Immune Basis and Clinical Implications of Threshold-based Phenotypes of Peanut Allergy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Immune Basis and Clinical Implications of Threshold-based Phenotypes of Peanut Allergy (NCT03907397), a randomized phase 2 study. The primary outcome is Percentage of Children That Tolerate the Full Challenge.

2. Clinical Question

The study also described secondary clinical aims: sustained unresponsiveness, quality of life, and safety. Its broader aims included characterizing allergic responses by threshold and exposure and investigating immune and molecular features. Those mechanistic aims are part of the study description, but the posted results do not report mechanistic analyses.

3. Trial Design

The registry record describes a phase 2, randomized, parallel-group study with no masking. It lists 73 participants as enrolled: 38 in the Peanut Protein group and 35 in the Avoidance group. The status is completed. The recorded study start date is 2019-08-05, and the primary completion date is 2023-11-17.

The intervention was peanut protein ingestion; the comparator was continued peanut avoidance, described as standard care. The study description says participants who ingested peanut could begin at different amounts depending on their reaction threshold. The registry record does not give further detail about the allocation procedure, how the starting amounts were chosen, or the frequency and supervision of ingestion. Those details are not reported in the registry record.

The primary endpoint was assessed at 72 weeks and used an endpoint double-blind, placebo-controlled oral food challenge. The challenge tested whether a participant reached a dose at least 2 steps above the baseline challenge dose or 9043 mg of peanut protein. A secondary sustained-unresponsiveness outcome was assessed at 96 weeks. The acute allergic reaction outcome and other posted secondary outcomes also have a 96-week timeframe.

Although the endpoint challenge was double-blind and placebo-controlled, the overall study is listed as having no masking. The registry record does not report further information about who was masked for particular assessments. It also does not report a sample-size justification, the detailed randomization procedure, or the planned analysis populations. These omissions limit how much can be said about design details beyond the listed randomized, parallel-group structure.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

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 Percentage of Children That Tolerate the Full Challenge, measured over 72 weeks.

The primary endpoint is the percentage of children who tolerated a specified peanut dose at the endpoint double-blind, placebo-controlled oral food challenge at 72 weeks. Success was defined as tolerating a dose at least 2 steps higher than the participant’s baseline challenge dose or reaching 9043 mg of peanut protein. The outcome is a threshold-based challenge measure, rather than a measure of everyday symptoms or peanut intake outside the challenge.

A challenge endpoint gives a structured way to compare participants’ responses under a defined assessment. In general, a threshold can summarize the amount tolerated during that assessment, while the endpoint’s success rule reduces that information to whether a participant crossed a prespecified criterion. This makes the outcome interpretable as a proportion of participants reaching the criterion, but it does not convey every individual’s threshold or response pattern.

The endpoint is used to address the study’s central comparison: whether the proportion reaching the challenge criterion differed between peanut ingestion and continued avoidance. The registry describes a protocol and statistical analysis plan that specified multiple imputation for missing primary endpoint data. Consequently, the posted comparison incorporates imputed values as well as observed endpoint information. The endpoint therefore needs to be read alongside the registry’s account of missing-data handling, not as a simple count of participants with observed challenge results alone.

7. Results Posted in the Registry

• Avoidance: 20.0% (7/35) versus 100.0% (38/38) on control; risk difference -80.0 percentage points (95% CI -93.3 to -66.7), risk ratio 0.20 (0.10–0.39); unadjusted two-proportion z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

The Percentage of Children That Achieve Sustained Unresponsiveness

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Avoidance: 8.6% (3/35) versus 68.4% (26/38) on control; risk difference -59.8 percentage points (95% CI -77.3 to -42.4), risk ratio 0.13 (0.04–0.38); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Avoidance than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Number of Children With Acute Allergic Reactions

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Avoidance: 31.4% (11/35) versus 65.8% (25/38) on control; risk difference -34.4 percentage points (95% CI -55.9 to -12.8), risk ratio 0.48 (0.28–0.82); unadjusted two-proportion z-test p = 0.003.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Avoidance than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcome pattern provides related but distinct context for the primary challenge result. The registry lists sustained unresponsiveness, acute allergic reactions, quality of life, skin-prick-test size, peanut-specific IgE, and peanut-specific IgG4 among the secondary outcomes. Results are posted for sustained unresponsiveness and acute allergic reactions in the material; the remaining listed outcomes do not have estimates. Their inclusion in the outcome list should not be mistaken for evidence of a particular result.

Sustained unresponsiveness is conceptually related to challenge tolerance but is a separate outcome, with a 96-week timeframe. The registry describes it as a surrogate term for tolerance without daily ingestion. It therefore addresses a different aspect of response than the primary endpoint’s challenge threshold at 72 weeks. Acute allergic reactions are also distinct: the posted measure counts participants with an acute reaction, rather than measuring challenge threshold or sustained unresponsiveness. The secondary findings should be read according to these differing definitions and follow-up times.

