Efficacy of MVA-NP+M1 in the Influenza H3N2 Human Challenge Model: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy of MVA-NP+M1 in the Influenza H3N2 Human Challenge Model (NCT03883113), a randomized phase 2 study. The primary outcome is Degree of Nasopharyngeal Viral Shedding as Determined by Quantitative Polymerase Chain Reaction qPCR.

2. Clinical Question

This is a question about a laboratory-measured response under a controlled human challenge, not a direct estimate of how vaccination changes influenza outcomes in everyday exposure settings. The study also recorded outcomes relating to virologically confirmed influenza-like illness, the attack rate of the challenge agent, and viral shedding. The registry lists 22 outcome measures and 17 posted secondary outcomes. Those additional measures give context to the trial’s broader scope, but the primary statistical comparison concerns the qRT-PCR viral-shedding endpoint.

3. Trial Design

NCT03883113 is a Phase 2 randomized, parallel-group study sponsored by Barinthus Biotherapeutics. The registry describes it as a single-center study in healthy adult volunteers and lists influenza as the condition. Its brief summary calls the study double blind, while the masking field records quadruple masking. The record does not explain the distinction between those descriptions.

Participants were assigned to MVA-NP+M1 or saline placebo during vaccination, and both groups received H3N2 challenge virus. The intervention descriptions specify intramuscular vaccination and a nasal-spray challenge. The challenge virus is identified as H3N2 (A/Belgium/2417/2015). The parallel design compares groups rather than describing a crossover in which participants receive both vaccination assignments.

The enrollment field records 145. The primary endpoint results are reported for several analysis labels: intention-to-treat (ITT), per-protocol (PP), and challenge. For the ITT comparison, the registry lists 71 for MVA-NP+M1 and 47 for placebo. For the PP comparison, it lists 70 and 47. The challenge comparison lists 71 and 47. The record does not define the membership rules for these analysis populations, explain why the analysis counts differ, or describe the allocation ratio.

The study status is completed. The registry lists a start date of 2019-06-03 and a primary completion date of 2019-12-16. The posted primary endpoint concerns measurements taken during the challenge phase, rather than a long-term clinical follow-up outcome. Details of the randomization procedure, concealment, and the operational meaning of the masking designation are not reported in the registry record.

4. Participant Flow and Accounting

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

Before assignment: Eight-hundred and nineteen (819) subjects were screened. One hundred and forty-five (145) subjects were actually enrolled and vaccinated. The study consisted of an outpatient vaccination phase and at least 6 weeks later an inpatient challenge phase.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: End point values

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 Degree of Nasopharyngeal Viral Shedding as Determined by Quantitative Polymerase Chain Reaction qPCR, measured over Throughout 9 days (Day2, Day3, Day4, Day5, Day6, Day7, Day8, Day9, Day10) after viral Inoculation (Day1) of the challenge phase. Nasal swabs taken twice a day (b.i.d) at least 8 hours apart..

The primary endpoint measures nasopharyngeal viral shedding using quantitative real-time polymerase chain reaction, or qRT-PCR. The registry describes the measure as viral area under the curve (vAUC): log viral particles per ml are plotted over time, and the area is calculated using the trapezoidal rule. The reported unit is hour*Log10 Viral Particles/ ml. The endpoint therefore combines the measured amount of virus and its course over the specified observation period into one continuous summary measure.

Measurements were scheduled throughout 9 days, from Day2 through Day10, after viral inoculation on Day1. Nasal swabs were taken twice a day, at least 8 hours apart. The registry identifies the reported result as an endpoint value and gives least-squares means with 95% confidence intervals.

In general, a repeated-measure viral-shedding summary can represent more of the measured challenge response than a result taken at a single time point. It can account for both the level and duration of detected shedding over the observation window, as reflected in the area-under-the-curve construction. It remains a laboratory endpoint: it does not by itself state whether a participant experienced symptoms, nor does it directly describe outcomes after naturally acquired influenza exposure. The registry’s primary endpoint is specifically the qRT-PCR vAUC measure, while other listed outcomes include illness and alternative viral-shedding measures.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is continuous: each participant’s viral-shedding response is summarized as a vAUC value. The registry reports a least-squares mean for each group and a 95% confidence interval. In general, a least-squares mean is a model-based group mean, sometimes called an adjusted mean. The registry does not describe the model, any covariates, or whether and how the reported means were adjusted. It is therefore not possible to infer from the record which features of the analysis produced these estimates.

