Treatment of Multifood Allergy With Omalizumab or Multiallergen Oral Immunotherapy

This randomized clinical trial investigates omalizumab and oral immunotherapy in the treatment of patients with multifood allergy.

Key Points

Question

How do omalizumab and oral immunotherapy (OIT) compare in the treatment of patients with multifood allergy?

Findings

In this randomized clinical trial comparing omalizumab with multiallergen OIT (MOIT) in 117 participants with multifood allergy, omalizumab was superior in terms of both efficacy and safety.

Meaning

Although the intention-to-treat analysis found a higher rate of treatment success in those receiving omalizumab, there was no difference in efficacy after accounting for the high rate of study withdrawals in the group of participants treated with omalizumab-facilitated MOIT.

Abstract

Importance

Food allergy is common, affecting up to 8% to 10% of children and adults. Treatment options include oral immunotherapy (OIT) and omalizumab, an anti–immunoglobulin E (IgE) monoclonal antibody.

Objective

To compare omalizumab with OIT for the treatment of patients with multifood allergy.

Design, Setting, and Participants

This was a double-blind, placebo-controlled, randomized clinical trial comparing omalizumab with omalizumab-facilitated multiallergen OIT (MOIT) in participants who completed stage 1 of the Omalizumab as Monotherapy and as Adjunct Therapy to Multiallergen OIT in Children and Adults With Food Allergy (OUTMATCH) trial, which led to the approval of omalizumab. The setting comprised 10 academic centers across the US. Included in this analysis were individuals aged 1 to 55 years with an allergy to peanuts and at least 2 other foods (milk, eggs, wheat, cashews, hazelnuts, walnuts). Eligibility was based on oral food challenge thresholds, requiring dose-limiting symptoms to cumulative doses of 144 mg or less of protein for peanuts and 444 mg or less for nonpeanut allergens. Data were analyzed from October 2024 to February 2026.

Interventions

Participants were randomized to receive MOIT with placebo omalizumab or omalizumab with placebo MOIT. All received 16 weeks of open-label omalizumab; at week 8, active or placebo MOIT was initiated and escalated to goal doses of 1000 mg per food. At week 16, participants transitioned to blinded omalizumab or placebo injections for 44 weeks.

Main Outcomes and Measures

The primary end point was cumulative tolerated dose (CTD) of 4044 mg or greater for all 3 foods. Predefined secondary end points included CTDs of 1044, 2044, 4044, 6044, or 8044 mg for 1, 2, or all 3 foods.

Results

A total of 117 participants (median [IQR] age, 7 [1-29] years; 64 male [55%]) were randomized to receive active MOIT (n = 58) or active omalizumab (n = 59). A total of 30 participants (51%) receiving active MOIT and 51 (88%) receiving active omalizumab completed the study. In the intention-to-treat (ITT) analysis, omalizumab was superior to MOIT (21 of 58 [36%] vs 11 of 59 [19%]; odds ratio, 2.6; 95% CI, 1.1-6.3; P = .03), with no differences in per-protocol analyses. Omalizumab superiority for CTDs of 4044 mg or greater was also demonstrated for 2 or more foods and for several individual foods. More participants taking active MOIT experienced adverse events (serious adverse events in 18 of 59 [31%] vs 0%; events leading to discontinuation in 13 of 59 [22%] vs 0%; events treated with epinephrine (22 of 59 [37%] vs 4 of 58 [7%]).

Conclusions and Relevance

Although the ITT analysis found a higher rate of treatment success in those receiving omalizumab compared with MOIT, results suggest that the difference was largely driven by the high rate of study discontinuation in the participants treated with MOIT, mostly related to adverse events.

Introduction

Food allergy is common and in the US has been estimated to affect up to 8% of children and 10% of adults, with a large proportion (30%-86%) allergic to multiple foods. It causes substantial morbidity and is a common cause of anaphylaxis. Successful avoidance of offending foods is difficult, particularly for patients allergic to multiple foods. Food allergy has negative impacts on quality of life and is a significant burden to patients, families, and health care systems.

