|Home | About | Journals | Submit | Contact Us | Français|
Oral immunotherapy (OIT) has been thought to induce clinical desensitization to allergenic foods, but trials coupling the clinical response and immunologic effects of peanut OIT have not been reported.
The study objective was to investigate the clinical efficacy and immunologic changes associated with OIT.
Peanut-allergic children underwent an OIT protocol including initial day escalation, build-up, and maintenance phases, and then oral food challenge. Clinical response and immunologic changes were evaluated.
Of 29 subjects who completed the protocol, 27 ingested 3.9 g peanut protein during food challenge. Most symptoms noted during OIT resolved spontaneously or with antihistamines. By 6 months, titrated skin prick tests and activation of basophils significantly declined. Peanut-specific IgE decreased by 12–18 months, while IgG4 increased significantly. Serum factors inhibited IgE–peanut complex formation in an IgE-facilitated allergen binding assay. Secretion of IL-10, IL-5, IFN-γ, and TNF-α from PBMCs increased over 6–12 months. Peanut-specific FoxP3 T cells increased until 12 months and then decreased thereafter. Additionally, T cell microarrays showed downregulation of genes in apoptotic pathways.
OIT induces clinical desensitization to peanut, with significant longer term humoral and cellular changes. Microarray data suggest a novel role for apoptosis in OIT.
In industrialized countries, peanut allergy affects 0.8% of children and 0.5% – 1% of the general population,1–3 and the prevalence appears to be increasing. Peanuts and tree nuts account for the vast majority of life-threatening or fatal reactions to foods.4,5 Currently, the primary treatment for peanut allergy is a peanut-free diet and ready access to self-injectable epinephrine and antihistamines.6 Strict avoidance diets can be complicated by difficulty in interpreting labels7 and undeclared allergens in commercially prepared foods.8,9 Up to 50% of food-allergic patients have an allergic reaction during a given two-year period.10 The combination of avoidance diets and risks of accidental exposures and life-threatening reactions creates a tremendous burden to patients and families.
Traditional subcutaneous immunotherapy is useful in treating forms of inhalant allergen sensitivity such as allergic rhinoconjunctivitis and asthma11 but is unsafe in food allergy.12,13 Oral immunotherapy (OIT) and sublingual immunotherapy (SLIT) have been reported by our group and others to result in induction of clinical tolerance to a variety of food proteins.14–16 Yet most studies have not attempted to couple clinical efficacy with long-term immunologic changes.
We conducted an open-label study of peanut OIT in children with peanut allergy. Our goals were to evaluate the ability of peanut OIT to induce clinical desensitization and to investigate the immunologic mechanisms associated with clinical efficacy. The term “desensitization” was used to mean a change in threshold of ingested food antigen needed to cause allergic symptoms, while “tolerance” referred to the induction of long-term immunologic changes associated with the ability to ingest food without symptoms and without ongoing therapy. We hypothesized that peanut-allergic subjects who underwent OIT would be shifted toward a Th1-type profile.
Subjects, ages 1 to 16 years, were recruited from the allergy and immunology clinics or surrounding community physician offices at Duke University Medical Center and Arkansas Children’s Hospital. Ethics approval was obtained through the Institutional Review Boards at Duke University Medical Center and University of Arkansas for Medical Sciences. Written informed consent was obtained in accordance with each institution’s ethics guidelines for research in children.
Included subjects had a clinical history of reaction to peanut within 60 minutes of ingestion, a positive peanut skin prick test (≥ 3 mm of negative control), and a peanut CAP FEIA ≥ 15 kU/L (Phadia AB, Pharmacia, Inc., Uppsala, Sweden). Subjects were also included if they had a CAP FEIA ≥ 7 kU/L and a clinical reaction within the previous 6 months. Subjects were excluded for history of severe, life-threatening anaphylaxis (with hypotension) to peanut, severe or poorly controlled asthma, or a medical condition preventing undergoing a food challenge.