In general, secondary outcomes can help describe dimensions of a study that the primary endpoint does not capture. They do not automatically confirm or explain the primary result. Here, the record lists no reported analyses for the posted secondary comparisons, and the posted p-values are unadjusted. When multiple outcomes are examined, unadjusted comparisons carry less confirmatory weight because the chance of finding at least one apparently notable result can rise across tests. The registry record does not report a multiplicity strategy or a formal hierarchy that would establish how the secondary outcomes were intended to support the primary analysis.

The secondary outcomes also differ in measurement scale and interpretability. A participant count, a percentage, a quality-of-life score, and laboratory measures cannot be combined into one overall conclusion without a specified method. Since not all listed outcomes have posted results here, the overall secondary pattern is necessarily partial. The appropriate reading is that the reported secondary results offer additional descriptive evidence, but they neither replace the primary endpoint nor establish unreported outcomes.

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

• The Percentage of Children That Achieve Sustained Unresponsiveness
• Number of Children With Acute Allergic Reactions
• Mean Change in Food Allergy Quality of Life Parental Burden Instrument
• Mean Change in SPT Wheal Size
• Mean Change in Peanut-specific IgE
• Mean Change in Peanut-specific IgG4

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary: each participant either met the challenge-based success criterion or did not. The group summaries are percentages, and the comparison is reported as a risk difference. In general, a risk difference subtracts one group’s event proportion from the other’s. Here, the comparison is labeled Avoidance against Peanut Protein, so the negative direction corresponds to a lower success proportion in Avoidance. The registry also reports p < 0.001 for this comparison.

What does the confidence interval describe?

In general, a confidence interval gives a range of values compatible with the data and the statistical method used, while reflecting sampling uncertainty. For a risk difference, an interval entirely on one side of zero indicates that equal group proportions are not among the values covered by that interval under the method used. It does not mean that every value in the interval is equally likely, and it does not account automatically for uncertainties such as missing-data assumptions or generalizability. The interval should be read alongside the direction of the contrast and the outcome definition.

How were missing primary outcomes handled?

The registry record states that the protocol and statistical analysis plan specified multiple imputation because primary endpoint data were missing. It says results from 30 completed-imputed data sets were combined using Rubin’s rule. In general, multiple imputation fills in missing values repeatedly to represent uncertainty about those values; pooling combines estimates across the imputed data sets rather than treating one filled-in data set as if it were complete information.

The registry’s account adds an important group-specific detail. In the Peanut Protein group, participants with missing data were imputed as successes across all imputations, resulting in a 100% success rate. In the Peanut Avoidance group, the proportion of successes varied across imputations, and those varying proportions were averaged. These assumptions and procedures are relevant to the reported comparison. The record does not provide further information about the reasons for missingness or sensitivity analyses using alternative assumptions.

How should the secondary comparisons be read?

The posted secondary comparisons include unadjusted p-values. In general, testing several outcomes creates multiple opportunities to see a small p-value by chance, so an unadjusted result is not automatically strong confirmatory evidence. The registry record does not report a multiplicity adjustment or a hierarchy for interpreting secondary endpoints. It also lists no reported analyses for the posted secondary outcomes. The absence of a reported adjustment should not be replaced with an assumption about what the study team planned.

Confidence intervals are available for the sustained-unresponsiveness percentages and for some posted comparisons. They show uncertainty for those particular summaries. The acute allergic reaction outcome is a count of participants with an event, not a count of all events. The other listed outcomes include quality-of-life and laboratory measures, but their results are not all displayed in the posted summary. Do not infer results for outcomes without posted estimates.

What are the sample-size illustrations?

The sample-size section is a planning illustration based on observed effects, not the trial’s own design calculation and not observed power. It uses a two-sided significance level of 0.05, equal allocation, and a normal approximation. Its rows show participants per arm for 80% and 90% power when planning for different fractions of the observed effect. In general, planning for a smaller effect requires a larger study to reach the same target power, under the assumptions used. These illustrations do not show that the completed study had a particular probability of detecting an effect, and they do not establish the original sample-size rationale, which is not reported in the registry record.

Over-dispersion is a general issue for count outcomes when variability exceeds what a simple count model assumes. It is not a reported feature of the binary primary endpoint here, and the registry record does not describe a count-model analysis for the posted acute-reaction outcome. No such modeling approach should be inferred from the outcome label alone.

11. Confidence Intervals and Precision

The primary result has a large separation between the posted group summaries, and its reported comparison has a confidence interval that remains below zero. The interval indicates that the direction of the group contrast is not especially ambiguous under the reported analysis. It does not eliminate uncertainty about the exact size of the difference or the influence of the imputation assumptions.