Omalizumab was approved in 2024 for the treatment of food allergy, with the indication “…for the reduction of allergic reactions, including anaphylaxis, that may occur with accidental exposure to 1 or more foods.” Only 1 other treatment for food allergy is US Food and Drug Administration (FDA) approved, a peanut oral immunotherapy (OIT) product, although OIT using commercial, non–FDA-approved products is commonly used in clinical practice, making it essential that controlled trials be performed to compare therapies. Studies have been conducted using OIT in combination with omalizumab, but there have been no trials directly comparing omalizumab and OIT. The Omalizumab as Monotherapy and as Adjunct Therapy to Multiallergen OIT in Children and Adults With Food Allergy (OUTMATCH) trial, which demonstrated the efficacy of omalizumab as monotherapy for multifood allergy, also included a second stage, placebo-controlled, randomized clinical trial comparing omalizumab with multifood OIT (MOIT).

Study Design and Methods

This study was approved by a single investigational review board at Johns Hopkins and was designed as stage 2 of the OUTMATCH trial, a multicenter, double-blind, placebo-controlled, randomized clinical trial of individuals with food allergies, aged 1 to 55 years, and allergic to peanuts and at least 2 other study-specified foods (milk, eggs, wheat, cashews, hazelnuts, or walnuts). Although a participant could be allergic to more than 2 other foods, only peanuts and 2 other foods were studied for each participant. In participants with more than 2 other food options, the included foods, other than peanuts, were chosen jointly by the participant and the study team. Written informed consent was obtained from all participants. Participants self-reported (or the parents of minor participants reported) the following races and ethnicities: Asian, Black or African American, Hispanic or Latino, not Hispanic or Latino, multiracial, Native Hawaiian or Other Pacific Islander, White, or unknown. Race and ethnicity were collected as descriptive variables due to potential variation across sites and to allow assessment of generalizability of the study findings. This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guidelines.

Protocol versions 8 to 10 are included with this article in Supplement 1; prior versions were previously published. Participants had to have a weight and total immunoglobulin E (IgE) level suitable for omalizumab dosing. After meeting allergy skin test and laboratory inclusion criteria, eligibility was based on oral food challenge (OFC) thresholds, requiring dose-limiting symptoms to a single dose of 100 mg or less of peanut protein (cumulative 144 mg) and 300 mg or less for each nonpeanut allergen (cumulative 444 mg). Key exclusion criteria included poorly controlled or severe asthma, history of life-threatening anaphylaxis to participant-specific foods, history of eosinophilic gastrointestinal disease, and recent food or other immunomodulatory therapy.

In stage 1 of the OUTMATCH trial, participants were randomized to receive 16 to 20 weeks of omalizumab or placebo injections, after which double-blind, placebo-controlled food challenges were repeated to each participant’s 3 foods and placebo to a cumulative dose of 6044 mg (eTable 1 in Supplement 2). The first 60 participants completing stage 1 were assigned to a 24-week open label extension of omalizumab, whereas the rest proceeded to stage 2. Additional participants entered stage 2 directly after stage 1 enrollment was closed after omalizumab efficacy had been demonstrated.

In stage 2, participants were randomized via Interactive Web Randomization System to receive double-blind MOIT with placebo omalizumab or omalizumab with placebo MOIT using a 1:1 allocation ratio and a permuted block randomization scheme with variable block sizes and stratified by stage 1 treatment arm. All participants began with 16 weeks of open-label omalizumab; at week 8, MOIT or placebo-MOIT was initiated, and at week 16, participants transitioned to blinded omalizumab or placebo injections for 44 weeks. As denoted in the study displays, MOIT refers throughout to omalizumab-facilitated MOIT. Participants were required to tolerate 9 mg or more of protein (3 mg of each food), with a maximum of 375 mg of each food of active or placebo MOIT during an initial dose escalation. They then entered a buildup phase where they had up to 24 weeks to reach a maximum maintenance dose of 1000 mg for each food (3000 mg total protein) (eTable 2 in Supplement 2). A minimum maintenance dose of 250 mg/food was required, after which the maximum dose achieved was continued. At the end of the 52-week treatment, participants completed double-blind, placebo-controlled food challenges to their 3 foods and placebo to cumulative doses of 8044 mg.

Study stopping rules are listed in eTable 3 in Supplement 2. Also, as detailed previously, the study experienced a brief interruption when mold growth was identified in a small number of food products being used for OFCs and MOIT. The duration of interruption for each participant is displayed in eFigure 1 in Supplement 2. After these interruptions, MOIT dose adjustments, when needed, were guided by the study leadership team in communication with the institutional review board, data safety monitoring board, and FDA (eTable 3 in Supplement 2).

