Epidemiology of food hyper-sensitivity in school children

Validation with double-blind placebo-controlled food challenges and biomarkers

The Obstructive Lung Disease in Northern Sweden (OLIN) Studies, Thesis XV

Anna Winberg

Department of Clinical Sciences, Pediatrics

Umeå University 2016

Responsible publisher under Swedish law: the Dean of the Medical Faculty

This work is protected by the Swedish Copyright Legislation (Act 1960:729)

ISBN: 978-91-7601-367-0

ISSN: 0346-6612-1763

Elektronisk version tillgänglig på http://umu.diva-portal.org/

Tryck/Printed by: Print och Media

Umeå, Sweden 2016

PROLOGUE

Bright eyed insight…

During our research travels in Northern Sweden, Åsa - my dear friend and fellow musketeer - and I - were still in denial regarding our uprising presbyopia. Due to our visual impairment we spent an absurd amount of time trying to decipher our study protocols, which, for some reason, were written in ridiculously small letters. During a food challenge series, one of our study participants sat watching us for a very long time and then he bursted out:

Ah! Now I understand what you mean by double-blind!

To all the amazing children who made this thesis possible…

i

Table of Contents

Table of Contents i Abstract iii Sammanfattning på svenska v Abbrevations viii Original Papers ix Introduction 1

Personal point of departure 1 Background 3

Definition of food hypersensitivity (FHS) 3 Prevalence of food hypersensitivity 3 Prevalence of food allergy 3 Prevalence of lactose intolerance 4 Factors associated with food hypersensitivity 5 Risk factors of food allergy 6 Risk factors of lactose intolerance 7 Immunological aspects of food allergy and tolerance development 8 IgE-mediated food allergy 9 Non-IgE-mediated and mixed-immune food allergies 11 Cytokine patterns in tolerance development and IgE-mediated food allergy 11 Allergic inflammation in the intestinal mucosa 12 Symptoms of food allergy 13 Symptoms of IgE-mediated food allergy 13 Symptoms of non-IgE mediated and mixed-immune food allergies 14 Establishing a food allergy diagnosis 15 Clinical history 15 IgE-tests 16 Oral food challenges 17

Open challenges 18 Blinded challenges 18

Complementary diagnostic tests and future prognostic biomarkers 19 Endoscopy 20 Inflammatory and regulatory makers in faeces 20 Atopy patch tests 20 Serological tests 21 Basophil activation tests 22 Cytokines 23

Objectives 25 Materials and methods 27

ii

1. Method: Paper I 27 Method: The OLIN studies 29 Study population 29 Questionnaires 30 Skin Prick Tests 30 Child interview 31 Serological tests 31 Statistical analyses 32 Definitions 33 2. Method: Paper II 37 3. Method: Paper III 39 4. Method: Paper IV 43 5. Method: Paper V 45

Results 49 Validation of new DBPCFC recipes (Paper I) 49 Incidence and remission of food hypersensitivity (Paper II) 50 Assessment of allergy to milk, egg cod and wheat (Paper III) 53 Milk hypersensitivity phenotypes (Paper IV) 54 Biomarkers in blood and stool samples (Paper V) 59

Discussion 63 Discussion of methodology 63 Sensory testing 63 Epidemiologic study design 63 Reported data 65 Analyses of associated risk factors 66 IgE-tests 67 Validation of reported data 68 The DBPCFC methodology 69 Cytokine mRNA expression in PBMC 69 Analyses of fecal biomarkers 70 Discussion of main results 71 Epidemiology of food hypersensitivity 71 Diagnosing food allergy 73 Ethical considerations 75 Personal point of closure 77

Acknowledgements 79 Funding 83 References 84 Supplement 101

Questions from structured interviews 101

iii

Abstract

Background

This thesis focuses on the incidence and remission of reported food

hypersensitivity in schoolchildren followed from 8 to 12 -years of age and the

prevalence of hypersensitivity to milk, egg, cod and wheat among 12-year

olds investigated by reported data, clinical investigation and double-blind

placebo-controlled food challenges and biomarkers.

Methods

The studies are mainly based on a population based cohort recruited in 2006

from three municipalities in Northern Sweden. All children in first and

second grade, aged 7-8 years, were invited to a parental questionnaire study

and 2585 (96% of invited) participated. The children in two of the

municipalities were also invited to a skin prick test with airborne allergens.

At age 11-12 years, there was a follow-up of the cohort using the same

methods, with the addition of a child interview and assessment of body mass

index (BMI).

At the follow-up, children who reported milk hypersensitivity were invited to

structured interviews and children reporting complete elimination of milk,

egg, cod or wheat due to perceived hypersensitivity were invited to a clinical

examination and blood sampling. According to test results, the children were

categorized into different food hypersensitivity phenotypes according to

preset criteria. Children categorized as current food allergy were then invited

to further evaluation with a double-blind placebo-controlled food-challenge

using newly developed recipes. Before their use, the recipes were successfully

validated regarding detectable sensorial differences between the active and

placebo substances in a separate cohort of healthy schoolchildren (n=275).

Before and after the challenge series blood samples were collected for

analyses of cytokine mRNA expression in peripheral blood mononuclear

cells including hallmark cytokines for the humoral allergy-promoting T

helper (Th) 2 response, cellular cytotoxicity-promoting Th1 response,

inflammatory-, and T regulatory responses. Fecal inflammatory biomarkers

were also analyzed before and after the challenge series.

iv

Results

Reported food hypersensitivity increased from 21% at age 7-8 years to 26%

at 11-12 years. There was a high incidence (15%) as well as a high remission

(33%) of reported food hypersensitivity. Risk factors associated with

incidence and remission were different for milk hypersensitivity and

hypersensitivity to foods other than milk. The agreement between reported

symptoms to milk, egg, cod, wheat, soy and peanut and sensitization to the

culprit food was poor. At 11 to 12-years of age the prevalence of reported

allergy to milk, egg, cod or wheat was 4.8% while the allergy prevalence

according to clinical evaluation was 1.4%. This figure was further halved

when evaluated with double-blind placebo-controlled food challenges.