The primary outcome display lists 100.0% in Peanut Protein and 21.0% in Avoidance. The accompanying imputation description reports group proportions of 100% and 20% for the comparison. These summaries are not identical as displayed. The record explains that missing values were imputed and pooled, but it does not provide additional detail that resolves the difference in the displayed figures. Both should therefore be described in their respective contexts rather than silently substituted for one another.

In general, confidence intervals tend to be more informative than a p-value alone because they convey a range of estimates compatible with the data and method. A small p-value does not indicate precision by itself, and a confidence interval does not capture every source of uncertainty. In particular, the primary interval does not resolve uncertainty about missing-data assumptions, whether results would generalize beyond this study, or how the result relates to outcomes not measured by the primary challenge criterion.

The secondary sustained-unresponsiveness percentages are accompanied by confidence intervals, and the posted tables show their extent. Those intervals convey uncertainty around each group’s estimated percentage; their width should be considered when reading the estimates. The acute allergic reaction comparison also has an interval, while the registry does not provide intervals for every arm-level outcome display. The posted material therefore offers different levels of precision across outcomes.

Sample size affects precision in general, but the planning illustrations elsewhere on the page are not a substitute for observed confidence intervals. They use the observed effect as an input and describe hypothetical planning under stated assumptions. They are not evidence of observed power or a measure of certainty in this completed trial.

12. Multiplicity, Missing Data and Analysis Populations

The registry record explicitly describes missing-data handling for the primary endpoint. It says that the protocol and statistical analysis plan specified multiple imputation, that 30 completed-imputed data sets were combined using Rubin’s rule, and that missing outcomes in Peanut Protein were imputed as successes across all imputations. For Avoidance, the proportions of successes varied across imputations and were averaged. This approach is central to understanding the posted primary comparison, particularly because the group-specific imputation behavior differed.

The record does not give the reasons for missing primary endpoint data, the amount of missingness by group, or results under alternative assumptions. It also does not report a sensitivity analysis showing how the comparison changes if missing outcomes are treated differently. These details are not reported in the registry record. Without them, readers cannot assess from the posted summary alone how robust the result is to other plausible missing-data assumptions.

Multiplicity is another unresolved aspect. The record lists several secondary outcomes and shows unadjusted comparisons for posted results. It does not report an adjustment for multiple testing, a prespecified hierarchy for secondary outcomes, or details of any interim analyses. These matters are not reported in the registry record. An unadjusted p-value for a secondary comparison should consequently be interpreted as an individual, exploratory-looking result rather than assumed to carry a family-wide error rate guarantee.

The registry record does not report the detailed analysis populations, such as whether every randomized participant was included in each analysis, or how participants with incomplete secondary data were handled. The enrollment and arm counts are posted, but those counts alone do not establish the analysis population for every endpoint. The record also does not report a detailed statistical analysis plan beyond the primary missing-data description. These gaps limit assessment of whether all analyses followed prespecified choices and how data availability affected each outcome.

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 0% reporting threshold.

Most frequent non-serious adverse event terms

The serious adverse event summary lists 0 serious adverse events in each group, among 38 participants at risk in Peanut Protein and 35 in Avoidance. It also lists 0 deaths in both groups. These figures describe the registry’s posted summary for the study; they do not establish that no safety events of any kind occurred.

The record separately lists other adverse events affecting 25 participants in Peanut Protein and 11 in Avoidance, among the same group sizes at risk. These are participant counts, not a measure of event severity, duration, or number of episodes. The registry summary does not provide serious-event terms because the serious-event count is zero in both groups, and it does not give further detail about the other events. The figures should not be treated as a complete characterization of the type or clinical significance of each event.

In general, a zero count in a study does not prove that an event cannot occur; it means that none was recorded in this study’s posted summary. The group sizes and study follow-up also constrain what can be concluded from these counts. The safety figures are descriptive and should be read alongside the study’s design and reporting limits, not as a recommendation about treatment or avoidance.

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

The primary result means that, under the registry’s reported analysis, the Peanut Protein group had a higher posted proportion meeting the study’s challenge-based success criterion than the Avoidance group. The endpoint concerns tolerating a defined dose at an endpoint oral food challenge at 72 weeks. The comparison is based on imputation as well as observed information, according to the registry’s account of missing-data handling.

It does not mean that every participant in the Peanut Protein group had the same threshold, or that the posted percentage describes day-to-day experience outside the challenge. A binary success criterion compresses a range of individual responses into whether each person crossed a threshold. It also does not establish the result for people who differ from the study participants or for settings and procedures not described in the registry record.