Outcomes

The cumulative tolerated dose (CTD) at the end of stage 2 was the primary outcome with the end point designed as a binary response (success or failure) for CTD greater than or equal to 4044 mg for each of the 3 foods. Not reaching a CTD of at least 4044 mg or withdrawal from the study before the final double-blind, placebo-controlled food challenges were considered failures. With regard to the respective food challenge, it was considered a treatment failure when participants either experienced dose-limiting symptoms at any dose of the placebo OFC or prematurely stopped any OFC.

Predefined secondary end points included CTDs of 1044, 2044, 4044, 6044, or 8044 mg for 1, 2, or all 3 foods and the number of foods with CTDs of 1044, 2044, 4044, 6044, or 8044 mg at the end of stage 2. To also examine potential protection from smaller food exposures, a post hoc analysis was added for a CTD of 444 mg for individual and multiple foods.

All reported adverse events (AEs) were assessed for their association to study procedures or treatment. To identify AEs associated with MOIT, all symptoms or events that occurred within 2 hours of administration were deemed related or possibly related. AEs occurring between visits were captured in daily diary logs.

Epinephrine administered for MOIT (or placebo MOIT) dosing reactions or omalizumab (or placebo) injections was considered a serious AE (SAE), as was any epinephrine use for reactions related to unintentional allergen exposures. Use of epinephrine for treatment of reactions during OFCs was not considered an SAE.

Statistical Analysis

Initial sample size calculations assumed that 128 participants (64 per arm) would be randomized into stage 2. This would have provided 86% power, based on a 2-sided Fisher exact test with a type I error rate of 5% and assuming primary end point success rate of 85% (MOIT arm) and 60% (omalizumab arm).

The intention-to-treat (ITT) population was used for the primary analysis of the primary end point. The ITT population included all participants who received at least 1 open-label omalizumab injection. The primary analysis was repeated using a per-protocol (PP) population, which included all participants who completed stage 2. The Cochran-Mantel-Haenszel method, controlling for stage 1 arm, was used to calculate the common odds ratio (OR) and 95% CI comparing the primary end point between the 2 arms. A 2-sided significance level of .05 was used without any further type I error control for supportive analyses for the primary end point.

The Hochberg method was used to control the familywise type I error at 0.05 across the 6 secondary end points defined by a CTD of 4044 mg or greater for (1) peanuts, (2) cashews, (3) eggs, (4) milk, (5) walnuts, and (6) the multifood end point of a CTD of 4044 mg or greater for at least 2 foods. There were no multiplicity adjustments for other secondary or exploratory analyses. Fisher exact test was used for subgroup analyses of the primary end point, with comparisons of the stage 2 arms within subgroups defined by age and by stage 1 arm.

AEs were summarized for the safety population (same as ITT) according to the number of events and the number and percentage of participants with events across the entirety of stage 2 and separately for the 3 segments of stage 2 (open-label omalizumab with no MOIT, open-label omalizumab plus double-blind MOIT, and double-blind injections plus double-blind MOIT). AEs were also grouped into clinically meaningful organ systems and summarized across the stage according to their relationship to MOIT.

Analyses were analyzed from October 2024 to February 2026 using SAS software, version 9.4 (SAS Institute) and figures were generated using R Statistical Software, version 4.4.1 (R Core Team 2024). The statistical analysis plan is available in Supplement 1.

Results

The study included 117 participants (median [IQR] age, 7 [1-29] years; 53 female [45%]; 64 male [55%]) in both the ITT and safety populations (Figure 1). Participants were randomized to active MOIT (n = 58) or active omalizumab (n = 59). A total of 30 participants (51%) receiving active MOIT and 51 (88%) receiving active omalizumab completed the study. Early study discontinuations accounted for the 36 participants (29 in the MOIT arm and 7 in the omalizumab arm) excluded from the PP population.

Figure 1. Consolidated Standards of Reporting Trials (CONSORT) Flow Diagram.

By protocol, the use of epinephrine (Epi) was considered a serious adverse event (SAE) if used for an allergic reaction that occurred during multiallergen oral immunotherapy (MOIT; or placebo MOIT) dosing or omalizumab (or placebo for omalizumab) injections. Recruitment into stage 1 of the study began in July 2019 and extended through October 2022. EOE indicates eosinophilic esophagitis.