The majority of children with reported allergy to milk, egg, cod and wheat

were categorized as other food hypersensitivity phenotypes, the most

common being probable lactose intolerance (40%) and outgrown food

allergy (19%). Even though reported milk hypersensitivity among the 11-12

year olds was 14.5%, only 3% were categorized as current milk allergy.

Current and outgrown milk allergy was associated with other atopic

disorders and a lower BMI (OR 0.8, 95% CI 0.80-0.98). Before the challenge

series, the mRNA expression of the cytokines IL-13 and IL-10 were higher

among children with a positive compared to a negative challenge outcome.

Conclusion

Reported food hypersensitivity was common among school children in

Northern Sweden and increased from 7-8 years to 11-12 years of age, and

both the incidence and remission of reported hypersensitivity was high.

There was an 8-fold difference in the prevalence of allergy to milk, egg, cod

or wheat when reported data was assessed by clinical examinations and

double-blind placebo-controlled food challenges. Allergy to milk, egg, cod

and wheat was an uncommon cause of complete avoidance of these foods

due to perceived hypersensitivity. Some of the analyzed biomarkers might

serve as prognostic markers for symptomatic, IgE-mediated food allergy but

need further validation.

v

Sammanfattning på svenska

Bakgrund

Födoämnesöverkänslighet håller på att bli ett stort och kostsamt

hälsoproblem i västvärlden. Prevalensen av rapporterad födoämnes-

överkänslighet bland barn ökar, men det är fortfarande oklart om detta

avspeglar en sann ökning i populationen. Det finns en stor spridning mellan

uppmätta prevalenstal i olika studier och i de få studier där man följt upp

rapporterade data med objektiva metoder ses en hög överrapportering.

Data saknas om reell prevalens av födoämnesöverkänslighet bland skolbarn i

Sverige. Befintliga prevalensdata baseras på rapporterad födoämnes-

överkänslighet och studier saknas där angiven födoämnesöverkänslighet i en

barnpopulationskohort validerats med objektiva metoder. Även om

dubbelblinda provokationer räknas som ”gold standard” används i praktiken

endast sjukhistoria, pricktest och analys av specifikt Immunoglubulin E

(IgE) samt öppna provokationer för diagnostik. Metoderna har flera felkällor

och mer tillförlitlig diagnostisk behövs, särskilt vid sena och svårtolkade

symtom. Korrekt diagnos är särskilt viktig när baslivsmedel har eliminerats

eftersom kostrestriktionerna riskerar att leda till negativa konsekvenser för

livskvalitet och intag av viktiga näringsämnen.

Syfte

Studierna i denna avhandling fokuserade på incidens och remission av

födoämnesöverkänslighet bland skolbarn i Norrbotten, som följdes från 7-8

år till 11-12 års ålder, och på prevalens av överkänslighet mot mjölk, ägg, fisk

och vete bland 12-åringar, undersökt med rapporterade data, klinisk

undersökning samt dubbelblinda placebokontrollerade födoämnes-

provokationer och biomarkörer.

Metod

Den huvudsakliga delen av det här projektet är baserat på en populations-

baserad kohort som rekryterades 2006 från 3 kommuner i norra Sverige.

Föräldrar till alla barn i klass 1 och 2 (7-8 år) bjöds in till ett frågeformulär,

som besvarades av 96% (n=2585) av de inbjudna. Barnen från två av

kommunerna, Luleå och Kiruna, bjöds också in till ett pricktest med 10

vanliga luftburna allergen och 90% (n=1700) av de inbjudna deltog. År 2010,

när barnen var 11-12 år, gjordes en studieuppföljning med samma metoder

och med ytterligare tillägg av en intervju med barnet och bestämning av

body mass index (BMI). Studiedeltagandet i enkäter och pricktest var lika

högt vid uppföljningen som vid studiestart.

vi

Vid studieuppföljningen bjöds barn med rapporterad mjölköverkänslighet in

till en strukturerad intervju och barn som helt undvek mjölk, ägg, fisk eller

vete på grund av upplevd överkänslighet, bjöds in till klinisk undersökning

och provtagning. Baserat på testresultaten kategoriserades barnen i olika

fenotyper av födoämnesöverkänslighet utifrån förutbestämda kriterier. Barn

som bedömdes ha en aktuell födoämnesallergi bjöds därefter in till vidare

utredning med dubbelblind placebokontrollerad födoämnes-provokation. De

recept som användes vid de dubbelblinda provokationerna hade

dessförinnan validerats avseende detekterbara smak- och konsistens-

skillnader mellan aktiv- och placebosubstans i en separat kohort av friska

skolbarn (n=275).

Före och efter den dubbelblinda provokationen samlades blodprover in för

analys av cytokin mRNA-uttryck i mononukleära celler. Analyserna

inkluderade cytokiner kännetecknande för humoralt allergidrivande T-

hjälpar 2 (Th2) svar, cellulärt cytotoxiskt drivande Th1 svar samt

inflammatoriskt- och T-reglerande svar. Vidare insamlades avföringsprover

för analys av inflammatoriska biomarkörer före och efter genomgången

provokationsserie.

Resultat

Prevalensen av föräldrarapporterad födoämnesöverkänslighet ökade från

21% vid 7-8 år till nästan 26% vid 11-12 års ålder. Incidensen av rapporterad

födoämnesöverkänslighet var hög (15%), liksom remissionen (33%).

Riskfaktorer associerade med incidens och remission var olika för mjölk-

överkänslighet och överkänslighet mot andra födoämnen. Vi såg också en

bristande samstämmighet mellan föräldrarapporterad överkänslighet mot

mjölk, ägg, fisk, vete, soja och jordnöt och IgE-sensibilisering mot det

aktuella födoämnet.