What does the confidence interval communicate?

In general, a confidence interval expresses the precision of an estimate under a particular statistical procedure and its assumptions. A narrower interval indicates greater precision than a wider interval for the same type of estimate. The intervals in this record quantify uncertainty around each group’s least-squares mean. They are not intervals for individual participants, and they are not a direct interval estimate of the difference between groups. No confidence interval for that contrast is reported in the registry record.

The means and intervals should be interpreted in the endpoint’s units. The endpoint is measured in hour*Log10 Viral Particles/ ml and is derived from log viral particles per ml plotted over time. The trapezoidal rule approximates the area under a curve by connecting observed time points and summing the areas between them. Since the endpoint combines readings across the stated observation period, the value is not a single swab result or a direct count of symptomatic illness.

What does the Wilcoxon comparison test?

The registry reports Wilcoxon (Mann-Whitney) comparisons for the ITT, PP, and challenge-labeled groups. In general, this is a rank-based method that compares the ordering of observations in two groups. It does not require the test to be described as a comparison of arithmetic means. Interpretation as a difference in medians requires additional assumptions about the shapes of the group distributions. The record does not give participant-level endpoint values or distribution summaries that would establish those assumptions.

The group results are presented as least-squares means and confidence intervals, while the reported p-values come from the Wilcoxon method. Those are distinct statistical summaries. A group mean describes a central summary on the outcome scale, while a rank-based test evaluates the relative ordering of observations. Without the raw data, a reported contrast estimate, and further analysis details, the p-values cannot be used to reconstruct a mean difference, a median difference, or the distribution of individual responses.

Why does the endpoint construction matter?

The vAUC calculation uses repeated measurements after challenge. In general, an area-under-the-curve endpoint can reflect both the level and persistence of a measured signal during its observation window. Its interpretation depends on the sampling schedule, the measurement scale, and how missing or below-detection observations are handled. The registry describes the sampling schedule and trapezoidal method, but handling of missing samples, assay limits, and other data-processing decisions is not reported in the registry record.

The endpoint uses log viral-particle measurements, so it should not be casually interpreted as a simple raw count accumulated over time. The named unit combines hours with a log-scale concentration. The clinical meaning of a particular vAUC value, including a threshold that would matter for symptoms or transmission, is not reported in the registry record.

How should the p-values be read?

In general, a p-value is calculated under a null hypothesis and a specified statistical model or test procedure. It indicates how unusual data at least as inconsistent with that null would be under those assumptions. It does not express the probability that a result occurred by chance, the probability that a treatment is effective, or the magnitude of a group difference. The record gives p-values of 0.171, 0.165, and 0.171 for the ITT, PP, and challenge comparisons, respectively, but does not report the test’s sidedness or the prespecified decision threshold for the primary analysis.

The registry lists 15 analyses and 22 outcome measures, along with 17 posted secondary outcomes. When many endpoints or comparisons are examined, the chance of at least one apparently unusual result can increase. In general, multiplicity procedures can address that issue, but the registry record does not report whether any adjustment or hierarchy was planned or used. The individual p-values therefore should not be assumed to account for all analyses across outcomes.

The result labels also matter. ITT, PP, and challenge indicate distinct analysis labels, but the record does not define their inclusion rules. The posted group counts vary between the ITT and PP labels for the MVA-NP+M1 group, while the placebo count is 47 in each listed primary comparison. Those labels and counts should be retained when describing the result; the record does not give enough information to infer exclusions or the consequences of population selection.

10. Analyses Reported by the Sponsor

The registry reports the trial’s own Wilcoxon (Mann-Whitney) analyses for the primary endpoint. The ITT-labeled and challenge-labeled comparisons have a reported p-value of 0.171, and the PP-labeled comparison has a reported p-value of 0.165. These are the inferential results recorded for those group comparisons.

A separate recalculated two-arm comparison is not available for this continuous endpoint in the posted summaries: the registry record indicates that values are missing or not estimable for such a comparison. The group least-squares means and their confidence intervals do not supply the missing test statistic or a confidence interval for the contrast. Accordingly, there is no recalculated comparison to place alongside the registry’s reported tests, and no discrepancy is flagged.