Age, race, ethnicity, stage 1 treatment, total IgE level, specific food distribution (Table) and baseline skin test size, food-specific IgE, and food-specific CTD results (eTable 4 in Supplement 2) were similar between treatment arms. Participants self-reported (or the parents of minor participants reported) the following races and ethnicities: 15 Asian (13%), 12 Black or African American (10%), 11 Hispanic or Latino (9%), 106 not Hispanic or Latino (91%), 23 multiracial (20%), 66 White (56%), 1 unknown (0.9%).

Table. Participant Characteristics.

Characteristic	Omalizumab-facilitated MOIT (n = 59)	Omalizumab + placebo MOIT (n = 58)
Sex, No. (%)
Female	26 (44)	27 (47)
Male	33 (56)	31 (53)
Age at stage 2 randomization, median (range), y	6.0 (1-28)	8.5 (1-29)
Age group at stage 2 randomization, No. (%)
≤5 y	23 (39)	19 (33)
6-11 y	23 (39)	25 (43)
≥12 y	13 (22)	14 (24)
Weight, median (IQR), kg	21.9 (16.0-35.5)	27.5 (18.0-39.5)
Body mass index, median (IQR)a	16.7 (15.3-18.3)	16.8 (15.8-19.0)
Ethnicity, No. (%)
Hispanic or Latino	5 (9)	6 (10)
Not Hispanic or Latino	54 (92)	52 (90)
Race, No. (%)
American Indian or Alaska Native	0	0
Asian	4 (7)	11 (19)
Black or African American	7 (12)	5 (9)
Multiracial	14 (24)	9 (16)
Native Hawaiian or Other Pacific Islander	0	0
White	34 (58)	32 (55)
Unknown	0	1 (2)
Total IgE at screening, median (IQR), IU/mL	623 (334–1222)	717 (431–1054)
Participant-specific food, No. (%)
Peanuts	59 (100)	58 (100)
Cashews	41 (69)	32 (55)
Eggs	23 (39)	20 (34)
Milk	17 (29)	18 (31)
Walnuts	27 (46)	29 (50)
Hazelnuts	6 (10)	11 (19)
Wheat	4 (7)	6 (10)
Stage 1 treatment arm, No. (%)b
Omalizumab	38 (64)	38 (66)
Placebo	20 (34)	19 (33)
Analysis populations, No. (%)
Intention to treat	59 (100)	58 (100)
Per protocol	30 (51)	51 (88)

Abbreviations: IgE, immunoglobulin E; MOIT, multiallergen oral immunotherapy.

^a Calculated as weight in kilograms divided by height in meters squared.

^b Two participants were enrolled directly into stage 2 with one assigned to each of the stage 2 arms. In analyses that control for stage 1 arm, these participants are considered to be in the placebo arm so they could be included in the analyses.

For the primary end point (ITT analysis), omalizumab was superior to MOIT with success in 21 of 58 participants (36%) vs 11 of 59 participants (19%; OR, 2.6; 95% CI, 1.1-6.3; P = .03) (Figure 2 and eTable 5 in Supplement 2), with no difference in the PP analysis (OR, 1.2; 95% CI, 0.5-3.3; P = .66). The difference between the ITT and PP results reflects substantial disparity in study discontinuations between the arms (MOIT, 29 of 59 [49%]; omalizumab, 7 of 58 [12%]) (Figure 1 and eTable 6 in Supplement 2). The majority of study discontinuations in the MOIT arm were associated with AEs, SAEs, and MOIT-associated symptoms (Figure 1 and eTable 6 in Supplement 2).

Figure 2. Bar Graphs Showing Successfully Consumed Cumulative Tolerated Dose (≥4044 mg) for All 3 Foods Without Dose-Limiting Symptoms (Primary End Point) in the Intention-to-Treat and Per-Protocol Analysis Populations.

Common odds ratio (OR) and 95% CI estimated using Cochran-Mantel-Haenszel method controlling for stage 1 treatment arm are for (odds of success for omalizumab + placebo multiallergen oral immunotherapy [MOIT]) / (odds of success for omalizumab-facilitated MOIT). Two participants enrolled directly into stage 2 are considered to have been assigned to placebo for stage 1 for the sake of these analyses.