Peanut protein (from Partially Defatted Peanut Flour 12% Fat Light Roast, Golden Peanut Company, Alpharetta, GA; 2 g flour = 1 g peanut protein) doses were pre-measured and mixed in a food vehicle of the subject’s choosing and taken over two or three bites. Approximately 240 mg of peanut protein equals one whole peanut.17
Peanut OIT consisted of three phases: initial day escalation, build-up, and maintenance. Patients were instructed to otherwise eliminate peanut protein from their diet. During dosing, subjects were asked to keep a diary of any missed doses or adverse symptoms. Self-administered epinephrine was provided to all patients. A member of the study team was available by pager and phone at all times throughout the study.
The initial day escalation phase was undertaken at the research unit at each institution. Intramuscular (IM) epinephrine, oral and intravenous (IV) doses of diphenhydramine, and albuterol were at the bedside at all times. Dosing began at 0.1 mg peanut protein, followed by an approximate doubling every 30 minutes, up to 50 mg. The highest tolerated single dose was the starting dose for the build-up phase, which was initiated in the research unit the following day.
Subjects were instructed to ingest the daily dose of peanut protein with other safe foods in two or three bites at home every day. Doses were increased 25 mg every 2 weeks until 300 mg. Subjects returned to the clinic for dose escalations. For subjects who stopped dosing at lower than 50 mg on the initial day escalation, their doses were doubled every 2 weeks until they reached 50 mg, and then the increases were 25 mg. Build-up dosing was delayed if subjects had evidence of illness (eg, viral infection) at the time of scheduled up-dosing; therefore, the time to reach maintenance dosing varied between subjects.
After reaching 300 mg peanut protein daily, subjects continued this dose until the food challenge. After oral food challenge, subjects were increased to a daily OIT dose of 1800 mg if the peanut IgE remained > 2 kU/L after 12 months on maintenance dose (this escalation occurred in all subjects reported). Subjects were evaluated every 4 months while on continued maintenance dosing (total of 36 months).
The first cohort of subjects (n = 7) underwent an open OFC to peanut protein after 13–22 months maintenance OIT, and the second cohort (n = 22) did after 4–7 months. The time to OFC was reduced because early basophil and skin test data and as well as OFC data indicated a lack of clinical reactivity sooner than hypothesized. Prior to the OFC, subjects were asked to restrict use of antihistamines (short-acting, 72 hours; long-acting, 7 days), beta-agonists (12 hours), theophylline (12 hours), and montelukast (12 hours). The OFC consisted of 4 doses (300 mg, 600 mg, 1200 mg, 1800 mg) of peanut protein given every 30 minutes up to a total of 3.9 g of peanut protein (7.8 g of peanut flour). The OFC was discontinued at 3.9 g or with objective symptoms.
Peanut proteins were extracted from defatted peanut flour (Golden Peanut Co.) in PBS, clarified by centrifugation (30,000 × g for 30 min), and sterilized by filtration. The major peanut allergen, Ara h 2, was purified and lyophilized as previously described,18 diluted in PBS, and sterilized. All protein concentrations were determined using the bicinchoninic acid assay (BCA, Pierce).
Titrated skin prick testing with peanut extract (Greer Laboratories, Inc., Lenoir, NC) and saline and histamine controls were performed at enrollment, after 4 months of maintenance therapy, and every 4 months thereafter. Tests to peanut were measured and followed at the same dilution (either 1:20, 1:200, 1:2,000 or 1:20,000) that initially showed a wheal > 5mm. Wheal size was the average of the largest diameter and the perpendicular midpoint diameter. Data were analyzed using a mixed model repeated measures ANOVA. The response variable was the highest dilution causing > 5 mm wheal at enrollment. Time and subject were treated as factors. Inferences about wheal size changes over time were made by comparing the mean at each time point back to time zero using a multiple comparisons procedure. Restricted maximum likelihood (REML) was utilized to fit the ANOVA model and estimate model parameters.