Vid 11-12 års ålder var prevalensen av rapporterad allergi mot mjölk, ägg,

fisk eller vete 4.8%, medan prevalensen baserad på klinisk undersökning och

provtagning var 1.4%. Prevalenssiffran halverades ytterligare när kliniskt

bedömd födoämnesallergi validerades med dubbelblinda placebo-

kontrollerade födoämnesprovokationer. Majoriteten av barnen med

rapporterad allergi mot mjölk, ägg, fisk eller vete klassificerades som andra

fenotyper av födoämnesöverkänslighet, varav de vanligast förekommande

var möjlig laktosintolerans (40%) och utläkt födoämnesallergi (19%).

Även om förekomsten av rapporterad mjölköverkänslighet bland 11-12

åringarna var så hög som 14.5%, kategoriserades bara 3% av dessa som en

aktuell mjölkallergi. Mjölkallergi, aktuell eller utläkt, var associerat med

vii

andra atopirelaterade tillstånd och ett lägre BMI (OR 0.82, 95% CI 0.80-

0.98) jämfört med barn som inte undvek mjölkprodukter.

Före den dubbelblinda provokationsserien var mRNA-uttrycket av den Th2-

relaterade cytokinen IL-13 och den regulatoriska cytokinen IL-10 högre

bland barn med provokationspåvisad födoämnesallergi jämfört med barn

med en negativ födoämnesprovokation. Såväl före som efter

provokationsserien kunde högre nivåer av inflammationsmarkörerna

eosinofil-deriverat neurotoxin (EDN) och kalprotektin uppmätas i

avföringsprover från barn med positivt provokationsutfall jämfört med barn

med negativ födoämnesprovokation. Skillnaderna i uppmätta nivåer av

biomarkörer i faeces uppnådde dock inte statistisk signifikans.

Slutsats

Rapporterad födoämnesöverkänslighet var vanligt förekommande bland

skolbarn i Norrbotten och ökade från 7-8 år till 11-12 års ålder. Incidensen av

rapporterad födoämnesöverkänslighet var hög, liksom remissionen.

Prevalensen av rapporterad allergi mot mjölk, ägg, fisk eller vete var 8

gånger högre än den prevalens som kunde påvisas med dubbelblind

placebokontrollerad födoämnesprovokation. Allergi mot mjölk, ägg, fisk och

vete var en ovanlig orsak till att barn helt undvek dessa födoämnen på grund

av upplevd överkänslighet. Några av de biomarkörer som analyserades innan

provokationsserierna visade lovande resultat som möjliga, framtida

prognostiska markörer för symptomatisk, IgE-medierad födoämnesallergi.

Dessa resultat behöver dock valideras med ytterligare studier.

viii

Abbrevations

BAT Basophil Activation Test

BMI Body Mass Index

CI Confidence Interval

DBPCFC Double-Blind Placebo-Controlled Food Challenge

EAACI European Academy of Allergology and Clinical Immunology

EDN Eosinophil-derived neurotoxin

ELISA Enzyme-Linked Immunosorbent Assay

EPX Eosinophilic cation protein X

FHS Food Hypersensitivity

FPIES Food Protein Induced Enterocolitis Syndrome

HBD2 Human Beta-Defensin 2

IgA Immunoglobulin A

IgE Immunoglobulin E

IgG Immunoglobulin G

IL- Interleukin

IFN-γ Interferon gamma

ISAAC International Study of Asthma and Allergy in Childhood

kU/L Kilo Units per Liter

MHC Major Histocompatibility Complex molecules

mRNA Messenger Ribonucleic Acid

OAS Oral Allergy Syndrome

OLIN Obstructive Lung Disease In Northern Sweden studies

OR Odds Ratio

PBMC Peripheral mononuclear blood cells

PUFA Polyunsaturated Fatty Acid

qRT-PCR Quantitative Real Time Polymerase Chain Reaction

SPT Skin Prick Test

Tc T-cytotoxic cell

Th T-helper cell

TGF-ß Transforming Growth Factor beta

tTGA Tissue Transglutaminase A

Treg T-regulatory cell

USA United States of America

ix

Original Papers

I Winberg A, Nordström L, Strinnholm Å, Nylander A, Jonsäll

A, Rönmark E, West CE. New validated recipes for double-

blind placebo-controlled low-dose food challenges. Pediatr

Allergy Immunol 2013;24:282-7.

II Winberg A, Strinnholm Å, Hedman L, West C.E,

Perzanowski M.S, Rönmark E. High incidence and remission

of reported food hypersensitivity in Swedish children followed

from 8 to 12 years of age – a population based cohort study.

Clin Transl Allergy 2014;4:32.

III Winberg A, West CE, Strinnholm Å, Nordström L, Hedman

L, Rönmark E. Assessment of Allergy to Milk, Egg, Cod, and

Wheat in Swedish Schoolchildren: A Population Based Cohort

Study. PLoS One. 2015;10(7):e0131804.

IV Winberg A, West CE, Strinnholm Å, Nordström L, Hedman

L, Rönmark E. Milk allergy is a minor cause of milk avoidance

due to perceived hypersensitivity among schoolchildren in

Northern Sweden. Acta Paediatr. 2015 Epub ahead of print

2015/11/01.

V Winberg A, Nagaeva O, Nagaev I, Lundell C, Arencibia I,

Mincheva-Nilsson L, Rönmark E, West CE. Dynamics of

cytokine mRNA expression and fecal biomarkers in school-

children undergoing a double-blind placebo-controlled food

challenge series. (In manuscript)

All published papers were reproduced with permission from the original

publishers. Paper II and paper IV are freely available (open acess).

x

1

Introduction

Personal point of departure

When I started working in pediatric Allergy around the millennium shift, one

of the greatest challenges in our clinic was the increasing number of children

reporting perceived food hypersensitivity. There was a lack of knowledge

about different types of food hypersensitivity, their underlying mechanisms

and the diagnostic tools available were insufficient. Much has happened

since then. Emerging interest and research in the area of food

hypersensitivity has broadened our understanding about the different

presentations of food hypersensitivity and how to manage this in the clinical

setting. Nevertheless, there is still a lack of knowledge about the prevalence

of hypersensitivity to different foods at different ages as well as the

prevalence of different phenotypes underlying reported food hypersensitivity

in children and adolescents. There is also a shortage of easily available

prognostic markers determining presence of tolerance.