After adjustments for multiplicity, superiority of omalizumab over MOIT for a CTD of 4044 mg or greater was also demonstrated in the ITT population for 2 or more foods (OR, 3.3; 95% CI, 1.5-7.3; P = .002) and for several individual foods (peanuts, OR, 4.3; 95% CI, 1.9-9.7; P <.001; milk, OR, 6.2; 95% CI, 1.4-27.6; P = .02; eggs, OR, 11.7; 95% CI, 1.9-70.2; P = .004) (eTable 7 in Supplement 2). Additional analyses in the ITT population also indicated higher omalizumab efficacy, including success at a CTD of 4044 mg or greater for 1 or more food items, success for all 3 foods at CTDs of 444 mg or greater (ad hoc) 1044 mg or greater and 8044 mg (Figure 3 and Figure 4), as well as differences in CTDs at different levels for 1 or more foods (Figure 4) and for individual foods (eFigure 2 in Supplement 2). For example, 42 of 58 participants (72%) treated with omalizumab vs 23 of 59 participants (39%) in the MOIT arm successfully tolerated 444 mg or more of all 3 allergens and 14 of 58 participants (24%) vs 6 of 59 participants (10%) tolerated 8044 mg or more for all 3 foods (Figure 4). There were no differences in the same analyses in the PP population.

Figure 3. Line Graphs Showing Success for 1, 2, and 3 Foods at Each Cumulative Tolerated Dose Level, Intention-to-Treat and Per-Protocol Analysis Populations.

Shown are the percentages and 95% Clopper-Pearson CIs (vertical axis) of the participants in the 2 groups who, during the end of treatment double-blind, placebo-controlled food challenges, consumed the indicated number of foods (≥1, ≥2, or 3) at the cumulative doses shown in the horizontal axis. MOIT indicates multiallergen oral immunotherapy.

Figure 4. Bar Graphs Showing Comparison of Efficacy for Select Prespecified and Ad Hoc End Points, Intention-to-Treat and Per-Protocol Analysis Populations.

Common odds ratio (OR) and 95% CI estimated using Cochran-Mantel-Haenszel method controlling for stage 1 treatment arm are for (odds of success for omalizumab + placebo multiallergen oral immunotherapy [MOIT]) / (odds of success for omalizumab-facilitated MOIT). Two participants enrolled directly into stage 2 are considered to have been assigned to placebo for stage 1 for the sake of these analyses. The primary end point was cumulative tolerated dose (CTD) ≥4044 mg for all 3 foods. The following was added as a post hoc end point: CTD ≥444 mg for all 3 foods.

Several post hoc analyses addressed areas of special interest. Recognizing prior evidence that OIT may work best in younger children, we explored the association between age and outcomes within each treatment arm. Although not statistically significant, point estimates suggest the possibility of higher OIT success in the 1- to 5-year age group (eFigure 3 in Supplement 2).

Second, recognizing that the protocol allowed for a wide range in MOIT maintenance doses (750-3000 mg), we assessed whether those taking the maximum dose (3000 mg) at the end of stage 2 had different outcomes than those taking lower maintenance doses. Although limited by the small sample size when focusing only on those who could tolerate MOIT, trends suggested potential benefits for those taking the highest maintenance dose (7 of 13 [54%] vs 4 of 17 [24%]; OR, 0.26; 95% CI, 0.04-1.59; P = .13) in the PP population (eTable 8 in Supplement 2).

Next, to examine whether treatment administered in stage 1 influenced the results of stage 2, we compared participants taking active vs placebo omalizumab in stage 1. This analysis suggested a possible advantage of having received omalizumab in stage 1 (eTable 9 in Supplement 2). Significant differences in success were seen for 2 end points (success for all 3 foods at CTDs of ≥1044 mg; OR, 3.8; 95% CI, 1.0-18.2; P = .048 and ≥4044 mg; OR, ∞; 95% CI; 2.2-∞; P = .005) and point estimates for success with other end points suggested similar trends.

Lastly, we assessed the reproducibility of stage 1 results in a new cohort of participants who had not previously received omalizumab. Results for this subgroup were similar to those observed in stage 1 (eTable 10 in Supplement 2). Further, the stage 2 results suggest improved efficacy with longer omalizumab treatment. For example, with peanuts, we found 67% success during stage 1 (16-20 weeks) and 84% in stage 2 among those on omalizumab in both stages (68-72 weeks) (eTable 10 in Supplement 2).

Among those receiving active MOIT, we explored factors that might explain the high rate of study discontinuation (eTable 11 in Supplement 2). Other than higher baseline peanut IgE level in participants who discontinued (median level, 64 vs 30 kilounits of allergen-specific IgE per liter; P = .04), no differences were detected for any other variables of interest, including the foods included in the MOIT, age, stage 1 treatment arm, or baseline peanut CTD.

