Oral Desensitization to Peanut in Peanut-Allergic Children and Adults Using Characterized Peanut Allergen OIT: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Oral Desensitization to Peanut in Peanut-Allergic Children and Adults Using Characterized Peanut Allergen OIT (NCT01987817), a randomized phase 2 study. The primary outcome is The Percentage of Subjects Who Tolerate at Least 300 mg (443 mg Cumulative) of Peanut Protein With no More Than Mild Symptoms at the Exit DBPCFC.

2. Clinical Question

The registry describes a comparison of the proportion meeting this challenge-based threshold in the AR101 and placebo groups. The endpoint is a responder outcome: each participant either met the stated tolerance criterion or did not. The main contrast is therefore between the groups’ responder proportions, rather than between average doses or symptom scores.

The primary outcome timeframe is listed as 6-9 months. The question is about tolerance at the exit challenge, not about outcomes beyond that assessment. The registry also lists secondary outcomes involving maximum tolerated dose, symptom severity at achieved doses, peanut-specific IgE and IgG4, and skin-prick-test wheal diameter. Those measures address additional aspects of challenge response and immune markers, but they are distinct from the primary responder definition.

3. Trial Design

The registry identifies this as a Phase 2, randomized, parallel-group trial with 56 participants enrolled. It describes a multicenter, randomized, double-blind, placebo-controlled study of the efficacy and safety of characterized peanut oral immunotherapy in people with peanut allergy. The masking field is recorded as triple masking, while the brief description uses the term double-blind. The record does not specify which roles were included in the triple masking designation.

Participants were assigned to AR101 powder provided in capsules or placebo powder provided in capsules. The AR101 capsules contained peanut protein; the placebo capsules contained inactive ingredients. The parallel design means participants were assigned to one of these groups rather than crossing between both interventions.

The registry gives a study start date of 2014-02-06, a primary completion date of 2015-01-07, and a completed status. The primary assessment was an exit double-blind, placebo-controlled food challenge, abbreviated DBPCFC in the record. The outcome timeframe is listed as 6-9 months.

The primary outcome analysis is described as intent-to-treat, defined in the record as participants who received at least one dose of randomized study treatment. The displayed group counts for this outcome are 29 for AR101 and 26 for placebo, while the enrollment field records 56. The registry does not explain the difference between the enrollment total and the number represented in the primary outcome groups. It also does not report the original sample-size justification or the detailed allocation procedure.

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 67 subjects signed the informed consent and went through the screening process. 11 subjects failed screening, resulting in 56 subjects being enrolled and initially randomized. The randomized population comprised 29 subjects in the AR101 group and 27 subjects in the placebo group. The final intent-to-treat (ITT) population (subjects who received at least 1 dose of randomized study treatment) had one fewer subject in the placebo group (n=26).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The number of participants analyzed per outcome measure reflects the intent-to-treat (ITT) population (subjects who received at least 1 dose of randomized study treatment)

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: These values reflect the intent-to-treat population (subjects who received at least 1 dose of randomized study treatment)

6. Primary Endpoint

The registry lists the primary outcome as The Percentage of Subjects Who Tolerate at Least 300 mg (443 mg Cumulative) of Peanut Protein With no More Than Mild Symptoms at the Exit DBPCFC, measured over 6-9 Months.

The primary endpoint counts participants who tolerated at least 300 mg of peanut protein, with 443 mg cumulative, at the exit DBPCFC and had no more than mild symptoms. The record calls participants who met this definition desensitization responders. The endpoint timeframe is 6-9 months.

This is a binary endpoint: a participant is counted as a responder if the full dose-and-symptom criterion is met, and otherwise is not counted as a responder. Presenting the outcome as a proportion gives a direct account of how many participants in each randomized group reached the specified challenge threshold. Comparing these proportions describes the difference between the groups for the prespecified responder definition.

A threshold endpoint is useful for making the outcome interpretable in terms of a defined challenge result. It also simplifies a range of individual challenge outcomes into a responder classification. That simplification has a trade-off: it does not show how far participants who did not meet the threshold progressed, or distinguish among different outcomes below the threshold. The registry lists change in maximum tolerated dose and symptom severity at achieved doses among secondary outcomes, which address different features of the challenge response.

The endpoint concerns tolerance under the conditions of the exit challenge. It is not a measure of every possible response to peanut, and the registry record does not report an assessment beyond the stated challenge or timeframe.