Basophil activation was measured as previously described.19 Briefly, peripheral blood was collected in sodium heparin tubes, aliquoted, and stimulated for 30 minutes with basophil medium alone (RPMI with 4 ng/mL human IL-3) or the medium with 10, 1, or 0.1 μg/mL peanut extract; 1 μg/mL anti-IgE (polyclonal rabbit anti-human, Bethyl Laboratories); or 2 μM fMLP (VWR Scientific). Cells were stained for 30 minutes at 4°C with the following monoclonal antibodies: CD63-FITC (clone H5C6), CD203c-PE (IM3575), CD123 PE-Cy5 (9F5), CD69-APC-Cy7 (FN50), CD3-APC (SK7), CD14-APC (M5E2), CD19-APC (H1B19), CD41a-APC (HIP8), and HLA-DR-PE-Cy7 (L243) (IM3575 Beckman Coulter, all others BD Biosciences). CD63 upregulation was assessed by flow cytometry. CD3, CD14, CD19, and CD41a positive events were excluded, and a minimum 1,000 CD123+ HLA-DR- events were acquired. Data were analyzed using FlowJo software (TreeStar, Ashland, OR).
Peanut-specific IgE, IgG, and IgG4 levels were measured in serum samples using the ImmunoCAP 100 instrument (Phadia) according to the manufacturer’s instructions. The same statistical approach as for the titrated skin tests analysis was carried out, except that the natural log of the immunoglobulin concentrations was taken to better meet the constant variance and normality assumptions of the ANOVA model.
Indicator serum containing high concentrations of peanut-specific IgE (RAST > 100 IU/mL) was purchased from PlasmaLab, Everett, WA. Equal volumes (10 μL) of serum obtained from the clinical study and indicator serum were incubated with peanut extract (0.04 μg/mL in 2.5 μL) for 1 hour at 37°C.20 Results of flow cytometry are expressed as relative binding, where binding observed by indicator serum alone is normalized to 100%, and changes in binding caused by the addition of patients’ serum to the indicator serum is related to this value. Statistical differences between pre-OIT and post-OIT sera were determined using SPSS 15.0 for Windows (SPSS Inc.). The 2-tailed Wilcoxon signed ranks test was used to compare pre- and post-OIT sera ability to inhibit peanut- specific FAB. P values < 0.05 were considered statistically significant.
PBMCs were isolated from ~25 mL heparinized blood using Ficoll-based density separation (LymphoH, Atlanta Biologicals). For cytokine assays, suspended PBMCs were distributed into 96-well flat-bottom plates at a concentration of 4 × 105 cells/well in triplicates and incubated with crude peanut protein (40 μg/well), Ara h 2 (20 μg/well), concanavalin A (8 μg/well, Sigma), or medium alone (RPMI-1640 with 2 mM L- glutamine, 25 mM Hepes buffer containing 10% human AB serum, 100 IU/mL penicillin and 100 μg/mL streptomycin, Mediatech). Cells were cultured at 37°C in 5% CO2 humidified atmosphere for 24, 48, and 96 hours. Culture supernatants were collected at each time point and analyzed in duplicates for 14 different analytes using a multiplex bead assay (R&D Systems) for the Luminex 100 platform. To analyze the cytokine data, linear mixed effects models were run in Splus (Insightful Co.) with subject as the random effect, and fixed effects given by culture condition, culture condition × months on immunotherapy, and time of culture. The response variable was log(y+1), where y is the mean cytokine concentration. Slope comparisons were against the null hypothesis that slope = 0 for RPMI. A positive or negative coefficient was considered statistically significant at the 0.05 level and was a measure of the trend over time of each cytokine.
For flow cytometry, PBMCs (2 × 106 cells/well) were cultured in 24-well plates under the same stimulation conditions as above. After 6 days, cells were collected and stained with fluorescent monoclonal antibodies: anti-CD3-PerCP, CD4-FITC, and CD25-PE (BD Biosciences). Additional intracellular staining with anti-Foxp3-APC was carried out after fixation/permeabilization of the cells (eBioscience). Isotype controls were included for each condition. The samples were run for 3-color detection in a FACSCalibur flow cytometer (Beckman-Coulter). At least 10,000 events were acquired for each experimental condition, and data were analyzed using the FlowJo software.