Three participants withdrew during the study pause associated with the mold problem, but these were not related to AEs, and there was no evidence that this pause affected outcomes (eTable 12 in Supplement 2).

Adverse Events

More participants in the MOIT arm experienced SAEs (18 of 59 [31%] vs 0%; risk difference, −30.5; 95% CI, −47.1 to −12.9), AEs leading to treatment discontinuation (13 of 59 [22%] vs 0%; risk difference, −22.0; 95% CI, −39.2 to −4.2), AEs treated with epinephrine (22 of 59 [37%] vs 4 of 58 [7%] overall; risk difference, −30.4; 95% CI, −47.1 to −12.9; 18 of 59 [31%] vs 0% related or possibly related to MOIT; risk difference, −30.5; 95% CI, −47.1 to −12.9), AEs meeting criteria for anaphylaxis (16 of 59 [27%] vs 1 of 58 [2%] overall; risk difference, −25.4; 95% CI, −42.4 to −7.7; 16 of 59 [27%] vs 0% related or possibly related to MOIT−27.1; 95% CI, −44.0 to −9.4), and AEs related or possibly related to MOIT (50 of 59 [85%] vs 14 of 58 [24%]; risk difference, −60.6; 95% CI, −73.8 to −44.0). Overall data are listed in eTable 13 in Supplement 2. eTable 14 in Supplement 2 summarizes the number of reported events grouped by system overall and separately for those related or possibly related to MOIT. Related events on active MOIT were reported more frequently compared with placebo MOIT (n = 418 vs n = 28), with specific categories including systemic allergic reactions and skin, gastrointestinal, upper respiratory, and lower respiratory symptoms. There were also 3 cases of biopsy-confirmed eosinophilic esophagitis in the active MOIT group.

eFigure 1 in Supplement 2 displays participant-specific timing of AEs related or possibly related to MOIT, as well as all study discontinuations and MOIT treatment interruptions related to the mold issue.

Discussion

Although omalizumab has been approved for the treatment of food allergy, questions remain regarding its optimal use and efficacy in comparison to other treatment options. OUTMATCH stage 2 afforded a direct comparison of omalizumab with MOIT in patients with multiple food allergies. The study was designed to provide the fairest possible comparison, including the use of open-label omalizumab before the initiation of MOIT.

Higher success rates were seen in the omalizumab arm for the primary end point, as well as a variety of other end points, in the ITT population. For example, 72% of participants in the omalizumab arm vs 39% in the MOIT arm successfully tolerated 444 mg or greater of all 3 allergens (relevant to protection from small unintentional exposures) and 24% vs 10% tolerated 8044 mg or greater of all 3 foods (serving size portions, eg, 1 cup of milk or approximately 30 peanuts).

Although the overall results of this trial are clear, it is important to note that the superiority of omalizumab over MOIT was largely driven by the high rate of study discontinuations in the participants receiving MOIT, the majority of which were due to AEs related to treatment. Closer examination revealed that participants receiving MOIT experienced significantly more AEs whether categorized as SAEs, AEs leading to discontinuation, or events treated with epinephrine. There were also 3 cases of eosinophilic esophagitis in the MOIT arm.

Although higher AE rates were expected in the population receiving MOIT, the frequency of study discontinuations was higher than in most other OIT studies, even some using MOIT and most without pretreatment with omalizumab. Comparing those participants taking MOIT who completed stage 2 with those who did not, no differences were found with regard to age, the specific foods included, or the treatment assignment in stage 1 (eTable 11 in Supplement 2). We suggest that possible reasons for the high MOIT discontinuation rate are (1) the nature of the study population, all documented to be highly reactive to multiple foods; (2) the potential to initiate MOIT at higher than typical doses and to escalate it more rapidly, given the pretreatment with omalizumab; (3) the MOIT goal dose, seeking to reach a maintenance dose of 1000 mg for each food; (4) protocol-imposed rigidity in dosing, including the fact that the 3 foods were managed as a composite without any option to discontinue or adjust dosing of individual foods; and (5) aversion to the taste and/or volume of food flours.

We also considered whether the treatment received in stage 1 (16-20 weeks of omalizumab or placebo) may have affected stage 2 results, especially in the arm receiving MOIT. Although no differences were detected regarding study discontinuations, outcomes of those who completed stage 2 appeared to be better for those who received omalizumab in stage 1. On the other hand, although pretreatment with omalizumab was assumed to be advantageous based on numerous prior studies, it is also possible it may have done more harm than good by allowing participants to begin MOIT at higher than usual doses.