7. Results Posted in the Registry

• AR101 Powder Provided in Capsules: 79.3% (23/29) versus 19.2% (5/26) on control; risk difference 60.1 percentage points (95% CI 38.9 to 81.2), risk ratio 4.12 (1.84–9.26); unadjusted two-proportion z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary because each participant either met the responder definition or did not. The registry gives the number of responders and the number analyzed in each group. For AR101, the recorded count is 23 of 29; for placebo, it is 5 of 26. Comparing the corresponding proportions is an appropriate way to describe a binary group outcome.

What does a risk difference describe?

A risk difference is the first group’s event proportion minus the comparator group’s event proportion. For this endpoint, the event is being classified as a responder. The measure is expressed in absolute terms, so it describes the gap between the two observed responder proportions rather than how many times one proportion is the other. The sponsor reports a risk difference of 60.0, with a 95% confidence interval from 35 to 79. The reported analysis identifies these as percentages. The posted recalculation is an unadjusted comparison based on the group counts and sizes; it is not an adjusted model that accounts for other participant characteristics.

Why report a confidence interval?

A confidence interval gives a range of estimates compatible with the observed data and the method used to calculate the interval. It communicates precision as well as direction: a narrower interval indicates less uncertainty in the estimate, while a wider interval indicates more. The sponsor’s interval is reported with a 95% confidence level. The registry also includes a recalculated interval for the unadjusted comparison. The two should be read in light of their respective methods, rather than treated as though every analysis necessarily uses an identical calculation.

What does the p-value address?

The sponsor reports a Fisher exact test and p-value of < 0.001. A Fisher exact test evaluates how compatible the observed group counts are with a null model of no association between randomized group and responder classification. The posted unadjusted recalculation reports p <0.001. A p-value is not the probability that a treatment effect is true, and it does not measure the size or practical importance of a difference. It is also not a substitute for the estimate and confidence interval.

Fisher’s exact test is a method for categorical count data. It calculates significance using the possible arrangements of counts under the null, conditional on the margins of the count table. This differs in construction from an approximation-based test. The registry’s reported result names Fisher exact; the recalculated comparison is identified as unadjusted. No additional adjusted analysis is reported.

Why can methods yield different results?

Different reasonable procedures can produce slightly different estimates, intervals, or p-values, particularly when group counts are not large. The sponsor’s reported analysis uses its stated method, while a simple recalculation from summary counts is descriptive and unadjusted. The registry does not report a discrepancy between them. The sponsor’s result remains the trial’s own reported analysis; the recalculation offers a way to inspect the basic comparison from the posted counts, not a replacement for the sponsor’s analysis.

Other effect measures can express a binary comparison in relative terms, such as a risk ratio or odds ratio. Relative measures answer a different question from the absolute risk difference and may appear numerically large or small depending on the baseline proportion. The registry record includes recalculated relative measures, but the primary reported analysis is framed as a risk difference. Interpretation should remain anchored to the responder definition and the observed counts.

For an outcome with variation beyond what a simple model expects, an over-dispersed model may be needed to avoid understating uncertainty. That is a general statistical consideration, not a feature reported for this analysis. The primary outcome is a participant-level binary classification, and the record names Fisher exact for the sponsor’s comparison. No over-dispersion model or alternative model is reported.

10. Confidence Intervals and Precision

The sponsor reports a risk difference of 60.0 percentage points with a 95% confidence interval from 35 to 79. This interval communicates uncertainty about the size of the group difference, not just whether the estimate is positive. The interval spans a range of values that are compatible with the data and the reported method. It should not be read as a range containing 95% of individual participant outcomes, nor as a guarantee about the result of a future study.

The interval’s endpoints are separated by a substantial range, so the estimate is not known with pinpoint precision. At the same time, both reported limits are above no difference. The interval therefore indicates that the direction of the observed comparison is more clearly determined than its exact magnitude. The sponsor’s Fisher exact p-value is < 0.001; the p-value addresses compatibility with a null comparison and is not a measure of precision or effect size.

The posted recalculation derives an unadjusted risk difference and interval from the reported summary counts. Recalculated intervals and sponsor-reported intervals can reflect different statistical procedures. The registry reports no discrepancy between its analyses, and the sponsor’s result is the trial’s own stated analysis. Neither interval removes uncertainty arising from the number of participants, the outcome definition, or limitations in the record’s reporting of missing outcomes and analysis populations.

Precision concerns the estimate for this particular responder definition and comparison. It does not quantify uncertainty about unreported outcomes or assessments outside the exit challenge. The interval should therefore be interpreted alongside the endpoint, group counts, and analysis population described in the registry.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total events. They are illustrations based on the observed effect in this trial, not the trial’s original design calculation. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The assumed effect to detect is a stated fraction of the observed effect. The trial’s own original sample-size assumptions and justification are not reported in the registry record.

