Omalizumab facilitates rapid oral desensitization for peanut allergy

Abstract

Background

Peanut oral immunotherapy (OIT) is a promising approach to peanut allergy but reactions are frequent and some patients cannot be desensitized. The anti-IgE medication omalizumab (Xolair) may allow more rapid peanut updosing and decrease reactions.

Objective

To evaluate if omalizumab facilitated rapid peanut desensitization in highly allergic patients.

Methods

Thirty-seven subjects were randomized to omalizumab (n=29) or placebo (n=8). After 12 weeks of treatment subjects underwent a rapid one-day desensitization of up to 250 mg of peanut protein, followed by weekly increases up to 2000 mg. Omalizumab was then discontinued and subjects continued on 2000 mg of peanut protein. They underwent an open challenge to 4000 mg peanut protein twelve weeks after stopping study drug. If tolerated, subjects continued on 4000 mg of peanut protein daily.

Results

The median peanut dose tolerated on the initial desensitization day was 250 mg for omalizumab versus 22.5 mg for placebo treated subject. Subsequently 23 of 29 (79%) subjects randomized to omalizumab tolerated 2000 mg peanut protein 6 weeks after stopping omalizumab versus 1 of 8 (12%) receiving placebo (p<0.01). Twenty-three subjects on omalizumab versus 1 on placebo passed the 4000 mg food challenge. Overall reaction rates were not significantly lower in omalizumab versus placebo treated subjects (OR=0.57 p=0.15), although omalizumab treated subjects were exposed to much higher doses of peanut.

Conclusion

Omalizumab allows subjects with peanut allergy to be rapidly desensitized over as little as 8 weeks of peanut OIT. In the majority of subjects, this desensitization is sustained after omalizumab is discontinued. Additional studies will help clarify which patients would benefit most from this approach.

Introduction

Food allergy remains the leading cause of anaphylaxis among children and its prevalence has been increasing over time.^1–4 In addition, food allergy is costly, increases anxiety, decreases patient and family quality of life, and may affect nutrition.⁵ Currently, the standard of care for food allergy is allergen avoidance and availability of self-injectable epinephrine. Peanut allergy is responsible for a large proportion of severe reactions to foods and, unlike milk and egg allergy, is typically life-long.^{6, 7} While recent work suggests that early introduction of peanut can protect against development of peanut allergy, almost 10% of high-risk infants evaluated in the first year of life showed sensitization to peanut, rendering them ineligible for early peanut exposure.⁸ Therefore, therapies for patients with established food allergies are clearly needed.

There has been substantial interest in oral immunotherapy as a treatment for peanut and other food allergies. Successful desensitization has been demonstrated for a number of foods including peanut, milk and egg.^9–17 Gradual exposure to increasing amounts of allergen can lead to the majority of subjects tolerating doses of food sufficient to prevent reaction upon accidental exposure. Unfortunately, most subjects undergoing OIT experience allergic reactions during desensitization with approximately 20% having severe reactions requiring injection of epinephrine and another 10–20% of patients being refractory to oral desensitization.^18–20 Furthermore, desensitization is less likely to be successful in patients with higher baseline food-specific IgE levels,^{10, 21} resulting in an increased risk of anaphylaxis with desensitization relative to continued avoidance.²⁰ While OIT allows some desensitized patients to develop sustained unresponsiveness, defined as persistence of protection after discontinuation of OIT, this occurrs in a small fraction of desensitized patients, generally those lower baseline food specific IgE levels.^{21, 22} Lastly, current food allergy OIT clinical studies are hampered by drop-out rates of up to 25% due to reactions and non-compliance over the long periods of desensitization (2–4 years).²³