The sponsor’s reported analysis should be distinguished from any calculation based only on group-level summaries. The record does not provide the participant-level values, model details, or enough information to reproduce the reported Wilcoxon tests independently. The posted p-values can therefore be described as the registry’s reported analyses, but they cannot be independently verified from the group means and confidence intervals alone.

11. Multiplicity, Missing Data and Analysis Populations

The registry presents results under ITT, PP, and challenge analysis labels, but does not define the membership rules for those populations. It does not explain the difference between the listed counts for the MVA-NP+M1 group in the ITT and PP comparisons, nor does it state how participants with missing swabs or other incomplete endpoint data were handled. The missing-data methods are not reported in the registry record.

The primary endpoint is based on scheduled repeated nasal swabs. In general, missing observations can affect an area-under-the-curve calculation because the observed time points contribute to the estimated trajectory. Decisions about interpolation, partial follow-up, exclusions, and assay values below detection limits can influence the endpoint or the analysis population. The registry describes the sampling schedule and trapezoidal calculation but does not state how these issues were managed.

The record lists 22 outcome measures, 17 posted secondary outcomes, and 15 analyses. These figures establish that the posted results extend beyond a single endpoint and comparison. They do not establish how many tests were prespecified or whether the reported analyses cover every planned test. The statistical plan, endpoint hierarchy, and any multiplicity adjustment are not reported in the registry record. Consequently, the primary p-value should be read as a reported test result, not presumed to be adjusted for the other outcomes or analyses.

The record also does not report a sample-size justification, the target effect used in planning, or whether interim analyses were conducted. Those details should not be inferred from the observed group counts or from planning illustrations based on an observed effect. In general, planning calculations describe assumptions for a future study; they are not observed power and do not revise the uncertainty of the results already reported. The registry record does not provide enough detail to assess how the enrollment target was chosen or whether the completed analysis followed a prespecified statistical plan.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: acute psychosis (2.1%), foetal death (1.1%).

The registry reports serious adverse event counts separately for vaccination and challenge phases. In the vaccination phase, the MVA-NP+M1 group lists 87 at risk and 1 affected; the saline placebo group lists 58 at risk and 0 affected. In the challenge phase, MVA-NP+M1 lists 71 at risk and 0 affected, while saline placebo lists 47 at risk and 1 affected. The record lists 0 deaths in each of these four group entries.

The registry also names Acute Psychosis at 2.1% and Foetal death at 1.1% among the top serious event terms. It does not connect those terms to a particular phase or treatment group in the information, so they should not be assigned to one. The event-term percentages should also not be treated as group-specific rates without a stated denominator and attribution.

In general, serious adverse event counts describe events meeting the study’s serious-event reporting criteria; they do not by themselves establish causation. The at-risk counts differ across phase and analysis entries, so each count must be read with its own group and phase label. The registry record does not provide event narratives, timing, relatedness assessments, or enough detail to reconcile the named event terms with the group-level counts. These entries are a descriptive safety account, not a standalone comparison of safety risk.

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

The posted results describe a controlled H3N2 human challenge study and a laboratory measure of nasopharyngeal viral shedding. The ITT-labeled group means are 649.7 for MVA-NP+M1 and 726.1 for placebo, with the reported 95% confidence intervals around each mean. The registry’s Wilcoxon p-value for that comparison is 0.171. The PP and challenge results are also posted, with their own analysis labels and test results.

These summaries do not establish a causal effect on the primary endpoint by themselves. The registry reports group-level estimates and rank-based p-values but does not report an effect estimate or confidence interval for the between-group contrast. The record also does not describe enough of the statistical model, data handling, or analysis populations to reconstruct the result. The numerical group means and p-values should therefore be treated as summaries of the posted analysis, not as a complete account of the evidence.

Nor does a non-extreme p-value demonstrate that the groups are equivalent. Equivalence or non-inferiority requires a specified margin and an analysis designed to evaluate it; those details are not reported in the registry record. Conversely, the numerical ordering of the group means is not proof that one assignment caused a difference. The size and precision of a treatment contrast cannot be read directly from group means alone.

