Safety and Immunogenicity (Non-inferiority) of a Purified Chick Embryo Cell Vaccine Vaccine Administered in Two Different Schedules (Conventional Versus Abbreviated Schedule): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Immunogenicity (Non-inferiority) of a Purified Chick Embryo Cell Vaccine Vaccine Administered in Two Different Schedules (Conventional Versus Abbreviated Schedule) (NCT00825305), a randomized phase 3 study. The primary outcome is Rabies Virus Neutralizing Antibody Concentrations on Day 14..

2. Clinical Question

The clinical question is whether the abbreviated Zagreb schedule of purified chick embryo cell vaccine (PCECV) produces day-14 rabies virus neutralizing antibody concentrations that are not unacceptably lower than those associated with the conventional Essen schedule. The registry describes the comparison as a non-inferiority study, but the non-inferiority margin is not reported in the registry record. Without that margin, the reported estimate and confidence interval cannot by themselves establish whether the prespecified non-inferiority criterion was met.

This is an immunogenicity question, not a comparison of clinical rabies outcomes. The posted primary endpoint is antibody concentration, expressed on a log2 scale, and the registry also lists secondary antibody concentration and seroconversion outcomes at other time points. Numerical secondary outcome results are not reported in the registry record. The available primary result therefore addresses a specific laboratory measure at a specified follow-up time; it does not, on its own, describe every immune response or safety outcome in the study.

3. Trial Design

The registry identifies this as a phase III, randomized, open-label, single-center study with a parallel-group design. The two groups received different PCECV schedules: abbreviated Zagreb or conventional Essen. Because masking is listed as none, the study was open label. The registry lists enrollment of 825 and a completed status. The recorded start is 2008-11 and primary completion is 2008-12.

The primary endpoint is rabies virus neutralizing antibody concentration on day 14. The reported immunogenicity population is not the full enrolled group: the registry says that, due to Chinese registration requirements, blood samples were drawn from only a subset of subjects, and that the per protocol population of this subset was analyzed. The posted day-14 groups contain 69 participants for Zagreb and 70 for Essen. These endpoint-specific analysis counts should not be treated as the trial's full enrollment or as the denominators for the safety results.

The safety record gives different at-risk counts: 549 for Zagreb and 275 for Essen. The registry does not explain in this record why the safety and immunogenicity counts differ, so they should be read as separate analysis populations rather than reconciled by assumption. Further details about participant flow, exclusions, and the reasons particular subjects were or were not included in each analysis are not reported in the registry record.

Random allocation supports a planned comparison between schedules, but the registry record does not report the allocation sequence method or concealment procedures. It does not report a detailed protocol, the planned non-inferiority margin, or a sample-size justification. The primary completion date indicates when the primary study period was recorded as complete; it does not provide information about the timing or conduct of analyses beyond what is stated in the 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: All enrolled subjects were randomized, but one subject in the Zagreb (2-1-1) group was not vaccinated.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Due to the Chinese registration requirements, only a subset of subjects bloodsamples were drawn and immunogenicity was evaluated. The per protocol population of this subset was analyzed.

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 Rabies Virus Neutralizing Antibody Concentrations on Day 14., measured over 14 days.

The primary endpoint is rabies virus neutralizing antibody concentration on day 14, reported in IU/mL but analyzed and presented on a log2 scale. Neutralizing antibodies are a laboratory measure of the antibody response to the vaccine. A day-specific concentration gives a quantitative outcome at a defined follow-up point and permits a comparison of the distributions or means between the schedules at that time.

The registry reports log-scale mean values of 3.1 for Zagreb and 3.3 for Essen, with respective 95% confidence intervals of 2.73 to 3.47 and 2.82 to 3.79. The registry notes that results are on the log2 scale and that returning to the original geometric, or multiplicative, scale requires raising the reported numbers to the basis of 2. No original-scale values are stated in the record, and none are calculated.

Using a log scale is common for concentration measures that may vary multiplicatively. In general, transforming a measure can make comparisons more suitable for statistical analysis when changes are naturally interpreted proportionally rather than as fixed additive amounts. A log-scale result is not itself an original-scale concentration, so it should not be read as a direct IU/mL value without the stated back-transformation.

Antibody concentration is an immunogenicity endpoint, not a direct count of clinical disease events. Its relevance is that it measures a vaccine-associated immune response. The registry record does not report the basis for selecting day 14 as the primary time point or the threshold, if any, used to define an acceptable response for the primary non-inferiority question.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical ideas help interpret the registry result. The explanations below are general unless they describe a specific feature of the record.

What does the log2 scale mean?