Recently, omalizumab (Xolair, Genentech South San Francisco), a monoclonal anti-IgE antibody, has been used to decrease reactions during oral desensitization. This medicine, initially approved for the treatment of asthma in adults and adolescents, binds free IgE and prevents allergic reactions.²⁴ We and others have reported in three open-label studies that omalizumab facilitates oral desensitization, in particular allowing more rapid and successful escalation of allergen dosing in subjects with high-risk food allergy.^25–27 Furthermore, preliminary data suggests that interrupting IgE signaling during oral immunotherapy/antigen exposure may reverse established allergy through induction of both Th1 and regulatory T cells that inhibit the allergic response.^28–31

Here we report the results of the Peanut Reactivity Reduced by Oral Tolerance in an anti-IgE Clinical Trial (PRROTECT), the first randomized, double-blind, placebo-controlled, multi-center study of the efficacy of omalizumab in conjunction with peanut oral immunotherapy.

METHODS

Study Subject Selection

This phase II study was conducted at four institutions with approval from their respective Institutional Review Boards and from the U. S. Food and Drug Administration through an investigational new drug application. Full inclusion and exclusion criteria are summarized in the supplemental methods. All patients had positive testing for peanut by both skin prick and specific IgE and a significant reaction (see supplemental material) to a ≤50 mg peanut protein dose (cumulative dose of 88 mg peanut protein) in a double-blind placebo-controlled food challenge (DBPCFC). No patients were undergoing immunotherapy to environmental allergens and none had received omalizumab in the past. Written informed consent was obtained from parents or guardians of all subjects along with assent from children of appropriate age.

Randomization and Treatment

From June 2013 through September 2014, a total of 37 eligible participants were enrolled and randomized in a 3.5 to 1 ratio to receive omalizumab (N=29) or placebo (N=8) in conjunction with peanut oral immunotherapy (OIT). The selected randomization ratio was based on the power analysis and on the strong therapeutic effect of omalizumab observed in previous open-label studies. The study design called for 28 patients in the omalizumab arm, but one participant randomized to omalizumab withdrew before receiving the first injection and was replaced per the protocol. Dosing for omalizumab (and placebo) was in accordance with the Xolair Global Dosing Schedule with slight modification.

Rapid Oral Desensitization and Updosing Schedule

12 weeks after the first dose of study drug, subjects began OIT with an initial, rapid oral desensitization day consisting of up to 10 doses (0.5, 1, 2, 4, 8, 15, 30, 60, 120, 250 mg) of peanut protein (one peanut contains 200–250 mg of peanut protein). The doses were spaced 30 minutes apart, until either the 250 mg dose (cumulative 490.5 mg) was tolerated or a significant reaction occurred (see supplementary material). The next day, the highest tolerated dose from the previous day was given under medical supervision; subjects were then discharged to take this dose once daily at home for the next week. All patients continued after the rapid desensitization day, regardless of the highest peanut dose achieved, and returned the following week for weekly updosing. Subjects who tolerated the 250 mg dose on the first day returned for weekly doses of 375, 500, 625, 750, 1000, 1250, 1625, and 2000 mg of peanut protein. Subjects who tolerated less than 250 mg peanut protein on the first day underwent updosing in smaller increments up to twice a week (see supplementary material).

Discontinuation of Study Drug, OIT Maintenance

At week 19, the last dose of blinded study drug was given to subjects who were tolerating the 1625 mg dose and they progressed to 2000 mg at week 20. Subjects who failed to reach 1625 mg received additional doses of study drug while continuing to updose for a maximum of 6 additional weeks. The primary outcome of the study was the ability to tolerate 2000 mg of peanut protein 6 weeks after withdrawal of study drug, defined as at least one day during the preceding week when this dose was consumed with no reaction. If the 2000 mg dose of peanut protein was not reached after the six-week extension, study drug was stopped but weekly updosing continued as tolerated. Those who could not tolerate 250 mg peanut protein by week 19 were considered treatment failures and had the option to receive open label omalizumab with peanut OIT, but initial therapy remained blinded (that is patients and investigators were not aware which treatment these subjects had initially received) Subjects who transitioned to open label omalizumab continued on the highest dose of peanut protein tolerated while receiving omalizumab for 12 weeks, and then followed the same updosing protocol as in the blinded portion of the study.