Also, despite improved diagnostic methods, the number of children with

perceived food hypersensitivity seems to be continuously increasing. There

are data suggesting that this, at least partially, could be explained by a true

increase and changing patterns of food allergy, though there is still a lack of

available data on the prevalence, incidence and remission of food

hypersensitivity in Sweden as well as globally (1). This awoke the idea in our

pediatric allergy team to investigate the epidemiology of food

hypersensitivity among school children in Northern Sweden and for this we

joined the Obstructive Lung disease in Northern Sweden (OLIN) studies.

The extensive experience of population based epidemiological research

within the OLIN group made the project in this thesis possible.

2

Figure 1. Definition of food hypersensitivity (FHS).

(Modified from Johansson et al. Allergy 2001 (2) and printed with permission from

Wiley, Copyright © Munksgaard 2001)

3

Background

Definition of food hypersensitivity (FHS)

There is inconsistency in the terminology of food hypersensitivity (FHS). In

this work we have chosen to use the terminology stated in the World Allergy

Organization (WAO) position paper published in 2001 (2), where FHS is

defined as an umbrella term for adverse reactions to food including reactions

of both immunological origin i.e. allergies and reactions driven by non-

immunological mechanisms i.e. intolerances (Fig 1).

Allergies – can be either Immunoglobulin E (IgE) mediated or non-IgE

mediated. While the immunological mechanisms in IgE-mediated allergies

are well described (3) less is known about the causative mechanisms in non-

IgE mediated allergies, though T-cells seem to play important roles (4-6).

Intolerances – can be either enzymatic, pharmacological or driven by other,

still unknown mechanisms. The most common intolerances are caused by a

shortage of or a decreased function of an enzyme necessary for the digestion

of a component in food (4,7).

Prevalence of food hypersensitivity

Reported FHS has emerged as a costly health problem in Western countries

(8) affecting 3-35% of the population (9-11). The prevalence in different

studies is difficult to compare due to heterogeneity in methodology, study

populations and definitions of FHS. Most studies focus on the prevalence of

IgE-mediated food allergies.

Prevalence of food allergy

Food allergy is thought to affect nearly 8 % of young children and 5 % of

adults in Western countries (12) and there are data suggesting that the

prevalence is increasing (13). In a recent European study the pooled life-time

prevalence and point prevalence of reported food allergy was 17.9% and 5.6%

respectively (14). The prevalence of self-reported food allergy is usually

higher compared to challenge-proven allergy (9,15). In another recent

European meta-analysis, the prevalence of self-reported allergy to the 8

common foods: milk, egg, wheat, soy, fish, shellfish, peanuts and tree nuts

was 6.0% while the prevalence of challenge-proven allergy to these foods was

4

0.6%. In this meta-analysis, allergies to most foods except peanut and soy

were more common in Northern Europe (16).

Independent of geographic setting, 90% of IgE-mediated allergies in

children are elicited by eight foods; cow’s milk, soy, hen’s egg, peanut, tree

nuts, wheat, fish and shellfish (17). Milk, egg and peanut are the main

triggers in infants and young children (16,18,19), while peanut, tree nuts, fish

and shellfish are the main food allergy triggers in adults (16). The majority of

cow’s milk and hen’s egg allergic children develops tolerance during infancy

and early school age (20-22).

Allergies to peanuts and tree nuts are more likely to become permanent and

less than 20% become tolerant (23,24). Allergy to fish and shellfish can

emerge in early childhood, but more commonly at school age (5,16). In

addition, raw fruits and vegetables are common triggers of the oral allergy

syndrome (OAS) affecting more than 50% of individuals with birch pollen

allergy, which is common in the Swedish population (25). Population based

studies from the north of Sweden show a prevalence of IgE-sensitization to

birch among young adults of 23% (26) and 13% in children aged 7-8 years

(27).

While allergies to hen’s egg and cod are predominantly IgE-mediated, IgE

and non-IgE mediated allergy to cow’s milk is equally common as is

tolerance development in both of these conditions (15,28,29). In a recent

European study, 69% of children diagnosed with cow’s milk allergy at one

year of age had become tolerant by the age of 2 years (15). In this study,

children IgE-sensitized to cow’s milk were less prone to develop tolerance

compared to children with no detectable milk-specific IgE. Less is known

about the prevalence and natural development of non-IgE-mediated allergy

to other foods, like soy and cereals, but studies indicate that most of the

affected children develop symptoms during the first years of life and the

majority develop tolerance during childhood (4,6,22).

Prevalence of lactose intolerance

The most common variant of food intolerance is the adult-type lactose

intolerance, which is caused by down regulation of the intestinal enzyme

lactase resulting in a decreased capacity to digest lactose (30). The onset of

lactose intolerance symptoms most commonly begins during adolescence

and rarely before school age (7). The down regulation of lactase is caused by

a specific genotype that is common in most parts of the world, except in

Northern Europe, where the prevalence of lactase persistence is high (31).

The prevalence of lactose intolerance has not been well studied, neither

5

globally nor in Sweden (7). One Swedish study showed a higher-than-

expected prevalence of lactase down-regulation among Swedish school

children, the prevalence increasing from 8% to 14 % during a 10-year period

(32). Others have shown that over-reporting of lactose intolerance is

frequent even in populations in which lactase down-regulation is common

(33), suggesting that symptoms of lactose intolerance often are caused by

underlying mechanisms other than lactase down-regulation (34).