For planning for 100% of the observed effect, the illustration gives 10 participants per arm for 80% power and 13 participants per arm for 90% power. For planning for 75% of the observed effect, it gives 18 participants per arm for 80% power and 24 participants per arm for 90% power. For planning for 50% of the observed effect, it gives 38 participants per arm for 80% power and 51 participants per arm for 90% power.

In plain language, the 80% power column describes the illustrated number of participants per arm under which the stated effect would be detected with 80% probability under the assumptions. The 90% column uses a higher target probability of detection, so it generally requires more participants for the same assumed effect. Smaller effects are harder to distinguish from random variation and therefore require more participants in these illustrations. These planning figures are not observed power.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.192; treatment proportion p₁ = 0.793; average p̄ = 0.493.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7070 = 1.3858.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5652 = 0.4757.
4. n = (1.3858 + 0.4757)² / (0.6008)² = 9.6, rounded up to 10 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.192307692307692                          # control proportion
p1 <- 0.793103448275862                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(1987817)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 10 (80% power), 13 (90% power)
 75% of observed effect: n per arm = 18 (80% power), 24 (90% power)
 50% of observed effect: n per arm = 38 (80% power), 51 (90% power)
power.prop.test: 9.6 
Simulated power at n = 10 per arm: 0.811 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary comparison as a Fisher exact test for risk difference. The reported estimate is 60.0, with a 95% confidence interval from 35 to 79, and p < 0.001. The analysis is labeled as a superiority analysis. The estimate and interval are presented by the registry as the trial’s own analysis.

The unadjusted recalculation uses the posted responder counts and group sizes to examine the same basic contrast. Its p-value is <0.001. Such a recalculation is not an adjusted analysis and does not replace the sponsor’s reported method. It is useful as a check on the comparison derived from the summary counts, while the sponsor’s reported test and estimate remain the primary trial results as recorded.

The registry does not flag a discrepancy between the sponsor analysis and the recalculation. Differences in methods can nonetheless lead to differences in the way uncertainty or significance is calculated. Interpretation should prioritize the sponsor’s own reported analysis and retain the endpoint definition, study population, and confidence interval as context.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists five secondary outcomes alongside the primary endpoint. These include change in maximum tolerated dose at the exit DBPCFC, the number of participants by maximum dose achieved with no or mild symptoms, changes in peanut-specific IgE and IgG4, and change in skin-prick-test mean peanut wheal diameter. The record also lists six outcome measures overall and two analyses.

Testing multiple outcomes or making multiple comparisons can increase the chance of at least one result appearing statistically notable by chance. Statistical plans sometimes address this through prespecified testing sequences or adjustment procedures. The registry record does not report whether any multiplicity adjustment, hierarchy, or other control strategy was planned. The existence of listed secondary outcomes alone does not establish how their analyses were handled.

The primary outcome population is defined as the intent-to-treat population, consisting of participants who received at least one dose of randomized treatment. The registry does not provide a detailed account of outcome availability for each participant, reasons for missing assessments, or the method used to handle missing primary endpoint data. It also does not report whether any imputation or sensitivity analyses were undertaken.

The enrollment field records 56 participants, while the primary outcome groups contain 29 and 26. The registry does not explain this difference. The group counts should therefore be read as the numbers represented in the recorded primary outcome comparison, not as a complete accounting of every enrolled participant. Without further information on missing data and analysis populations, it is not possible to assess from the record alone how robust the result is to alternative assumptions about participants without an observed endpoint.

15. 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: hypersensitivity (3.8%).

The serious adverse event counts are 1 of 29 participants in the AR101 group and 1 of 26 in the placebo group. The registry records zero deaths in each group. It also lists hypersensitivity among the serious adverse event terms, with 3.8% shown for that term. The registry does not provide further detail in these summary fields about the timing, course, or relationship of the listed events to study treatment.

Other adverse event counts are recorded for 28 of 29 participants in the AR101 group and 22 of 26 in the placebo group. These counts are separate from the serious adverse event counts. They should not be interpreted as a measure of severity or as evidence about causation without additional event-level information.

These figures describe events recorded in the trial groups, not the full range of potential risks or outcomes beyond the study. They do not establish whether an event was caused by an intervention. The registry record does not report detailed event descriptions or exposure-adjusted rates in the information. Safety interpretation is therefore limited to the counts and term listing recorded in the registry.