The primary outcome is a concentration measured in IU/mL, but the reported values are on a log2 scale. A logarithmic transformation expresses multiplicative changes in a different form: equal steps on a log scale correspond to equal multiplicative changes on the original scale. This can be useful when concentration measurements are skewed or when relative rather than absolute differences are of interest. The registry explicitly says that the original geometric scale is obtained by raising numbers to the basis of 2. The values reported for the groups and their comparison should therefore be read as log-scale results, not as direct concentrations in IU/mL.

The analysis description labels the comparison as a “ratio of log2 mean.” The reported estimate and interval are retained in that form rather than being converted into another contrast. The record does not give further mathematical detail about that parameter, so its exact construction cannot be clarified from the registry entry alone. In general, an analysis should be interpreted using the definition of its estimand—the quantity it was designed to estimate—and the scale on which it is reported.

What does the confidence interval convey?

The registry reports 95% confidence intervals for both group summaries and the between-schedule analysis. In general, a confidence interval conveys the uncertainty around an estimate under the statistical model and sampling process. A narrower interval indicates greater precision than a wider interval on the same scale, while a wide interval leaves a broader range of values compatible with the data and method. A confidence interval is not a guarantee that the true value lies within its endpoints, and it is not a probability statement about the effect after observing the data.

For the group summaries, the reported intervals are 2.73 to 3.47 for Zagreb and 2.82 to 3.79 for Essen. For the sponsor's comparison, the interval is -0.81 to 0.4. The comparison interval is the relevant reported uncertainty measure for the schedule contrast; the separate group intervals do not replace it. The registry does not report a p-value for the sponsor comparison.

What does ANOVA contribute?

The sponsor's analysis is identified as ANOVA. In general, analysis of variance is a model-based method for comparing group means. Its interpretation depends on how the outcome and contrast are defined and on the assumptions used by the analysis. With a transformed concentration endpoint, the reported mean comparison applies to the transformed scale unless the analysis specifies otherwise. The registry record does not provide details about model diagnostics, covariate adjustment, or sensitivity analyses, so those elements cannot be assessed from the posted entry.

Randomization creates a design for comparing the assigned schedules, but it does not eliminate the need to describe the analysis population and methods. The registry says that only a subset had blood samples drawn for immunogenicity evaluation and that the per protocol population of that subset was analyzed. It does not state the full criteria for the per protocol population or how missing measurements were handled. Those details matter because an analysis restricted to participants meeting specified conditions can differ in interpretation from an analysis of all randomized participants.

Why does non-inferiority need a margin?

In general, a non-inferiority study asks whether a new or alternative approach is not worse than a comparator by more than a prespecified amount. The margin defines what degree of difference is considered acceptable for the study's statistical question. The estimate and its confidence interval are assessed relative to that margin and the chosen direction of comparison. A finding that an interval includes zero, or that a conventional difference test is not statistically significant, does not by itself demonstrate non-inferiority. The registry describes this study as non-inferiority but does not report the margin or enough detail about the decision rule. That omission prevents confirmation of the non-inferiority conclusion from the registry result alone.

How should multiplicity and missing results be considered?

The registry lists secondary antibody outcomes at additional time points and seroconversion percentages at several time points, as well as multiple outcome measures and analyses. When multiple hypotheses are tested, the chance of at least one false-positive finding can increase if the analyses are treated as independent confirmatory tests. General approaches include prespecifying a hierarchy or adjusting inference, but the registry record does not report a multiplicity strategy. No conclusion about whether such a strategy was used should be inferred from that absence.

Numerical results for the listed secondary outcomes are not reported in the registry record. The record also identifies values as missing or not estimable for a two-arm comparison for another endpoint. These gaps mean those outcomes cannot be interpreted from the posted results. They do not establish that the outcomes were unmeasured or that a particular result occurred.

Why are the analysis counts different?

The primary endpoint has 69 participants in the Zagreb group and 70 in the Essen group, while safety at-risk counts are 549 and 275, respectively. The registry explicitly attributes the restricted immunogenicity population to blood sampling in a subset under Chinese registration requirements, but does not explain the complete relationship between the safety and immunogenicity counts. In general, different outcomes may have different eligible populations and denominators. Readers should use the counts attached to each analysis, rather than substituting the overall enrollment or one outcome's denominators for another.

There are no sample-size planning results reported in the registry record for interpreting the primary contrast. General planning calculations are prospective illustrations based on assumptions, not observed power and not evidence that a study did or did not detect an effect. For this result, the reported estimate, uncertainty, analysis population, and missing non-inferiority margin are the key features for statistical interpretation.

10. Analyses Reported by the Sponsor

The registry's sponsor-reported comparison is an ANOVA analysis described as a ratio of log2 means. It reports an estimate of -0.2 and a 95% confidence interval from -0.81 to 0.4. No p-value for that analysis is reported in the registry record. The group summaries are also reported on the log2 scale, with values of 3.1 for Zagreb and 3.3 for Essen.