Factors associated with food hypersensitivity

A number of intrinsic and environmental factors have been associated with

reported food hypersensitivity. These include female sex, number of siblings,

allergic heredity, allergic sensitization and presence of atopy related

conditions like asthma, rhinitis and eczema (35-37). The prevalence of

reported food hypersensitivity also seems associated with age and native

country of the study population (10,14,38). Since the risk factors associated

with food hypersensitivity are related to the background mechanisms of the

different types of food hypersensitivity, risk factors will be discussed

separately for food allergy and lactose intolerance.

In conclusion:

Food hypersensitivity (FHS) is common in children as well as in

adults, though the prevalence varies. This is likely due to the use

of different study methods, age and origin of the study

population and the definition of FHS. Most studies focus on IgE-

mediated allergies, which affect around 8% of young children

and 5% of adults in Western countries and the prevalence of

food allergy seems to be increasing. The prevalence of reported

food allergy exceeds the prevalence of allergy established by an

oral challenge.

6

Risk factors of food allergy

Food allergy is a polygenic, complex disease in which the risk of developing

symptoms seems to depend on a number of genetic and environmental

factors (39). The strong relationship between allergic heredity and

development of food allergy is a compelling evidence of the contribution of

genetic factors in this disease (39,40). Nevertheless, the specific genetic loci,

which modulates food allergy remains to be identified (41) as does the roles

of gene-environment interaction, gene-gene interaction and epigenetics (39).

Most studies investigating risk factors of food allergy are focusing on IgE-

mediated allergies and early life environmental exposures, given their

possibility to modulate developing immune responses and tolerance (42)

and being plausibly modifiable. Since the introduction and extension of the

hygiene hypothesis (43), these possible early modulating factors include

exposures affecting gut microbial composition and diversity: such as family

size and place of upbringing (44,45), early infections (46,47), delivery mode

(48), use of antibiotics, breast feeding and pre- and probiotics (49,50). In

addition, many studies have investigated dietary components as vitamin D

(51,52) and polyunsaturated fatty acids (PUFA) (53) as modifiers of allergy

development.

Still, there are many questions in the search of the origin of food allergies.

Recent data suggest that early introduction of allergenic foods in a child’s

diet can steer the immune system towards tolerance development (54,55)

while a route of IgE-sensitization through a disrupted skin barrier might do

the opposite (56). Childhood eczema is a risk factor associated with

development as well as persistence of food allergy (36,57,58). In addition,

food allergies often co-exist with other atopy related conditions like asthma,

rhinitis and allergic sensitization, which suggests that these conditions may

have common risk factors (12,59,60).

Low-level IgE-sensitization and IgE-sensitization to one or a few food

allergens may predict remission of a food allergy, while multi-sensitization

and high levels of specific IgE is associated with development and

persistence of food allergy symptoms (20,23,61). As discussed above, food

allergy in children is strongly associated with allergies and atopic diseases in

their parents (59). Except for genetic factors, other variables that could affect

this association are exposure patterns including allergen avoidance in

families at risk of having a food allergic child and report bias (39,63,64).

7

Some of the differences in the prevalence of food allergy between

populations have been related to race. Studies from the USA have shown a

higher risk of IgE-sensitization to common food allergens and higher levels

of IgE-sensitization in children with self-reported black race and/or African

ancestry compared to Hispanic or Caucasian children (65). Further, the

prevalence of food allergy may be affected by sex (58,63). Even though food

allergies and allergic sensitization are more common among boys than girls

(59) there is a shift during adolescence making food allergies more common

among women (37,66). The reason for this shift is still unknown, though

contribution of hormonal factors has been suggested (67).

Similar to IgE-mediated allergies, non-IgE mediated food allergies are

thought to emerge from a failed immunological tolerance development (68).

However, the risk factors specifically related to non-IgE mediated food

allergies are currently un-explored (64).

Risk factors of lactose intolerance

As in non-IgE mediated food allergies, there is a lack of studies investigating

the risk factors of lactose intolerance (7). Since lactose intolerance is caused

by a genotype coding for down-regulation of the intestinal enzyme lactase

and different mutations of the lactase down-regulating gene are more or less

prone to cause symptoms (69), the occurrence of this condition is strongly

associated with heredity. Notably, large differences in the prevalence of

lactose intolerance have been found between populations (30,70).

In most cases, symptoms of lactose malabsorbtion develop successively

during adolescence making lactose intolerance uncommon in younger

children (71). Even though the distribution of the lactase down-regulating

genotypes is similar among men and women, symptoms of lactose

intolerance are more frequently reported by women than men (70,72).

8

Immunological aspects of food allergy and tolerance development

Food allergy is a general term for adverse immunological reactions to food

and includes a number of different disease phenotypes (4,5). A food allergy

can be IgE-mediated, non-IgE mediated or a combination of the two. Even

though the underlying immunological mechanisms are diverse, and in the

case of non-IgE mediated food allergies largely unknown, they are all

thought to emanate from failed or abrogated tolerance, which involve an

aberrant regulatory T-lymphocyte response (73).

The development of tolerance normally occurs in the immune system of the

gut (74). Even though most of our knowledge about tolerance emanates from

animal studies, the reigning current paradigm is that the interaction between

the gut microbiota, the immune cells of the gut and exposure of food

allergens through an oral route are essential for a successful tolerance

development also in humans (49).

In conclusion:

Factors including sex, race, allergic heredity and atopy related

diseases e.g. eczema, asthma and rhinitis have been associated with

reported food hypersensitivity. Most studies on risk factors of food

hypersensitivity focus on IgE-mediated allergy, which is a polygenic,

complex disease dependent on a number of genetic and environmental

factors. Many uncertainties remain regarding risk factors of food

allergy, though early introduction of “allergenic” foods seem to steer

the immune system towards tolerance development, while

sensitization through a disrupted skin barrier might do the opposite.