Challenge to 4000 mg peanut protein

Study subjects tolerating a daily oral dose of 2000 mg peanut protein underwent an oral food challenge with peanut 12 weeks after their last dose of study drug (week 31 for subjects who tolerated 2000 mg of peanut protein at week 20). Three doses of peanut protein were given 30 minutes apart (1250, 1250, and 1500 mg) for a cumulative dose of 4000 mg. Subjects who reacted during the challenge maintained a 2000 mg daily dose of peanut protein for the reminder of the study. Subjects who did not react continued on 4000 mg peanut protein daily.

Assessments

Total peanut-specific IgE levels and SPT to peanut extract were assessed at baseline, 12 weeks after the last dose of study drug and at the end of the study. Adverse events (AEs), as defined by FDA regulation, were documented for the duration of the study and recorded and graded according to NCI-CTCAE version 4.03.³² Events not included in the NCI-CTCAE listing were graded grade 1–5 (mild, moderate, severe, life-threatening, death) according to the General Grade Definitions. Allergic reactions to peanut OIT were graded following the Bock criteria (see supplement).³³

Study Registration

The study was registered as NCT01781637 at www.clinicaltrials.gov

Statistical Analysis

Full details of the statistical analysis are in the supplementary appendix. Briefly, primary and secondary endpoints were compared between the study arms using Fisher’s exact test. Prespecified analyses were conducted using intention to treat (ITT). Reactions per dose were compared between study arms using Generalized Estimated Equations with logit link. P-values ≤0.05 were considered significant.

Role of the funding source

The sponsors of the study had no role in study design or data collection. After the study started DTU changed employment from Boston Children’s Hospital to Genentech, one of the sponsors. He subsequently was involved in data analysis, data interpretation and writing the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Results

Baseline characteristics and initial food challenge

Sixty-nine subjects underwent baseline food challenge to peanut. Thirty-seven of these subjects were randomized, 29 to omalizumab and 8 to placebo (Figure 1). Baseline characteristics, including dose of peanut protein eliciting a reaction, of the 37 subjects enrolled in the study are presented in Table 1 and were similar between groups. The median age of the subjects was 10 years, and the median peanut-specific IgE levels were 91 and 88 kU_A/L in the omalizumab and placebo groups, respectively. The median cumulative peanut dose eliciting a reaction during the baseline DBPCFC was 38 mg (range 0.5–88 mg) for subjects randomized to omalizumab and 88 mg for those randomized to placebo (range 13–88 mg). Seventeen subjects required treatment with epinephrine during the baseline food challenge.

Figure 1.

Study Enrollment, Randomization, Procedures and Outcomes. OL=Open Label and indicates subjects who were considered to have failed peanut OIT and went on to receive open-label omalizumab.

Table 1.

Baseline characteristics

Baseline Characteristics	N	Omalizumab	Placebo	Odds ratio (95% CI)	p-value
Age at baseline, years, median (range)	37	10 (7, 19)	10 (6, 17)		0.56
Female gender, N (% of study group)	15	11 (37.9%)	4 (50%)	OR 0.61 (0.13, 2.95)	0.69
Male gender, N (%)	22	18 (62.1%)	4 (50%)	Reference	.
Total IgE level, IU/L, median (range)	37	430 (65, 1172)	391 (125, 1511)		0.78
Peanut-specific IgE level, kUA/L, median (range)	37	91 (10, 535)	88 (26, 483)		0.57
IgE ratio, median (range)	37	0.28 (0.03, 0.58)	0.29 (0.06, 0.73)		0.73
Peanut skin test wheal size, mm, median (range)	37	11 (5, 18)	10 (7, 13)		0.49
Peanut skin test erythema size, mm, median (range)	37	33 (0, 58)	31 (24, 46)		0.80
DBPCFC Cumulative failure dose (peanut protein), mg, median (range)	37	38 (1, 88)	88 (13, 88)		0.08