The endpoint is a challenge-phase virologic measure over the scheduled swabbing period. It does not directly answer whether vaccination changes the risk or severity of naturally acquired influenza in a broader population. The study is described as enrolling healthy adult volunteers, and the challenge model is a controlled setting. Generalizing beyond that population or setting would require evidence not contained in these endpoint summaries.

The record’s safety entries are also descriptive. Serious adverse events and deaths are listed for each phase and group, but the information does not establish attribution, comparative risk, or the full clinical context for the reported events. Overall, the result is best read as a limited report of primary endpoint summaries and tests from a specific challenge study, with important analytic and interpretive details not reported in the registry record.

14. Why This Trial Matters Statistically

This trial illustrates why the endpoint, statistical summary, and test need to be kept distinct. The primary outcome is a continuous vAUC measure derived from repeated qRT-PCR observations, but the group summaries are least-squares means while the reported comparison is rank-based. Those statistics answer related but not identical questions. A careful reading avoids treating the reported p-value as a test of the displayed mean difference.

The analysis labels also matter. ITT, PP, and challenge are not interchangeable labels, and their counts should be reported as recorded rather than merged. The registry gives different MVA-NP+M1 counts for ITT and PP, and does not define the populations. That gap limits interpretation of what changed between analyses and whether the same participants contributed to each summary.

The endpoint demonstrates the trade-off involved in summarizing repeated biological measurements. A vAUC can capture a pattern across time, but it also depends on sampling and data-processing choices. The registry identifies the observation schedule and trapezoidal method, while leaving missing-data handling and several model details unreported. This is a useful reminder that a named outcome and formula do not disclose every analytic decision.

Finally, a challenge study has a deliberately controlled exposure context and a laboratory-focused primary outcome. Its estimates should be described within that context. The posted findings are not a direct substitute for clinical outcome evidence in other settings, and the registry’s reported p-values do not by themselves settle effect size, precision, or practical importance.

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 leaves important details of the analysis unspecified. It does not report the model used to obtain least-squares means, any adjustment variables, the prespecified primary-analysis threshold, or a confidence interval for the between-group contrast. The Wilcoxon comparisons are reported with p-values, but the test’s sidedness and the assumptions used to interpret its result are not stated.

The ITT, PP, and challenge populations are named but not defined. The record does not explain the differences in listed counts, how missing swabs were handled, or how the endpoint was calculated when measurements were incomplete. It also does not describe handling of assay limits or other data-processing decisions that could affect a repeated-measure vAUC.

Multiplicity procedures, interim analyses, and a sample-size justification are not reported in the registry record. This limits assessment of how the primary analysis relates to the other listed outcomes and analyses, and how the enrollment target was selected. The registry lists secondary outcomes and analyses, but does not give enough information to determine which comparisons were prespecified or whether any adjustment was used.

The trial is described as a single-center challenge study in healthy adult volunteers. Its primary measure is qRT-PCR viral shedding during a defined post-inoculation window, rather than a long-term clinical outcome. These features define the scope of the posted result and limit generalization to populations or settings not described by the study. The record does not provide detailed information on participant characteristics, protocol deviations, or the context needed to interpret the reported serious adverse event terms by group.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Efficacy of MVA-NP+M1 in the Influenza H3N2 Human Challenge Model — NCT03883113.
• Registry record last updated 2021-03-18; status last verified 2021-03.
• Results first posted 2021-03-18.
• 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

NCT03883113 is a completed Phase 2, randomized, parallel-group study of MVA-NP+M1 versus saline placebo in healthy adult volunteers receiving H3N2 challenge virus. The registry identifies the primary endpoint as nasopharyngeal viral shedding measured by qRT-PCR vAUC throughout 9 days after inoculation, with least-squares means and 95% confidence intervals reported for ITT, PP, and challenge-labeled groups.

The ITT-labeled means are 649.7 for MVA-NP+M1 and 726.1 for placebo, with a reported Wilcoxon p-value of 0.171. The PP-labeled comparison reports means of 646.5 and 726.1 and p-value 0.165. The challenge-labeled comparison repeats the ITT summaries and p-value. No recalculated two-arm comparison is available from the posted continuous-endpoint summaries. Definitions of analysis populations, missing-data methods, multiplicity procedures, and key model details are not reported in the registry record.