Risk factors of non-IgE mediated allergies and lactose intolerance are

yet to be explored.

9

T-lymphocytes have been given their designation letter because they mature

in the thymus (75). There are three subpopulations of T-cells: T-helper cells

(Th), T-cytotoxic cells (Tc) and T-regulatory cells (Treg). Following

recognition of cell membrane bound antigens called major

histocompatibility complex molecules (MHC), Th-cells differentiate into

effector cells that enhance the activation of various other cells that

participates in the immune response e.g. B-cells, Tc-cells and macrophages.

Alternatively, some Th-cells differentiate into memory cells (76). The

immune environment of cytokine and chemokine signaling steers the

differentiation of Th-cells into different subsets (77). The first two subsets

discovered were Th1 and Th2. A predominate Th1-activation is associated

with autoimmunity while a dominant Th2-response relates to IgE-mediated

allergy (78).

In recent years a number of additional Th-subsets have been discovered,

including Th3 and Th17, contributing to the complexness of immune

response patterns (73,79). Th17 cells are involved in autoimmunity and has

also been suggested a role in neutrophilic asthma (80), whereas Th3 cells are

part of T-regulatory cells. T-regulatory cells can suppress the immune

response, either by cell to cell contact and/or by the production of down-

regulatory cytokines (81,82). T-regulatory cells include CD4+CD25+ T-

regulatory (Treg) cells, inducing suppressive effects via cell contact and type

1 regulatory (Tr1) cells and Th3 cells mediating their effects via cytokines e.g.

interleukin 10 (IL-10) and transforming growth factor beta (TGF-β) (83).

Treg cells are thymus-derived and express CD25 (the IL-2 receptor) and the

transcription factor FoxP3. The suppressor function of Treg cells appear to

depend in part on TGF-β, but less on IL-10 (84). Th3 cells are gut derived

and characterized by production of TGF-β (+ IL-10), exerting their effects by

mediating mucosal tolerance and antigen-specific IgA production (85). Tr1

cells are derived from the periphery and have the capacity to produce IL-10.

However, their origin is unclear and as of yet it is not known whether they

represent a distinctive developmental pathway or are derived from Th or

Treg cells (86).

IgE-mediated food allergy

The allergic process starts with exposure of an allergen to the immune

system, in food allergy usually via the intestinal mucosa (87,88). In the

surface of the mucosa, dendritic cells can sample luminal allergens and

present them to naïve T-cells in nearby lymph nodes (88). An environment

of Th2-related cytokines and chemokines facilitates clonal expansion and

differentiation of the naïve T-cells into Th2 subsets and memory cells.

10

Further, some of the encountered food allergens are transferred and

presented to B-cells, triggering them to mature into plasma cells, that

subsequently produce and secrete allergen specific IgE. This process is called

allergic sensitization (87,89) (Figure 2).

Re-exposure to the food allergen leads to allergen binding to circulating and

cell surface bound specific IgE-antibodies. Cross-linking of IgE on the

surfaces of mast cells and basophils can, under the right circumstances, lead

to release of inflammatory mediators e.g. histamine and tryptase, capable of

triggering an allergic response. Partly due to cross-binding IgG4-antibodies

and cross-reactivity between allergens, IgE-sensitization to food allergens is

far more common than food allergy (90, 91).

Figure 2. Schematic illustration of allergic sensitization resulting in

circulating and receptor-bound food antigen specific IgE-antibodies.

11

Non-IgE-mediated and mixed-immune food allergies

Less is known about the background mechanisms resulting in a non-IgE

mediated food allergy (6). The non-IgE mediated food allergies are a

heterogeneous group of diseases thought to be driven by cell–mediated

immunity that can be separated into different phenotypes following

improved methods of clinical investigation including increased use of

endoscopy. These phenotypes include: food protein induced reflux,

eosinophilic esophagitis, eosinophilic gastroenteritis and colitis and food

protein induced proctitis (6,89). Non-IgE mediated food allergies often

comprise symptoms from the gastrointestinal tract (92) and in some of these

conditions signs of inflammation and disrupted cytokine responses have

been detected in the local intestinal mucosa (93).

In recent years, it has been shown that in some food allergic diseases, IgE

and non-IgE mechanisms can co-occur. In these conditions, T-cell mediated

mechanisms are thought to cause a disrupted barrier in the skin or gastro-

intestinal mucosa and thereby facilitate allergic sensitization (94). Examples

of these mixed immune food allergies are eosinophilic gastrointestinal

diseases e.g. eosinophilic esophagitis and atopic dermatitis. Much is yet to be

discovered about these complex diseases (64).

Cytokine patterns in tolerance development and IgE-mediated

food allergy

Cytokines are mediators and communicators of the immune system, steering

the immunological responses to antigens in predisposed directions (95,96).

Food allergens are mainly presented to the immune system in the gut where

interactions between the gut microbiota and T-regulatory cells induce a

tolerogenic IL-10/TGF-ß response that appears essential for tolerance

development (49,97,98). IL-10 is a major regulatory cytokine of

inflammatory responses (99) that inhibits the production of IgE and

upregulates the secretion of IgG4 (100) and seems to play a critical role in

tolerance development and outgrowth of food allergy (101-103). In addition,

the cytokine TGF-ß has been shown to enhance tolerance development by

down-regulation of the immune system e.g. by inhibiting proliferation of T-

cells and activation of macrophages (104). IL-10 and TGF-ß gene

polymorphisms are associated with persistent milk allergy in children (105).

