The intervals extend to 100.0, consistent with the very high observed percentages and the natural upper limit of a percentage. The interval widths differ across groups; the reference-group analysis counts are smaller than the VRVg-2 plus HRIG count.

For the sponsor’s comparison with Verorab plus HRIG, the reported 95% confidence interval for the difference in percentage is -2.33 to 4.35. For the comparison with Imovax Rabies plus HRIG, it is -1.32 to 6.50. These intervals describe uncertainty about contrasts, rather than uncertainty about each group percentage separately. They include values on both sides of zero, so the posted interval estimates do not isolate a single direction of difference.

General statistical explanation: confidence intervals depend on the analysis method and assumptions. They are not guarantees that a parameter lies in the particular interval, and they do not show the distribution of individual participant responses. Their interpretation also depends on the target population and analysis set. Here, the primary analysis is described as per-protocol, so its precision pertains to that analysis population as defined in the registry.

For a non-inferiority conclusion, precision must be judged in relation to the margin. Because the margin is not reported in the registry record, the sponsor’s intervals cannot be independently evaluated against that criterion. The close point estimates are not, on their own, proof that the underlying responses are equal or that every potentially relevant difference has been ruled out.

11. Analyses Reported by the Sponsor

The registry reports two sponsor analyses labeled as non-inferiority analyses. For VRVg-2 plus HRIG versus Verorab plus HRIG, the reported difference in percentage is -0.42, with a 95% confidence interval from -2.33 to 4.35. For VRVg-2 plus HRIG versus Imovax Rabies plus HRIG, the reported difference is 0.86, with a 95% confidence interval from -1.32 to 6.50. The registry does not report a p-value for either sponsor analysis, nor does it state the non-inferiority margin or analysis method.

The unadjusted re-calculations from posted summary statistics give p=0.57 and p=0.403 for the two comparisons. Those calculations are not the sponsor’s non-inferiority analyses: they are unadjusted re-calculations and do not test a stated margin. Their p-values therefore cannot replace or validate the sponsor’s decision rule. The record does not flag a discrepancy between the posted results and the re-calculations, but the analyses answer different statistical questions.

The sponsor’s own analysis is the relevant reported analysis for the stated non-inferiority objective. The available registry entries are not sufficient to reconstruct its method or independently confirm a non-inferiority conclusion. The results can be reported as the sponsor presents them, while noting that the margin and method are not reported in the registry record.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary non-inferiority endpoint and multiple secondary outcome types, including antibody threshold percentages, geometric mean titers, geometric mean titer ratios, virus neutralization categories, and adverse-event counts. It also describes immune-response assessments at Day 14, Day 28 and Day 42. The number of outcome measures is listed as 11, and the registry reports 10 secondary outcomes posted.

There are two sponsor-reported primary comparisons: VRVg-2 plus HRIG versus Verorab plus HRIG, and VRVg-2 plus HRIG versus Imovax Rabies plus HRIG. The record does not report whether or how multiplicity across these comparisons, secondary outcomes, or assessment times was addressed. This matters because interpreting several tests or intervals together can differ from interpreting each one in isolation. No adjustment method should be assumed from the information recorded.

The primary analysis population is described as the per-protocol analysis set. It included participants who received at least one dose and excluded those with protocol deviations or who met per-protocol exclusion criteria. The registry does not report detailed participant flow into and out of this set, the number excluded for each reason, or a comparison with an intention-to-treat analysis. The endpoint counts are reported, but the full accounting of missing endpoint measurements is not reported in the registry record.

The record does not report how missing outcome data were handled, whether any imputation was used, or whether sensitivity analyses assessed the effect of missingness. It also does not report the non-inferiority margin, the sponsor analysis method, or a sample-size justification. Interim analyses and stopping rules are not reported in the registry record. These omissions limit independent assessment of the analysis plan and how robust the conclusions are to alternative assumptions.

13. 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: meningioma benign (1.0%), ischaemic stroke (1.0%), endometriosis (1.0%), general physical health deterioration (0.9%), erysipelas (0.9%).

The registry reports serious adverse-event counts for each group alongside the number at risk: 4 of 320 in Group 1, 4 of 107 in Group 2, 0 of 107 in Group 3, and 3 of 104 in Group 4. Deaths are recorded as 0 in each group. These are counts as posted; they should not be converted into a comparison of risk without accounting for the study’s safety definitions, observation period, and analysis details.

The record also lists terms among the most frequent serious events, including Meningioma Benign, Ischaemic Stroke and Endometriosis at 1.0%, and General Physical Health Deterioration, Erysipelas and Dyskinesia at 0.9%. The registry record does not establish from these listings whether an event was caused by a vaccine or HRIG. A serious adverse event is not automatically an adverse reaction attributable to treatment.

The different group sizes and the small number of serious events make descriptive counts more appropriate than strong comparative claims. The registry also reports counts in another adverse-event category, but the displayed figures alone do not provide a complete account of event timing, severity, relatedness, or duration. Absence of a recorded death does not mean that no other serious event occurred.

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

The primary endpoint results show high proportions of participants reaching the RVNA threshold at Day 28 across the three HRIG groups. The reported percentages are 99.6% for VRVg-2 plus HRIG, 100.0% for Verorab plus HRIG, and 98.7% for Imovax Rabies plus HRIG. The sponsor’s reported contrasts and confidence intervals provide the registry’s non-inferiority results, but the margin and method needed to independently assess that claim are not reported in the registry record.