RNA isolated from resting PBMC CD3+ T cells (EasySepTM T cell Enrichment, Stem Cell Technologies, Inc., Vancouver, Canada) with the RNeasy Total RNA Isolation kit (Qiagen, Inc., Valencia, CA) was used for target preparation and hybridization with the GeneChip human genome U133 Plus 2.0 array (Affymetrix, Inc., Santa Clara, CA) according to the manufacturer instructions. Hybridized microarrays were scanned using an Affymetrix GeneChip 3000 scanner. Microarray assays and statistical analyses of experimental data were performed by Expression Analysis, Inc., Durham, NC, and included assessment of data quality by standard quality checks and principal components analysis (PCA) by sample of the probe-level data, along with normalization and signal summarization using the robust multi-array (RMA) algorithm. Determination of differential expression of genes in subject samples before and after OIT was performed by repeated measures analysis accounting for multiple testing using a variant of Significance Analysis of Microarrays (SAM)21 to detect statistically significant transcripts. Enrichment analysis of the set of transcripts identified as being differentially expressed between subjects by repeated measures analysis was then performed by GeneGo, Inc., (St. Joseph, MI) utilizing the MetaCore software suite (GeneGo). This enrichment analysis matched Entrez (National Center for Biotechnology Information, National Institutes of Health, Bethesda, MD) geneIDs for common, similar, and unique sets of the OIT subjects differentially expressed transcript list with Entrez gene IDs in functional ontologies in MetaCore. The ontologies included canonical pathway maps, GeneGo cellular processes, gene ontology cellular processes, and disease categories. The degree of relevance to different categories for the OIT subjects dataset was defined by P values, so that the lower P value received higher priority.
Resting PBMC CD3+ T cell RNA isolated for microarray assays was used for cDNA synthesis and quantification of experimental and control (18s rRNA) transcripts by Expression Analysis, Inc., using a 7900 HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA) and TaqMan Gene Expression Assay (Applied Biosystems) gene-specific primer and probe sets. Experimental transcript levels were normalized to those for 18s rRNA in each sample.
Thirty-nine subjects were enrolled. Twenty five (64%) were male. The median age at enrollment was 57.5 months (range, 12–111 months). The median age at first reaction to peanut was 15 months (range, 8–48 months). All but 3 subjects had allergic disease other than food allergy, including atopic dermatitis (69%), asthma (62%), and allergic rhinitis (62%). And 54% had an additional food allergy: 38% tree nuts, 23% egg, 13% cow’s milk, 5% fish, and 3% soy.
All 39 subjects completed the initial day escalation protocol. Ten (25%) subjects subsequently withdrew. Six discontinued for personal reasons, including transportation issues, parental anxiety, and failure to perform home dosing. These six had reactions during the initial escalation day that were similar to patients who continued in the study. The remaining 4 subjects discontinued because of allergic reactions to the OIT that did not resolve with continued treatment or dose reduction. Three had gastrointestinal complaints, and 1 had symptoms of asthma. Twenty-nine subjects completed all 3 phases of the study and peanut challenges.
During the initial day escalation, 10 (26%) subjects tolerated the highest dose of 50 mg peanut protein (Table 1), 15 (38%) tolerated 25 mg, 6 (15%) tolerated 12 mg, 5 (13%) tolerated 6 mg, 1 (3%) tolerated 3 mg, and 2 (5%) tolerated 1.5 mg. Thirty-six patients (92%) experienced some symptoms during the initial escalation day. Most common were upper respiratory symptoms, with 27 patients (69%) reporting mild sneezing/itching and mild laryngeal symptoms. No patients experienced severe upper respiratory or laryngeal symptoms. Seventeen patients (44%) reported mild to moderate nausea or abdominal pain, and 8 (21%) of patients suffered mild diarrhea/emesis. Twenty-four subjects (62%) had mild or moderate skin symptoms. A total of 6 patients experienced chest symptoms during the initial escalation day; all 6 had mild wheezing, and 2 progressed to moderate wheezing. Three of the subjects with chest symptoms during the initial day escalation also had a prior diagnosis of asthma.