A separate recalculated contrast is not reported in the registry record. Accordingly, there is no independently stated recalculated estimate to compare numerically with the sponsor's analysis. The sponsor result should be read as the posted inferential comparison, not as a recalculation. The registry does not flag a discrepancy between the sponsor analysis and another comparison.

The sponsor estimate and its confidence interval communicate the reported contrast and its precision, but they do not resolve the non-inferiority question without the margin. That margin, the detailed model specification, and the analysis decision rule are not reported in the registry record. The result should therefore be described in the terms and scale used by the registry, without inferring an unreported criterion or converting the result to a different effect measure.

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies a primary day-14 antibody concentration outcome and lists secondary antibody concentration and seroconversion outcomes. The secondary outcomes include antibody concentrations on day 7 and day 42, and the percentages meeting the stated threshold of equal to or above 0.5 IU/ml on days 7, 14 and 42. Numerical results for these secondary outcomes are not reported in the registry record. The registry also notes values missing or not estimable for a two-arm comparison for another endpoint, without giving a numerical result.

When a study evaluates several outcomes or time points, multiplicity becomes relevant if multiple tests are used to support confirmatory claims. In general, testing many hypotheses without an adjustment can increase the chance that at least one result appears statistically notable by chance. A prespecified hierarchy or an adjustment can address that concern, depending on the study's aims. The registry record does not report whether a hierarchy, adjustment, or other multiplicity strategy was planned or used. No particular strategy should be assumed.

Missingness is also important. The registry states that only a subset of subjects had blood samples drawn and that the per protocol population of that subset was analyzed for immunogenicity. It does not report why individual samples were unavailable, how missing values were handled, whether imputation was used, or whether sensitivity analyses assessed the impact of missing data. Those details are not reported in the registry record. The available record therefore supports identifying the analyzed subset, but not evaluating the assumptions underlying its selection or the possible effect of missing measurements.

The primary immunogenicity counts are 69 for Zagreb and 70 for Essen. The safety at-risk counts are 549 and 275. These are outcome-specific denominators; the record does not fully explain their relationship. The registry lists enrollment of 825, but that total should not be substituted for either endpoint-specific analysis population. Participant flow and detailed reasons for inclusion or exclusion are not reported in the registry record.

Finally, the non-inferiority margin and the statistical decision rule are not reported. This is particularly consequential for a non-inferiority interpretation, because the reported confidence interval must be judged against a prespecified boundary. Without that boundary, the posted primary comparison can be described, but the registry information is insufficient to verify the formal non-inferiority conclusion.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

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: injury of soft tissue in the leg and shoulder (0.4%).

The registry reports serious adverse event counts of 0 among 549 at risk in the abbreviated Zagreb group and 1 among 275 at risk in the conventional Essen group. It lists “Injury of soft tissue in the leg and shoulder” as the top serious term, at 0.4%. The record also reports 52 other adverse events among 549 at risk for Zagreb and 26 among 275 at risk for Essen. The safety denominators are distinct from the smaller immunogenicity analysis groups.

These are descriptive counts as posted. They should not be read as a complete comparative safety analysis: the record does not provide a statistical comparison, confidence interval, or detailed timing and duration information for these events. The listed serious term is reported at 0.4%; the record does not establish from that listing alone how the term relates to each group's serious-event count. The deaths entries are not populated, so deaths are not reported in the registry record rather than being established as zero.

In general, adverse-event counts are interpreted in relation to their definitions, observation periods, denominators, and methods for collecting and classifying events. Those details are not fully reported in the registry entry. The figures describe what is posted, but do not by themselves quantify the comparative safety of the schedules.

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

The result means that the registry records a randomized, parallel-group comparison of two PCECV schedules and posts a day-14 immunogenicity analysis for a per protocol subset. The Zagreb group summary is 3.1 on the log2 scale, with a 95% confidence interval of 2.73 to 3.47; the Essen summary is 3.3, with a 95% confidence interval of 2.82 to 3.79. The sponsor's reported ANOVA comparison is -0.2, with a 95% confidence interval from -0.81 to 0.4.

Those figures do not establish whether the abbreviated schedule met the study's non-inferiority criterion. The registry calls the study non-inferiority, but the margin and decision rule are not reported in the registry record. The comparison interval must be interpreted relative to that margin; a confidence interval on its own cannot answer the formal non-inferiority question when the required boundary is unavailable. Nor should the result be turned into a claim of equivalence: non-inferiority and equivalence are distinct statistical questions and require prespecified criteria.