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

The primary result describes a difference between randomized groups in the proportion of participants meeting a challenge-based responder definition at the exit DBPCFC. The recorded counts are 23 of 29 for AR101 and 5 of 26 for placebo. The sponsor’s reported risk difference is 60.0, with a 95% confidence interval from 35 to 79 and p < 0.001. These statistics characterize the observed comparison for this outcome and analysis.

The outcome is specific: participants had to tolerate at least 300 mg of peanut protein, 443 mg cumulative, with no more than mild symptoms. The result does not describe every possible dose response, the full distribution of symptoms, or outcomes outside the exit challenge. A responder proportion also does not show how participants who did not meet the threshold differed from one another.

The comparison does not, on its own, establish the result for people outside the trial population or at other follow-up times. The registry reports a Phase 2 study with 56 enrolled participants and an outcome timeframe of 6-9 months. The primary outcome groups account for 29 and 26 participants, and the reason for the difference from enrollment is not reported.

The result should not be read as a complete account of benefits and harms. The registry lists selected safety counts and secondary outcomes, but the available record does not provide the full detail needed to interpret every outcome, missing assessment, or event. Statistical evidence for a group difference is not the same as a complete judgment about clinical importance. The posted data support interpretation of the specified endpoint and comparison, while leaving broader questions dependent on information not reported in the registry record.

17. Why This Trial Matters Statistically

This trial illustrates how a clinically defined challenge outcome can be translated into a binary endpoint. The responder definition combines a dose threshold with a symptom-severity condition, making the primary outcome a count of participants who satisfy both criteria. The resulting comparison is direct, but it necessarily compresses a more detailed range of challenge experiences into two categories.

It also demonstrates why an absolute effect measure and its confidence interval matter alongside a p-value. The sponsor reports a risk difference with a confidence interval, while the Fisher exact p-value addresses compatibility with a null comparison. These statistics answer related but different questions: one describes the estimated separation between groups and its uncertainty, and the other summarizes evidence against the null under the test method.

The record further shows the value of keeping the analysis population and denominator visible. Enrollment is recorded as 56, while the primary outcome groups contain 29 and 26 participants. Because the record does not explain the difference or describe missing outcome handling, interpretation requires attention to the population represented in the outcome analysis.

Finally, the planning illustrations make clear that sample-size needs depend on the effect selected for planning and the target power. They are not evidence about the trial’s original design or its observed power. Separating planning calculations from the trial’s reported analysis helps prevent a retrospective illustration from being mistaken for a prospective design decision.

18. 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 does not report the original sample-size justification, detailed allocation procedure, or full statistical analysis plan. Although the primary outcome analysis is described as intent-to-treat, defined as receiving at least one dose, the record does not explain why the primary outcome groups contain 55 participants when enrollment is listed as 56.

Detailed handling of missing primary endpoint data is not reported. The record does not state whether all participants completed the exit DBPCFC, how unavailable assessments were treated, or whether sensitivity analyses evaluated alternative assumptions. These omissions make it difficult to assess how missing outcomes might affect the group comparison.

The registry lists secondary outcomes but does not report a multiplicity strategy. It is therefore not possible to determine from the record whether secondary analyses were adjusted for multiple testing or interpreted under a prespecified hierarchy. The listed outcomes also cover several different kinds of measures, which should not be treated as interchangeable with the primary responder endpoint.

The trial is identified as Phase 2 and has 56 participants enrolled. The result is tied to its defined population, intervention, comparator, and exit challenge timeframe. The registry record does not report outcomes beyond the stated assessment or enough event-level safety detail to characterize the timing, severity, and relationship of each event. These limitations constrain what can be concluded from the posted summary statistics, without changing what the reported primary comparison shows.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Oral Desensitization to Peanut in Peanut-Allergic Children and Adults Using Characterized Peanut Allergen OIT — NCT01987817.
• Registry record last updated 2021-11-30; status last verified 2021-11.
• Results first posted 2021-11-02.
• 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.

22. Record Summary

NCT01987817 is a completed Phase 2 randomized, parallel-group study of AR101 powder in capsules versus placebo capsules in children and adults with peanut allergy. The record describes the study as multicenter and placebo-controlled, lists triple masking, and records 56 participants enrolled.

The primary endpoint was the proportion tolerating at least 300 mg of peanut protein, 443 mg cumulative, with no more than mild symptoms at the exit DBPCFC. The primary outcome groups contain 29 participants in the AR101 arm and 26 in the placebo arm. The sponsor reports a risk difference of 60.0, a 95% confidence interval from 35 to 79, and p < 0.001.

The record lists five secondary outcomes and selected adverse event counts. It does not report the trial’s original sample-size justification or detailed handling of missing outcomes.