Initiation of Study Drug and Oral Desensitization

Of the 29 subjects randomized to omalizumab [intention to treat (ITT) population], one subject withdrew before receiving the first dose of omalizumab and was replaced per the protocol; another withdrew before commencing OIT due to a hospitalization for an eating disorder judged to be unrelated to the study drug. Twenty-three of the 27 subjects remaining in the omalizumab group completed the 250 mg desensitization versus 1 of the 8 subjects treated with placebo (p=0.0003, Figure 2). The median tolerated dose after this first day of the desensitization process for subjects receiving omalizumab was 250 mg (490.5mg cumulative) versus 22.5 mg for those on placebo.

Figure 2.

Highest dose of peanut protein tolerated during initial rapid desensitization

The 24 subjects who successfully tolerated the rapid first day of the desensitization process (23 omalizumab treated and 1 placebo treated) had their peanut dose increased weekly up to 2000 mg of peanut protein (equivalent to 8–10 peanuts) over 8 weeks. Twenty-one of 29 omalizumab and 1 of 8 placebo treated subjects reached 2000 mg of peanut protein by study week 20 (p<0.01, ITT population).

Primary endpoint: ability to tolerate 2000 mg peanut dose 6 weeks after withdrawal of study drug

The primary endpoint of this study was the ability to tolerate a 2000 mg dose of peanut protein 6 weeks after stopping omalizumab or placebo injection. This was achieved in 23 of 29 of subjects randomized to omalizumab (79% of the ITT population,) and 23 of 27 who actually received peanut immunotherapy (85%) compared to 1 of 8 (12.5%) receiving placebo (p<0.01 for ITT population, Table 2).

Table 2.

Tolerance of 2000 mg of peanut protein after discontinuation of study drug

Outcome	Omalizumab	Placebo	Odds ratio (95% CI)	p-value
	N (% of treatment group)
Tolerated 2000 mg of peanut protein	23 (73.9%)	1 (12.5%)	26.8(2.7,262.3	<0.01
Failed to tolerate 2000 mg of peanut protein	6 (20.7%)	7 (87.5%)
Withdrew from study	3 (50.0%)	0 (0.0%)
Up-dosed to <250 mg, went on open-label	2 (33.3%)	6 (85.7%)
Up-dosed to x, where 250 mg ≤ × <2000 mg	1 (16.7%)	1 (14.3%)

Treatment failures

Six subjects (75%) in the placebo group and two (7.4%) in the omalizumab group were unable to tolerate 250 mg of peanut protein after 8 weeks of desensitization. These subjects were considered treatment failures, and received open label omalizumab and peanut OIT, as per protocol. All 8 were able to tolerate 2000 mg of peanut protein while receiving open-label omalizumab. In total, 31 of the 35 (89%) subjects [31/33 (94%) who received omalizumab and peanut OIT] were able to tolerate 2000 mg of peanut 6 weeks after stopping omalizumab. The two subjects (one initially randomized to omalizumab and one to placebo) who failed to reach this endpoint both had symptoms suggestive of Eosinophilic Esophagitis (EoE), one documented by esophageal biopsy (discussed below).

Oral challenge with 4000 mg peanut protein

Subjects able to tolerate 2000 mg of peanut protein continued on this dose and underwent an open peanut challenge with a cumulative dose of 4000 mg of peanut protein 12 weeks after discontinuation of study drug (a secondary endpoint). Twenty-two of 29 subjects randomized to omalizumab (22 of 27 who actually received peanut immunotherapy) and 1 of 8 randomized to placebo passed this challenge (Table 3, p= 0.002, ITT population). They continued to consume 4000 mg of peanut protein daily, in a form of their choosing. One subject who failed this challenge resumed taking peanut at 2000 mg daily. Sixteen of the 22 omalizumab treated subjects and the single placebo treated subject who passed the 4000 mg challenge maintained this dose for an additional 21 weeks (the end of the study). Six of the omalizumab treated subjects who passed the 4000 mg challenge subsequently had allergic reactions on the 4000 mg maintenance dose including 4 requiring epinephrine. All of these subjects were successfully converted to a 2000 mg peanut maintenance dose.