The endpoint is a laboratory measurement of antibody concentration at one follow-up time. It is not a clinical disease outcome, and the record does not report comparative counts of rabies illness. The secondary immunogenicity results are also not reported numerically. Therefore, this primary finding should not be extended to clinical protection, to unreported time points, or to other populations beyond the healthy adults in China described in the registry.

The safety information is similarly limited to posted counts and a listed serious-event term. The registry does not provide a full comparative safety analysis or enough detail to interpret the counts as a complete assessment of risk. The appropriate statistical reading is narrow: report the endpoint, scale, analysis population, estimate, interval, and limits of the registry information without converting them into a treatment recommendation or a broader efficacy claim.

14. Why This Trial Matters Statistically

This trial illustrates why the statistical question must be read together with the study design and endpoint definition. The comparison is framed as non-inferiority, so the point estimate alone is not the decision criterion. The confidence interval needs to be evaluated against a prespecified margin, and that margin is not reported in the registry record. Without it, the result can be accurately described but its formal non-inferiority status cannot be independently assessed from the registry entry.

It also demonstrates the importance of the analysis population. Although the registry lists enrollment of 825, the primary immunogenicity analysis uses a per protocol subset with 69 Zagreb and 70 Essen participants. The record explains that only a subset had blood samples drawn for immunogenicity evaluation under Chinese registration requirements. Safety counts use different at-risk denominators. Keeping these populations separate prevents the full enrollment figure from being mistaken for the number contributing to a particular endpoint.

The endpoint scale matters as well. The antibody summaries are presented on a log2 scale, while the stated measurement unit is IU/mL. That distinction affects what the reported means represent and how the comparison should be communicated. The registry's explicit scale note helps readers avoid reading transformed values as original-scale concentrations.

Finally, the record lists secondary outcomes but does not report their numerical results or a multiplicity strategy. That makes clear why a primary result cannot stand in for all outcome analyses. A careful statistical reading distinguishes what is estimated, what remains uncertain, which analyses are available, and which design details are absent. These distinctions are central to interpreting a non-inferiority trial without overstating what the posted result establishes.

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.

The registry record has important limitations for interpreting the primary result. Most notably, it does not report the non-inferiority margin or the decision rule. Because those are necessary to assess the stated non-inferiority question, the reported comparison cannot establish from the registry entry alone whether the formal criterion was met.

The analysis population is a per protocol subset of participants with blood samples drawn for immunogenicity assessment. The registry attributes this subset to Chinese registration requirements, but does not provide complete participant flow, the per protocol criteria, or details about missing samples. The primary analysis counts of 69 and 70 are substantially different in purpose from the safety at-risk counts of 549 and 275; the record does not fully explain the relationship between these sets of counts.

The sponsor comparison is labeled as a ratio of log2 means, but further details defining the contrast and its model are not reported. The registry does not report diagnostics, covariate adjustment, sensitivity analyses, or the handling of missing endpoint values. It also provides no p-value for the posted primary comparison. These omissions limit evaluation of the analysis beyond the estimate and confidence interval as stated.

Numerical results for the listed secondary antibody outcomes are not reported in the registry record, and the record notes another endpoint with values missing or not estimable for a two-arm comparison. A multiplicity approach is not reported. The available record therefore does not permit a complete account of all outcome findings or how multiple analyses were handled.

For safety, the registry posts event counts and a serious-event term but does not report a comparative statistical analysis or detailed event observation and classification methods. Death entries are not populated. The record's single-center, open-label design is also relevant context, while further operational details such as allocation concealment are not reported. These limitations restrict interpretation to the posted summaries and prevent broader conclusions than those supported by the registry information.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Safety and Immunogenicity (Non-inferiority) of a Purified Chick Embryo Cell Vaccine Vaccine Administered in Two Different Schedules (Conventional Versus Abbreviated Schedule) — NCT00825305.
• Registry record last updated 2011-12-13; status last verified 2011-12.
• Results first posted 2010-05-19.
• 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

NCT00825305 is a completed phase III, randomized, open-label, single-center, parallel-group study comparing abbreviated Zagreb 2-1-1 and conventional Essen 1-1-1-1-1 PCECV schedules in healthy adults in China. Enrollment is listed as 825; the recorded start is 2008-11 and primary completion is 2008-12.

The primary endpoint is day-14 rabies virus neutralizing antibody concentration, reported on a log2 scale. The per protocol immunogenicity subset includes 69 Zagreb participants and 70 Essen participants. The sponsor-reported ANOVA comparison is -0.2, with a 95% confidence interval from -0.81 to 0.4. The non-inferiority margin is not reported in the registry record, so the formal non-inferiority conclusion cannot be verified from the posted result alone.

Secondary antibody outcomes are listed, but their numerical results are not reported. Safety counts use separate at-risk denominators. Further analysis details, participant flow, missing-data handling, and a multiplicity strategy are not reported in the registry record.