Subjects had symptoms after 46% of build-up doses. Subjects were on maintenance dosing at home prior to oral food challenge for a median of 4.7 months (range, 4–22). All subjects experienced rare, and typically minor, symptoms during some point of home dosing (3.7% of 14,773 doses given) (Table 1). Upper respiratory (1.2%) and skin (1.1%) were the most common. Treatment was given with 0.8% of home doses. Only 2 subjects received epinephrine after home dosing, and each of the two had only one such incident.
Twenty-nine subjects participated in the open OFC to peanut. Overall, 27/29 (93%) reached the total peanut dose of 3.9 gm with no more than mild symptoms, suggesting successful desensitization to peanut protein. Two subjects did not ingest the maximal dose and stopped after 2.1 g. One stopped because of parental anxiety, and the other because of mild urticaria and one vomiting episode.
Titrated skin prick tests showed a significant decrease of 4 mm beginning at 6 months (P < 0.0001) and remained decreased throughout the study.
Basophil reactivity to peanut antigen was evaluated at three peanut concentrations, 10 μg/mL, 1 μg/mL, and 0.1 μg/mL, and four time points: prior to OIT (n = 15), < 4 mos (n = 9), 4–6 mos (n = 6), and > 6 mos (n = 4) (Figure 1). At a peanut concentration of 10 μg/mL, basophil activation was significantly reduced within 4 months (P < 0.001). We also evaluated basophil reactivity in subjects in an observational study of peanut allergy, and these patients did not experience reduced basophil activity.
The initial median concentration of serum peanut-specific IgE was 85.4 kU/L (range, 9.1 to 840.0 kU/L). After 3 months of treatment, median peanut-specific IgE increased nearly three-fold (249.0 kU/L, P < 0.0005). At 12 and 18 months, no significant decrease from baseline was found, but for all subsequent time points (up to 33 months), peanut-specific IgE levels were significantly decreased (P < 0.0005) (Figure 2A).
Median baseline serum peanut-specific IgG was 9.7 mg/L (range, 2.5 to 56.0 mg/L). A significant increase (P < 0.0005) in specific IgG levels also started at 3 months (Figure 2B). Specific IgG levels remained high until 24 months and slowly returned to baseline by 33 months.
The peanut-specific IgG4 followed a slightly different trend (Figure 2C). Initial concentrations were low, with a median of 0.3 mg/L (range, 0.1 to 1.4 mg/L). Peanut-specific IgG4 concentrations increased initially, reaching statistical significance at 3 months (2.0 mg/L vs. 0.3 mg/L, P < 0.0005), and continued elevated at the end of the study (P < 0.0005).
FAB inhibition by serum factors was tested in 20 patients at baseline and after 12 months treatment. A decrease in percent relative binding following 12 months of OIT was measured in 18 of the 20 subjects (Figure 3). Peanut-allergic subjects from our egg OIT study16 currently avoiding peanuts were used as controls and showed no change in relative binding (data not shown). For the peanut OIT subjects, the percent mean relative binding decreased from 87.6 ± 23.4% at baseline to 69.3 ± 23.3% by 12 months (P < 0.001).
A panel of 14 cytokines was measured in the supernatants of PBMCs incubated for 24, 48, and 96 h with peanut, Ara h 2, Con A, or medium alone (RPMI) for the first 5 subjects every 6 months for a period of 2 years on OIT. As expected, a more robust secretion of cytokines was measured after ConA stimulation, enabling measurement of otherwise undetectable cytokines (Figure 4). Cytokines including IL-5, IL-10, IFN-γ, and TNF-α significantly increased, as did the growth factor G-CSF, whereas IL-2 declined. IL-4 and IL-17 were undetectable at baseline and remained so, while many inflammatory mediators (IL-1β, IL-6, IL-8, MIP-1β, and GM-CSF) were found at saturating levels (data not shown).
Following peanut stimulation, a number of inflammatory cytokines/chemokines were significantly increased over time, including IL-1β, IL-5, TNF-α, MIP-1β, as well as the growth factors G-CSF and GM-CSF (Figure 5). Saturating levels of IL-6 and IL-8 prevented the delineation of a trend, and no significant change in the monocyte chemoattractant MCP-1 was observed during immunotherapy (data not shown). After peanut stimulation, no detectable levels of secreted IL-2, IL-4, IL-10, IL-17, and IFN-γ were measured (data not shown). Cytokine levels detected following stimulation with a single allergen, Ara h 2, were not different from those with medium alone (data not shown).