Table 3.

Oral food challenge (OFC) 12 weeks after last dose of study-drug

Outcome	Omalizumab	Placebo	Odds ratio (95% CI)	p-value
	N (% of treatment group)
Passed 4000 mg peanut protein OFC	22 (75.9%)	1 (12.5%)	22.0 (2.3, 211.1)	<0.01
Did not pass 4000 mg peanut protein OFC	7 (24.1%)	7 (87.5%)
Experienced allergic reaction during OFC	1 (14.3%)	0 (0.0%)
Tolerating <2000 mg and did not receive OFC	1 (14.3%)	1 (14.3%)
Withdrew from study prior to time point	3 (37.5%)	0 (0.0%)
Switched to open-label prior to time point	2 (28.6%)	6 (85.7%)

Safety data and reactions

Safety data were characterized by body system and Bock grade.³³ Focusing on the initial rapid desensitization (since most placebo subjects never tolerated 250 mg of peanut protein) 4 of 28 omalizumab treated subjects versus 6 of 8 placebo treated subjects had significant (Bock grade 2) reactions. Grade 1 reactions were seen in 16 of 28 omalizumab treated subjects versus 5 of 8 placebo treated. Overall, reactions occurred after 7.8% of OIT doses given in the omalizumab arm versus 16.8% in placebo despite omalizumab subjects receiving higher doses of peanut. This difference was not statistically significant (OR=0.57, p=0.15). Most reactions were mild (Bock grade 1) and localized to the gastrointestinal (GI) tract. The majority of reactions occurred during the initial desensitization day and subsequent dose escalation visits. No significant reactions were attributed to omalizumab or placebo administration.

There were seven grade 3 reactions judged to be study-related: 3 in subjects on blinded omalizumab, 1 on placebo and 3 in subjects on open-label omalizumab (see supplementary table 4). All reactions were reversible and were thought to be due to peanut exposure. A total of 14 reactions requiring epinephrine treatment occurred in 8 subjects on OIT. Seven reactions occurred in subjects on blinded therapy: 3 reactions in 2 patients in the placebo arm and 4 reactions in 3 patients in the omalizumab group. The remaining 7 reactions occurred in subjects on open label omalizumab. All subjects who received epinephrine recovered with no clinical sequelae and continued in the study. NSAID (ibuprofen) ingestion, viral infections and exercise were noted to be possible cofactors. One reaction requiring epinephrine occurred in a college student in the maintenance phase who did not take the peanut doses for 6 days and then, without contacting the site, took 4000 mg of peanut protein.

Three subjects developed persistent GI symptoms consistent with EoE and stopped peanut dosing. At study week 14, an omalizumab treated subject developed GI symptoms, that were consistent with EoE and was withdrawn from the study at week 17 due to persistent symptoms; endoscopy was not performed. A second omalizumab treated subject withdrew at extension week 2 due to persistent GI symptoms and an esophageal biopsy taken 11 days after stopping peanut while on proton pump inhibitor therapy showed 21 eosinophils/high power field, consistent with a diagnosis of EoE. A placebo treated subject developed persistent GI symptoms requiring reduction of peanut dose. Endoscopy done at week 23 while on high dose proton pump inhibitor therapy demonstrated EoE (peak 29 eos/hpf) and eosinophilic gastritis; peanut OIT was stopped. In all 3 subjects, symptoms resolved with discontinuation of peanut intake; follow up endoscopies were not performed.