Similarly to children with IgE-associated asthma and eczema, children with

IgE-mediated food allergy have a cytokine response pattern with increased

expression of Th2-related cytokines e.g. IL-4, IL-5 and IL-13 and reduced

12

production of the Th1-related cytokine interferon gamma (IFN-γ) (106). This

dysregulated cytokine response is usually present already at a very young age

(107, 108). Th2-associated cytokine patterns have even been detected in cord

blood from children who eventually developed food allergy (109). Following

stimulation with relevant food allergens, elevated levels of pro-inflammatory

cytokines e.g. IL-6 have been detected in challenge-proven milk and egg

allergic children compared to children who had developed tolerance to these

foods, indicating an inflammatory response triggered by the food allergen in

children with persistent allergy (110).

Allergic inflammation in the intestinal mucosa

The majority of children with food allergy have at least one symptom from

the gastrointestinal tract, which comports with the common route of food

allergen exposure (92). In gastrointestinal allergies signs of allergic

inflammation can often be detected where the vulnerable intestinal mucosa

has been exposed to the culprit food allergen. By examining mucosal

biopsies, a skilled pathologist can distinguish eosinophilic inflammation

caused by food allergy from other inflammatory diseases of the gut e.g.

gastroesophageal reflux and celiac disease (6,92). The increased use of

endoscopy in food allergy diagnosis opens up the possibility of measuring

cytokines and other biomarkers at the site of the allergic inflammation

(6,111). In addition, decomposing products emanating from an allergic

inflammation in the intestinal mucosa may be detected in stool samples from

food allergic children (112).

In conclusion:

Food allergy is a general term for a heterogeneous group of disease

phenotypes resulting from immunological reactions to food with failed

or abrogated tolerance and aberrant regulatory T-cell response as

common nominators. Cytokines are mediators and communicators of

immunological responses. From very early life, dysregulated cytokine

patterns can be detected in children eventually developing IgE-

mediated food allergies.

13

Symptoms of food allergy

Food allergy can cause a broad palette of symptoms and many food allergic

children develop symptoms from more than one organ system (5). The most

common locations for food allergy symptoms are the skin and the

gastrointestinal tract including the oral mucosa (36). Symptoms of food

allergy can occur within minutes of allergen exposure or require repeated

exposures over time to be elicited. In addition, food allergy mostly causes

benign symptoms but it can also be a life-threatening disease (3,113).

To further complicate the management and diagnostics of food allergies, the

liability to develop symptoms may differ in the same individual due to co-

factors like infections, drugs, simultaneous exposure to other allergens and

physical activity (114,115). Even though symptoms caused by different types

of food allergies may be confusingly similar, there are some reaction patterns

and combination of symptoms that may help in the differential diagnostics

(116).

Symptoms of IgE-mediated food allergy

The first symptoms of an IgE-mediated allergy classically occur within 2

hours of exposure and often within minutes (5). As a result, IgE-mediated

allergies are also called immediate-type allergies. In some cases, the IgE-

mediated food reaction may be biphasic with a symptom relapse within 1-6

hours and in rare cases within days of allergen exposure (91).

Symptoms associated with IgE-mediated food allergies are rapidly emerging

itching of the mouth, vomiting, urticaria, angioedema and symptoms from

the airways (5,94). IgE-mediated food allergies are also strongly associated

with acute life-threatening anaphylaxis, which comprises airway and/or

cardiovascular symptoms (117). However, even though food allergies are

common triggers of anaphylactic reactions in childhood, these reactions can

also be triggered by mechanisms other than IgE-sensitization (118).

Following IgE-sensitization to tree or grass pollens, birch pollen in

particular, the pollen-specific IgE-antibodies may bind to other proteins

similar to the original allergen occurring in numerous plants and vegetables

(119). The majority of these pollen cross-reactive IgE-sensitizations do not

cause clinical symptoms (120). Risk factors for symptom development are

clinically relevant pollen IgE-sensitizations and IgE-sensitization to multiple

pollens. This secondary type of IgE-mediated food allergy is called oral

allergy syndrome (OAS) or pollen-associated food allergy syndrome (94).

14

The clinical symptoms of this condition are usually limited to itching and

sometimes erythema of the oral mucosa shortly following allergen exposure.

Except for some rare conditions, symptoms of OAS are usually local and

benign. The oral symptoms may however be confused with oral symptoms as

a first sign of a primary IgE-mediated food allergy (5).

Symptoms of non-IgE mediated and mixed-immune food

allergies

The most common symptoms of a non-IgE mediated food allergy are skin

and/or gastrointestinal symptoms occurring hours and sometimes days after

allergen exposure (6,92). These symptoms are also common in mixed-

immune food allergies, though symptom onset may be quicker due to the IgE

component (94). The symptoms of non-IgE mediated and mixed immune

food allergies e.g. eczema, diarrhea, malabsorption, feeding difficulties and

failure to thrive are usually not as typically associated to food allergy as the

symptoms of an IgE-mediated allergy (5,94). Together with the late onset

symptoms, these non-specific symptom patterns often lead to a delayed

diagnosis (6,64).

During recent years, we have learned that food allergies affecting the gut can

be sorted into different phenotypes according to their location and/or

distinct symptom patterns (6). Examples of such gastrointestinal food allergy

phenotypes are: food-induced reflux, eosinophilic esophagitis, eosinophilic

enterocolitis or food induced proctitis, the latter strictly limited to the most

distal part of the colon resulting in blood-tinged stools in an otherwise

unaffected infant (6, 94).

As in IgE-mediated food allergies, the non-IgE and mixed immune food

allergies contain a broad variety of disease severity, ranging from mild

eczema to a potentially life-threatening disease (92). The food protein

induced enterocolitis syndrome (FPIES) is a non-IgE mediated food allergy

resulting in dramatic symptoms including profuse repeated vomiting, pallor

and apathy and in some cases hypovolemic shock. The symptoms of FPIES,

that usually occur 1-4 hours following allergen exposure, are often

misinterpreted as caused by other acute onset diseases like septicemia,

meningitis, severe gastroenteritis or an acute gastrointestinal event (121).