In 10 subjects who received peanut OIT for up to 36 months, a subpopulation of FoxP3 positive T cells were investigated by flow cytometry in the lymphocyte gates of PBMCs incubated for 6 days with medium alone (RPMI), peanut, or Ara h 2 (Figure 6). During OIT, the number of FoxP3 T cells increased approximately 1.5-fold in peanut-stimulated cells at 6 and 12 months (P < 0.05) and then decreased thereafter, returning to baseline levels by 20 months. Ara h 2 stimulation created a similar yet less pronounced increase.
Genome-wide oligonucleotide microarray analyses compared transcription patterns in T cells obtained from 6 unrelated subjects prior to starting OIT and 6 months after uncomplicated OIT. Differential expression of genes in subject samples before and after OIT was determined by repeated measures analysis and yielded 450 transcripts with a false discovery rate of < 7%. A reduced, nonrepetitive subset of 334 genes having a well-described Entrez Gene ID (Supplementary Table 1) was then submitted to GeneGo, Inc., for enrichment analysis. The three canonical signaling and metabolic pathways most affected by OIT were all involved in apoptosis, and all differentially expressed transcripts in these pathways were downregulated after 6 months of OIT (Table 2). Quantitative real-time PCR of selected samples confirmed the observed downregulation of BCL2L11, GADD45A, TNFSF8, and RELA gene expression in three subjects following OIT (data not shown). Further enrichment analysis of 110 cellular and molecular processes whose content is defined and mapped by GeneGo, with each process representing a preset network of protein interactions characteristic for the process, also demonstrated a statistically significant alteration of apoptosis networks following OIT.
In this clinical and mechanistic study, peanut OIT induced clinical desensitization in the 29 peanut-allergic subjects who completed the study. Ninety-three percent successfully completed an OFC to 3.9 g peanut protein, and all subjects had a significant increase in the amount of peanut they tolerated during food challenge. Peanut OIT was also safe; mild symptoms were relieved with diphenhydramine or albuterol. Our results are consistent with prior studies in which OIT led to clinical desensitization to foods such as egg and cow’s milk.14–16,22 Furthermore, the humoral and cellular responses associated with peanut OIT suggest that OIT also induces the transition from short-term desensitization to long-term tolerance. For this analysis, we did not perform OFCs after cessation of therapy, when sensitivity could return. Per the protocol, these definitive challenges are planned for subjects who complete 3 years of maintenance therapy and have a significant drop in serum IgE. However, compared with prior OIT studies, our study had a longer duration of maintenance therapy, which we hypothesize has a significant impact on the immunologic parameters indicative of long-term tolerance.
In our study, titrated prick skin tests showed a significant decrease at 6 months and remained decreased throughout the study. Similarly, basophil activation, a measure of IgE-dependent response, decreased significantly within 4 months, and the decline continued beyond 6 months. IgE-mediated hypersensitivity responses are known to be downregulated during drug desensitization,23 and chronic FcεRI signaling induces a downregulation of Syk-dependent signal transduction in vitro.24
With peanut OIT, peanut-specific IgE, IgG, and IgG4 increased by approximately 3 months, and then the IgE declined by 18 months. IgG began to decrease by the end of the study, while IgG4 remained elevated. Increased levels of specific IgG4 with or without decreased IgE have been associated with successful venom immunotherapy,25 lower levels of atopy in the presence of parasite infection,26 transient rather than persistent milk allergy,27 and the apparent protective effect of high levels of cat allergen exposure.28,29 Previous reports have demonstrated that fractionated IgG4 antibodies from serum of patients who received grass pollen immunotherapy inhibit IgE–FAB binding to B cells,30–33 suggesting a functional role of IgG4 in inhibiting IgE–FAB.