Changes in immunologic parameters

Subjects treated with omalizumab showed decreased wheal size on skin testing but increased peanut-specific IgE values at week 31 compared to baseline, (P<0.0001 for both) consistent with data from other desensitization studies (Figure 3).^{9, 31, 34–36} Although Week 31 (11–12 weeks after stopping omalizumab) occurred at about 3 half-lives after last omalizumab dose (the half-life of omalizumab in the circulation is about 26 days)³⁷, the small amounts of omalizumab remaining in circulation at that time is thought to only minimally affect skin test reactivity.³⁸ Increased peanut-specific IgE and decreased SPT wheal size were also seen in subjects who received open label omalizumab. The single placebo treated subject who tolerated the 4000 mg food challenge, showed increases in both peanut-specific IgE and SPT.

Figure 3.

Changes in Immunologic Characteristics. Wheal size for peanut skin prick testing (A) and peanut specific IgE (B) are shown at baseline and week 31 of the study-after 12 weeks off study drug.

DISCUSSION

This randomized, double-blind, placebo-controlled study demonstrates that a short treatment course of omalizumab improves the safety and tolerability of peanut updosing during peanut OIT, allowing a large majority of subjects to achieve a 2000 mg maintenance dose of peanut protein in a median time of 8 weeks. As the first placebo controlled trial of omalizumab with peanut OIT, this study extends our previous work²⁵ and that of others,³⁹ and provides rigorous scientific confirmation that omalizumab can facilitate rapid OIT of high-risk peanut allergic subjects that is maintained even after omalizumab treatment is discontinued. Thus, omalizumab pretreatment allowed 23 of 29 subjects to tolerate 250 mg of peanut protein (cumulative dose 490 mg), roughly the amount found in 2 peanuts, during the first day of desensitization, compared to 1 of 8 placebo treated subjects (median dose 22.5 mg). Furthermore, 23 of 29 (76% of the ITT population) omalizumab treated subjects were able to quickly achieve a 2000 mg maintenance dose of peanut protein and maintain non-reactivity for 6 weeks after discontinuing omalizumab versus 1 of 8 placebo treated subjects. Although subjects likely retained some protection from omalizumab at this time point (the half-life of omalizumab is 26 days), successful achievement of this endpoint indicated that omalizumab could indeed facilitate rapid dose escalation to a 2000mg dose of peanut. Moreover, 16 omalizumab subjects tolerated 4000mg of peanut protein and an additional 7 tolerated 2000 mg 33 weeks after discontinuing therapy (>9 half-lives after the last omalizumab dose) demonstrating that the presence of omalizumab was not required to maintain the peanut-desensitized state.

The rapid desensitization in subjects treated with omalizumab in our study contrasts with the results of investigations utilizing oral peanut desensitization alone. For example, in a previous study of conventional OIT (without omalizumab), only 10 of 39 subjects achieved the highest initial dose of 50 mg of peanut protein on the initial desensitization day.⁹ In another study, Blumchen et al reported that only 5 of 22 subjects were able to tolerate 125 mg of peanut protein after a 7 day rush desensitization.²³ The final dose tolerated after this 7-day desensitization was not significantly higher than the dose subjects reacted to at baseline.

Moreover, in our study, the final target dose of peanut (2000 mg) was achieved in the majority of omalizumab-treated patients (21 of 29 subjects randomized to omalizumab, of whom 27 received both omalizumab and OIT) after only 8 weeks of desensitization. In contrast, other recent studies not utilizing omalizumab reported much slower increases in peanut dosing, ranging from 41⁴⁰ to 44 weeks¹¹ to reach a final dose of 800 to 4000 mg peanut protein. For example, Anagoustou et al¹⁰ demonstrated that peanut dosing could be increased to 800 mg of peanut protein over approximately 16 weeks, but only 24 of 49 subjects were able to tolerate 1400 mg peanut protein after 26 weeks of therapy. Thus, use of omalizumab allowed rapid desensitization to a high dose of peanut.