Although success rates in participants who were able to continue MOIT were similar to those treated with omalizumab, they still fell below the predictions used in protocol development and sample size calculations, having predicted 85% success for MOIT vs 60% for omalizumab. We cannot offer a clear explanation for this difference, although there is a suggestion that within participants receiving MOIT who completed the study, success was lower in those who did not achieve the intended 3000-mg maintenance dose (24% vs 54%; P = .13).

Limitations

The study has some limitations. The sample size was smaller than originally planned as enrollment in stage 1 was stopped early with the demonstration of omalizumab efficacy and the ensuing FDA approved. Our choice of primary end point may have been overly optimistic, and although differences were detected, we also investigated other relevant end points, most of which favored omalizumab and none of which favored MOIT, even in the PP population. Finally, given the multifood allergies and low baseline CTDs, study participants may not be representative of the general food allergic population.

Conclusions

In summary, in this randomized clinical trial, although the ITT analysis demonstrated higher rates of treatment success with omalizumab across numerous end points, this difference was largely driven by the high rate of MOIT-related adverse events and study discontinuations rather than a difference in efficacy among those who completed treatment. Both omalizumab and omalizumab-facilitated MOIT represent viable treatment options, each with distinct risk-benefit profiles that must be weighed in the context of individual patient needs and preferences. Additional studies comparing OIT with anti-IgE strategies will be needed, including detailed cost-benefit analyses, but in the interim we will continue to gain valuable information as we personalize these treatments, or combinations thereof, in the clinical setting.

Supplement 1.

Trial Protocol and Statistical Analysis Plan.

Supplement 2.

eTable 1. Dosing Schedule for DBPCFCs

eTable 2. Multi-OIT Initial Dose Escalation (IDE) and Build-Up Dosing Schedule

eTable 3. Study Stopping Rules, Results, and Outcome and Algorithm for the Resumption of Study Activities After Lapses Due to Food Product Recall

eTable 4. Food-Specific Baseline Characteristics, Intention-to-Treat Analysis Population

eTable 5. Consumed at Least 1 Dose of 2000 mg (CTD ≥4044 mg) for All 3 Foods Without Dose-Limiting Symptoms, Intention-to-Treat and Per-Protocol Analysis Populations

eTable 6. Stage 2 Disposition, Intention-to-Treat Analysis Population

eTable 7. Successfully Consumed CTD ≥4044 mg for at Least 2 Foods and Individual Foods With Prespecified Multiplicity Adjustment, Intention-to-Treat Analysis Population

eFigure 1. Individual Participant Flow Through Study, Intention-to-Treat Population

eFigure 2. Success for Individual Foods at Each CTD Level, Intention-to-Treat Analysis Population

eFigure 3. Comparison of Efficacy for Select Prespecified and Ad Hoc End Points Between Age Groups, by Treatment Arm, Intention-to-Treat Analysis Population

eTable 8. Association of Last Recorded MOIT Dose Prior to Stage 2 DBPCFC and Consumption of at Least 1 Dose of 2000 mg (CTD ≥4044 mg) for All 3 Foods (Primary End Point) at the End of Stage 2 DBPCFC Within the Omalizumab-Facilitated MOIT Arm, Per-Protocol Analysis Population

eTable 9. Stage 2 Results Comparing Those on Placebo in Stage 1 With Those on Omalizumab in Stage 1, Within the Stage 2 Omalizumab-Facilitated MOIT Arm, Intention-to-Treat Analysis Population

eTable 10. DBPCFC Outcomes at the End of Stage 1 and Separately, DBPCFC Outcomes at the End of Stage 2 Intention-to-Treat Analysis Populations

eTable 11. Consideration of Population Differences Between Stage 2 Completers and Those With Early Withdrawal Within the Omalizumab-Facilitated MOIT Arm, Intention-to-Treat Analysis Population

eTable 12. Potential Effect of Study Interruption on Outcomes, Intention-To-Treat Analysis Population

eTable 13. Summary of Stage 2 Adverse Events for the Initial Open-Label Omalizumab Period (First 8 Weeks, Weeks 8-16, Weeks 16-60) Safety Analysis Population

eTable 14. Adverse Event by Category and Relatedness to MOIT/Placebo for MOIT Across Stage 2, Safety Analysis Population

Supplement 3.

Data Sharing Statement.