These findings concern an antibody measurement in healthy adults following the study’s simulated post-exposure regimen. They do not report the incidence of rabies, outcomes after an actual exposure, or direct clinical effectiveness. Meeting an antibody threshold is the endpoint chosen for this trial, not a clinical outcome observed in the registry results.

The results also do not show that the three vaccine regimens are identical. Close observed percentages and intervals that include zero do not prove equality. A lack of statistical evidence for a difference is distinct from evidence that any difference is sufficiently small under a prespecified non-inferiority margin. Since the margin is not reported, that distinction is especially important when interpreting the sponsor’s label for its analyses.

The trial includes a VRVg-2 group without HRIG, but the primary non-inferiority comparisons concern VRVg-2 with HRIG against the two reference vaccines with HRIG. The results for the HRIG groups should not be generalized to the separate regimen or to populations not described in the registry. The study’s analysis set is per-protocol, and detailed missing-data handling and the full participant accounting are not reported.

Safety counts are descriptive. The registry records serious adverse events in the groups and no deaths, but these figures do not establish causation or quantify all dimensions of safety. Overall, the posted data characterize selected immunogenicity and safety outcomes for the trial groups; they do not answer every clinical question about rabies prevention.

15. Why This Trial Matters Statistically

This trial illustrates why a high response percentage is not enough to assess a non-inferiority question. The primary outcome is binary and approaches the upper bound in all three HRIG groups. The group percentages are therefore close, but the inferential question still depends on the contrast estimate, its uncertainty, and the prespecified margin. Without the margin, readers cannot independently check whether the sponsor’s interval met its decision criterion.

It also illustrates how two kinds of analysis can coexist without being interchangeable. The sponsor reports analyses labeled non-inferiority, while simple unadjusted re-calculations from the posted summary statistics provide p-values for ordinary contrasts. A p-value for a difference test does not substitute for a non-inferiority evaluation. Clear reporting of the margin, method, analysis population, and multiplicity approach would make the relationship between these analyses easier to assess.

The endpoint itself demonstrates a trade-off in statistical summaries. A threshold percentage is straightforward to understand, but it groups all titers above the threshold together and does not preserve the full variation in measured titers. The registry’s secondary outcomes include geometric mean titers and other immune-response summaries, which address different aspects of the response.

Finally, the record distinguishes enrollment from the analysis counts reported for the primary endpoint and defines a per-protocol set. Understanding who contributed to an estimate is essential for interpreting its precision and scope. The registry’s incomplete reporting of missing data, participant flow, and sample-size justification limits the extent to which readers can independently assess those issues.

16. 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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

Several trial-specific reporting gaps constrain interpretation. The registry labels the two sponsor comparisons as non-inferiority analyses but does not report the non-inferiority margin or the method used. Without those details, the sponsor’s intervals cannot be checked against the decision criterion, and the unadjusted re-calculations cannot supply that missing information.

The primary analysis is described as per-protocol, but the record does not give a detailed flow of participants into the analysis set or the number excluded for each reason. Although the endpoint analysis counts are posted, handling of missing endpoint data and sensitivity analyses are not reported in the registry record. The sample-size justification, interim analyses, and statistical plan details are also not reported.

Multiplicity procedures are not reported despite the two primary comparisons and multiple secondary outcomes and assessment times. The registry does not explain whether any adjustment or hierarchical testing approach was used. The available summary therefore does not support an independent assessment of family-wise error or the interpretation of multiple secondary findings.

There is also a timing detail worth reading carefully: the primary endpoint description defines Day 28 as 14 days after the fourth injection, while the intervention descriptions list injections on Days 0, 3, 7, 14 and 28. The registry separately describes Day 42 as 14 days after the last injection. The record does not explain this wording or reconcile it, so the visit and dose timing should be understood as stated rather than inferred beyond the descriptions.

Finally, the study concerns healthy adults in France and a simulated post-exposure regimen. The endpoint is antibody response rather than observed rabies disease. The registry record does not establish the applicability of these results to other populations or to outcomes after an actual exposure.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Study of Purified Vero Rabies Vaccine Compared With Two Reference Rabies Vaccines in a Simulated Post-Exposure Regimen in Adults — NCT03965962.
• Registry record last updated 2025-09-18; status last verified 2025-09.
• Results first posted 2022-03-11.
• 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.

20. Record Summary

NCT03965962 is a completed randomized, quadruple-masked, parallel-group Phase 3 study in healthy adults in France, with enrollment listed as 640. It compared VRVg-2 plus HRIG with Verorab plus HRIG and Imovax Rabies plus HRIG for the primary Day 28 RVNA threshold endpoint. A fourth group received VRVg-2 without HRIG.

Posted percentages reaching the threshold were 99.6%, 100.0% and 98.7% in the three HRIG groups, respectively. The sponsor reported two non-inferiority analyses; unadjusted re-calculations give p=0.57 and p=0.403. The non-inferiority margin and analysis method are not reported in the registry record.

The registry also reports safety counts, including serious adverse events and deaths. Missing-data handling, multiplicity procedures, detailed participant flow, and the trial’s sample-size justification are not reported. The results describe immunogenicity and safety outcomes in the analyzed trial participants, not clinical rabies outcomes after actual exposure.