Wood et al.⁴¹ recently reported that omalizumab can facilitate milk OIT, and that rates of allergic reactions were significantly lower in omalizumab treated subjects, utilizing a long desensitization period (>2 years of treatment with omalizumab) and a post-treatment oral food challenge. In this study, 71.4% of placebo treated subjects (and 89% of omalizumab treated subjects) passed the final 10,000 mg oral food challenge, consistent with previous studies showing that slow, persistent OIT updosing without omalizumab can be successful. Successful treatment at Month 32 in this study was associated with lower milk-specific IgE levels (median 24 kUa/L), as has been previously suggested.¹⁰ Our study included patients with high peanut specific IgE levels (median 91 kUa/L) who may be at increased risk of having difficulty with conventional immunotherapy.^{21, 23, 34, 42–45}

The use of omalizumab allowed rapid desensitization to 2000 mg of peanut protein with low rates of peanut reactions and adverse effects. Although the reaction rates were not statistically lower with omalizumab (7.8% of doses) compared to placebo (16.8%), omalizumab treated subjects were exposed to much higher doses of peanut protein. For example, during the initial rapid desensitization day, grade 2 reactions were observed in 4 of 27 omalizumab treated subjects versus 6 of 8 placebo treated subjects, but omalizumab treated subjects reached a median dose of 250 mg peanut protein compared to 22.5 mg for placebo. The safety of omalizumab in facilitating OIT therefore may be underestimated because of the study design. It is important to note, however, that rates of reactions are likely to be increased in peanut allergic inidividuals undergoing OIT compared to continued avoidance.

Three subjects (2 treated with omalizumab) developed persistent GI symptoms consistent with EoE. Peanut consumption was stopped in these subjects leading to complete resolution of symptoms. EoE has been observed in other trials of OIT at rates comparable to those seen in this study,⁴⁶ consistent with the idea that omalizumab is not effective for treatment of EoE⁴⁷, although this study was not powered to evaluate rates of EoE. It is possible the rapid updosing used in this study increases the risk of EoE. Of note, of 27 patients randomized to omalizumab and who received peanut OIT, 23 tolerated 2000 mg of peanut protein during the double-blind phase of the trial, 2 more tolerated this dose after open label omalizumab, and 2 failed to achieve this endpoint due to EoE like symptoms. The inability to tolerate high doses of antigen due to persistent GI symptoms consistent with EoE remains a significant obstacle to successful use of OIT⁴⁸ and omalizumab does not seem to significantly decrease this risk.

Interestingly, a single placebo treated subject was able to meet all study endpoints, and 2 subjects who were treatment failures on blinded omalizumab were able to meet all study endpoints when they received open-label omalizumab. Baseline characteristics of these subjects were unremarkable. The latter 2 subjects suggest some subjects may benefit from a prolonged course of omalizumab prior to OIT; alternatively, the likelihood of developing allergic symptoms during OIT may be lower in the setting of open-label omalizumab administration.

A potential concern with use of omalizumab is that subjects may be at risk for allergic reactions once omalizumab treatment is stopped (week 19 of this study, just before subjects received their first 2000 mg dose of peanut). Six of 22 omalizumab treated subjects who passed the 4000 mg food challenge at week 31 had to reduce their peanut dose at some point over the following 21 weeks because of allergic reactions. However, all were able to then safely tolerate a daily maintenance dose of 2000 mg of peanut protein, suggesting the 2000 mg dose may be easier to tolerate. Missed doses of peanut, use of NSAIDs, exercise and concurrent infections may all increase the risk of reactions. Whether using omalizumab beyond the 20 weeks employed in our study would reduce the frequency of these reactions is not clear.