Traditional allergen-injection immunotherapy appears to act through downregulation of allergen-specific Th2 responses or increased Th1 responses or through the induction of T regulatory cells. Populations of both thymus-derived CD25+ “natural” T cells and antigen-specific T cells become CD25+, express FoxP3, secrete IL-10, and have suppressive function. IL-10+ T cells are induced during venom, dust mite, birch, and grass pollen immunotherapy.34–36 In our study, FoxP3 regulatory T cells increased after the induction of OIT and then eventually decreased. IL-10 was significantly increased over 6 to 12 months, as were a number of inflammatory cytokines/chemokines, such as IL-1β, IL-5, TNF-α, MIP-1β, and the growth factors G-CSF and GM-CSF. These changes did not reflect the typical transition toward a Th1 profile that we expected. However, the early induction of regulatory T cells expressing FoxP3 and the associated increase in IL-10 indicate an immunologic change induced by OIT, with transition away from a Th2-type profile that was seen with both nonspecific (ConA) and antigen-specific (peanut) stimulation over time.
Our microarray data demonstrating downregulation of genes in several apoptosis pathways in patient T cells following 6 months of OIT are intriguing and may reflect involvement of programmed cell death in peanut OIT. However, it is unclear from these results if the observed changes in total peripheral blood T cell transcription patterns include altered apoptosis of antigen-specific T cells. To help clarify this point, studies are underway to compare transcript patterns before and after OIT in peanut-specific T cells isolated using MHC class II/Ara h 2 peptide tetramers. The lack of treatment-related changes in expression of Treg-, Th1-, or Th2-specific genes by microarray versus by protein assays likely reflects the small number of FoxP3-producing cells and low cytokine transcription levels in unstimulated CD3+ T cells analyzed in microarrays.
To our knowledge, no other oligonucleotide microarray analyses of patient T cell transcription patterns pre- and post-treatment of food allergy have been reported. In one microarray study of PBMC transcripts in 8 subjects with allergic rhinitis, several apoptosis-related genes were underexpressed compared with control PBMC prior to allergen immunotherapy.37 A small number of studies have noted increased in vitro apoptosis of stimulated peripheral blood Th2 cells following standard allergen immunotherapy in subjects with either grass pollen allergy38,39 or dust mite–sensitive asthma.40 We plan similar flow cytometric analyses of apoptosis in patient T cells following in vitro stimulation with peanut antigen.
Taken together, our results suggest that OIT induces a progression toward tolerance starting with desensitization at approximately 3 months. During this time, the threshold of antigen needed to induce an allergic response changes drastically, as reflected by diminished reaction to skin prick tests and activation of basophils. Subsequent immunologic changes over 6-to-12 months reflect a pro-inflammatory, rather than Th2, profile.
In our study, results of titrated prick skin tests, levels of allergen-specific IgE, IgG, and IgG4 over time, and FAB data are similar to those reported from studies of traditional subcutaneous immunotherapy.25,30–36 Our cytokine data, with a significant increase in IL-10 and a number of inflammatory cytokines/chemokines, are not reflective of the typical transition toward a Th1 profile. The increase in IL-10 could support an initial increase in Treg cells, leading to tolerance, but the overall increase in the inflammatory cytokines/chemokines is not really suggestive of this expected change. The inflammatory response may result from the oral versus subcutaneous route of exposure, although exactly how is unclear. No similar studies comparing OIT and traditional subcutaneous allergy immunotherapy have been done that might provide a context for our basophil and microarray data.
Clinical desensitization, which we defined as raising the threshold of food antigen needed to cause allergic symptoms, can provide an improved margin of safety in case of accidental food ingestion. This is an important therapeutic benefit to patients and their families. Blinded, placebo-controlled studies with peanut OIT are underway now, as are studies to determine the ability of OIT to induce long-term clinical tolerance after discontinuing OIT.
Sources of support: Food Allergy and Anaphylaxis Network, Gerber Foundation, National Institutes of Health - 1R01-AI068074-01A1, Arkansas Biosciences Institute, Dorothy and Frank Robins Family, Food Allergy Project, Clinical and Translational Science Award 5M01-R000030-45
We would like to acknowledge discussions about his project with the late Larry Katz, PhD (formerly J.B. Duke Professor of Neurobiology and Investigator, Howard Hughes Medical Institute).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.