The primary result does not, by itself, establish why group outcomes differed. The study description includes immune and molecular objectives, but the posted results do not provide mechanistic findings. Nor does the primary endpoint answer every question about sustained unresponsiveness, quality of life, laboratory measures, or safety. These are separate outcomes with their own definitions and evidence.

The comparison should also not be read without its missing-data context. Missing primary outcomes were handled using multiple imputation, with all missing Peanut Protein outcomes imputed as successes across the imputations and varying averaged proportions in Avoidance. That is part of the reported analysis. The registry record does not provide enough information to judge the effect of alternative assumptions.

Finally, a statistically small p-value is not a measure of clinical importance, and statistical significance does not remove uncertainty. The posted result is evidence about this study’s defined comparison and endpoint. It is not, on its own, a universal conclusion about peanut allergy or an individual-level treatment decision.

15. Why This Trial Matters Statistically

This trial illustrates how an endpoint can shape the statistical interpretation of a clinical study. Rather than treating response as a continuous threshold value, the primary outcome classifies participants according to whether they reach a defined challenge criterion. That makes the comparison straightforward to express as percentages and a risk difference, while losing detail about variation among participants who meet the criterion.

It also shows why missing-data methods need to be read as part of the result rather than as a technical footnote. The registry reports a planned multiple-imputation approach and describes different imputation behavior by group. Because missing Peanut Protein outcomes were imputed as successes across all imputations, while Avoidance proportions varied, a reader needs that context to understand what the posted comparison represents. The record does not report sensitivity analyses that would show how the result behaves under other assumptions.

The design also highlights the distinction between a randomized group comparison and the limits of a registry summary. Randomization is listed, but masking is listed as none, and the posted result cannot convey all implementation details, individual trajectories, or unreported analyses. Confidence intervals and p-values summarize statistical uncertainty under the reported analysis; they do not settle those broader issues.

Finally, secondary endpoints demonstrate the importance of outcome-specific interpretation and multiplicity. Sustained unresponsiveness and acute reactions address different questions from challenge threshold, while other listed outcomes are not accompanied by posted estimates. Unadjusted secondary comparisons should not be treated as independent confirmation of the primary result. Taken together, these features make the trial a useful example of interpreting a registry result through its estimand, missing-data procedure, uncertainty, and reporting scope.

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.

The posted record leaves several trial-specific questions unresolved. It lists randomization, but does not describe the allocation procedure or report the planned sample-size justification. It reports no masking overall, while the primary endpoint uses a double-blind, placebo-controlled challenge; further details about masking for assessments are not reported in the registry record. The study description also does not provide a complete account of ingestion schedules or adherence in the posted summary.

The primary result depends on the stated imputation procedure. Missing outcomes in Peanut Protein were imputed as successes in every imputation, while the Avoidance proportions varied. The registry record does not report the reasons for missing data, the amount missing by group, or sensitivity analyses under alternative assumptions. The outcome display lists 21.0% for Avoidance, while the accompanying comparison reports an imputation-derived proportion of 20%; the posted material does not further explain this difference.

There are also reporting limits for outcomes and analyses. The record lists several secondary measures, but the posted results cover only some of them. It reports no analyses for the posted secondary comparisons and does not describe a multiplicity adjustment or hierarchy. The detailed analysis populations and handling of missing secondary outcomes are not reported in the registry record.

Finally, the primary endpoint is a binary challenge-based measure at 72 weeks. It does not represent every aspect of response, and the secondary sustained-unresponsiveness measure has a different definition and timeframe. These limitations do not negate the posted comparison, but they define what it can and cannot establish from the available registry summary.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Immune Basis and Clinical Implications of Threshold-based Phenotypes of Peanut Allergy — NCT03907397.
• Registry record last updated 2025-02-07; status last verified 2025-02.
• Results first posted 2025-02-07.
• 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

The registry record describes NCT03907397, a completed phase 2 randomized, parallel-group study with no masking. It lists 73 participants: 38 assigned to Peanut Protein and 35 to Avoidance. The primary endpoint was the percentage meeting a challenge-based peanut tolerance criterion at 72 weeks.

The outcome display lists 100.0% for Peanut Protein and 21.0% for Avoidance. The accompanying comparison reports imputation-derived proportions of 100% and 20%, with p < 0.001; the registry describes multiple imputation and pooling for missing primary endpoint data. Secondary results include sustained unresponsiveness and acute allergic reactions. The record also lists quality-of-life and laboratory outcomes, but their estimates are not presented in the summary.

The record reports 0 serious adverse events and 0 deaths in both groups. It lists other adverse events affecting 25 participants in Peanut Protein and 11 in Avoidance. The posted findings should be interpreted in light of the challenge endpoint, missing-data assumptions, unadjusted secondary comparisons, and details not reported in the registry record.