In our current study, we did not examine sustained unresponsiveness, although we plan to evaluate for this following additional time on maintenance therapy and after peanut-specific IgE levels fall to <10 kU/L (median peanut-specific IgE at week 52 was 70kU/L). Unfortunately, sustained unresponsiveness after OIT appears to be difficult to achieve even with prolonged intake of high daily doses of peanut, except in patients who begin OIT with low food-specific IgE levels^{21, 22, 41}. Whether high (e.g. 2000 mg) or low (e.g. 250 or 300mg) dose of peanut protein are optimal for induction of sustained unresponsiveness remains unknown. Further study of the most appropriate maintenance dose and the length of omalizumab and maintenance therapy would help to optimize this treatment. Given that peanut OIT does not consistently lead to sustained unresponsiveness, the benefits of protection against accidental exposure versus the risks of allergic reactions occurring during OIT must be balanced before considering the use of omalizumab facilitated OIT.

There are a several weaknesses of our study. First, the sample size was relatively small and included only 8 placebo subjects. We chose this low number of placebo treated subject because the power analysis, based on our previous open-label study, indicated that 8 placebo subject was more than sufficient, and because we hoped to limit the number of subjects exposed to the rapid updosing of peanut on the first day of desensitization (250mg maximum dose, cumulative 490 mg) that was expected induce allergic reactions in placebo treated subjects. While only 37 subjects were randomized, this is one of the largest trials of peanut desensitization to date and was sufficient to demonstrate a highly significant difference between the placebo and omalizumab treated groups in achieving the 2000 mg maintenance peanut dose.

Second, given the small number of placebo patients, the potent therapeutic effect of omalizumab in this situation, subjects may have deduced which therapy they were receiving. The ability to tolerate 250 mg on the initial updosing day would not be expected in placebo subjects and may have been a clue that subjects were getting active therapy. In addition, omalizumab may have improved control of concomitant asthma or allergic rhinitis. However, the 250 mg dose was achieved in one placebo-treated patient, and four omalizumab-treated patients did not achieve the 250 mg dose on the first updosing day, suggesting that the blinding was effective.

Finally, three previous studies showed that treatment alone with an anti-IgE monoclonal antibody (mAb) led to increased tolerance to peanut with reactions at median doses of 1315 mg³⁹, 500 mg⁴⁹ or 6500 mg of peanut⁵⁰, with wide variation within each study. It should be noted that subjects in the three studies had significantly lower median peanut-specific IgE levels (14.3 and 25.9 kUa/L, and one not reported) than in our study (91 kUa/L), smaller skin prick test to peanut (only reported in one study: 7.8 vs. 11 mm) and tolerated higher doses of peanut at baseline food challenge (e.g. 80 mg in the study by Savage et al.⁵⁰) Although some of our omalizumab-treated patients might have been able to tolerate 250mg or even 2000 mg of peanut in one day after omalizumab treatment alone, our experience is that most of our patient, particularly those with high baseline peanut specific IgE, require desensitization over several weeks to reach doses of 2000 mg. Most importantly, the goals of these three studies were significantly different from those of our omalizumab-facilitated OIT approach, in which omalizumab treatment is limited to 5 to 6 months as opposed to indefinitely.

OIT remains a promising approach for patients with peanut and other food allergies, but it is still not recommended for general use, in part because reaction rates are high and many patients are sensitized to more than one food allergen. However, our study demonstrates that peanut OIT in combination with omalizumab allows for rapid, effective desensitization in the majority of peanut allergic patients, including those with high peanut-specific IgE levels. While omalizumab is expensive, requires repeated injections and is associated with a risk of hypersensitivity reactions, we suggest that the benefits of omalizumab-enabled OIT may outweigh these downsides; additional studies will be required to further evaluate this proposition.

Clinical Implications.

Omalizumab allows rapid, safe desensitization of patients with peanut allergy. Desensitization is maintained after omalizumab treatment is stopped.

Capsule Sumary.

This placebo-controlled trial demonstrates that use of omalizumab permits rapid oral desensitization of patients with peanut allergy and that the desensitized state is maintained after omalizumab therapy is stopped.

List of abbreviations

DBPCFD

double-blind, placebo-controlled food challenge

EoE

eosinophilic esophagitis

GI

gastrointestinal

ITT

intention to treat

OIT

oral immunotherapy

OL

open label