15

Establishing a food allergy diagnosis

Diagnosing food allergy is a challenge, in every meaning of the word

(64,116). There is a broad variety in symptoms, severity and background

mechanisms of food allergy. This, together with insufficient diagnostic tools,

means that investigation of a suspected food allergy requires knowledge of

the different disease phenotypes, potential differential diagnoses as well as

the possibilities and limitations of the tests available (122,123). To date, the

most reliable method to diagnose or rule out a suspected food allergy is to

perform an oral food challenge and the double-blind placebo-controlled food

challenge is considered gold standard (5,116,123).

Clinical history

Even though it cannot be used as a solitary tool for food allergy diagnostics, a

thorough clinical history can often narrow down the area of search to a few

differential diagnoses and be helpful in the choice of proper diagnostic

methods (5,116,123). A clinical history for food allergy diagnostics should

include: possible food triggers, type and severity of symptoms, time to onset

and duration of symptoms after allergen exposure, extrinsic co-factors, age

of onset, allergic heredity, other atopy related diseases, growth deviation and

eating habits of the child (5,116).

In conclusion:

There is a broad variety in symptoms and severity of food allergy,

ranging from local benign symptoms to life-threatening anaphylaxis.

Most food allergic children have symptoms from more than one organ

system, of which the skin and gastrointestinal tract are the most

common. IgE-mediated food allergies are usually characterized by

immediate onset reactions while non-IgE mediated food allergies

usually occur hours and sometimes days following allergen exposure.

16

Recently, a standardized diet history tool has been developed and published

in order to support the diagnosis of food allergy (124).

IgE-tests

The most common laboratory tests in allergy diagnostics are measurement of

allergen specific IgE in patient sera and skin prick tests (SPT), measuring the

reaction of allergen specific IgE clad mast cells in the subcutaneous tissue

when exposed to allergen through a disrupted skin barrier (90,125) (Figure

3). Both of these tests demonstrate IgE-sensitization and not allergy, which

also requires the presence of symptoms upon allergen exposure (5,126).

In population-based studies a relationship can be seen between the level of

IgE-sensitization and the presence of allergic symptoms following allergen

exposure (127). Due to the lack of reliable tests of individual tolerance and

diverse allergenic properties in commercially available allergens, there are

no valid cut-off levels of specific IgE that guarantee or exclude food allergy

(128,129). Nor can the level of specific IgE be used for determining the

disease severity (113). Since these tests measure IgE-sensitization, the results

must always be interpreted in light of the clinical history (5) and the tests are

consequently of no use in diagnosing a non-IgE mediated food allergy (6).

The first generation of IgE-assays was based on crude allergen extracts

derived from natural sources, including both allergenic and non-allergenic

compounds (90). Today, many allergenic molecules have been identified and

by the use of recombinant DNA technology allowing cloning and large-scale

production of recombinant allergens, component-resolved analyses of

molecular allergens are now available for a number of food allergens

(126,130). This new technique has improved the sensitivity and specificity of

the IgE-assay, among other things it helps in differentiating cross-reactions

from true IgE-sensitization (126). However, even though component

resolved diagnosis allows a more individualized IgE-testing, a thorough

clinical history is still essential for its interpretation (131).

17

Oral food challenges

In some cases of food allergy, such as severe immediate onset allergy to a

single food trigger, the allergy diagnosis can be based on the clinical history

in combination with a concordant IgE-test (132). In most cases of food

allergy, however, the clinical history and the results of skin prick tests or

serological analyses are not as clear-cut. In these cases further evaluation of

the suspected food allergy is needed and the recommended method is the use

of a diagnostic elimination diet followed by an oral challenge (5,116).

Figure 3. Schematic illustration of IgE-tests: 1. Measurement of free

circulating food allergen specific IgE-antibodies in patient sera and 2. Skin

prick test measuring the effects of mast cell degranulation in the subcutaneous

tissue following allergen exposure through a disrupted skin barrier.

18

Most commonly when performing a diagnostic food elimination, a single

suspected food allergen is eliminated from the child’s diet over a limited

period of time. The duration of the food elimination is determined by the

type and severity of symptoms, though an elimination period of 2-4 weeks is

usually sufficient (116,132). Even if the child’s symptoms resolve during the

elimination period, a food allergy diagnosis cannot be established until a

food challenge has been performed, resulting in recurrence of symptoms

(116,123,132).

In most cases of food allergy where the food-induced symptoms are benign,

the diagnostic food challenge can be performed at home. However, if the

parents are not properly instructed and motivated the diagnostic

elimination/provocation test is often not completed and the child remains on

an elimination diet (133). Therefore, the advice on food elimination should

always be evaluated (132). Children with severe food allergic reactions

should be followed by a pediatrician and food challenges, when appropriate,

should be performed in a hospital setting (5,116).

Open challenges

Open challenges are most commonly used in clinical practice. Without

blinding and the use of placebo the open challenge is easier to perform and

less time consuming. It is, however, limited by the chance of bias, which may

result in falsely positive outcomes (116,134). The risk of false positive results

is highest if the patient is anxious about the challenge and if the patient’s

previous symptoms have been late onset and/or subjective (132).

During a food challenge the food is introduced in gradually increasing doses,

usually aiming at a final dose close to a portion size of the challenge food. In

most cases a standardized challenge schedule is used, though the food

amounts and dose intervals might need to be adjusted according to the

severity and eliciting doses of previous reactions (5,123).

Blinded challenges

In a single-blinded challenge the serving of challenge food or placebo is

blinded to the patient but not to the personnel responsible for the testing.

This procedure might reduce the risk of bias on the part of the patient, but

can still affect the observer in the interpretation of the challenge outcome

(132,135).

19

Therefore, the double-blind placebo-controlled food challenge (DBPCFC) is

considered gold standard in diagnosing food allergy and it is the preferred

method for research protocols (5,116,135). Double-blind challenges reduce

the patient and observer bias to a minimum and is particularly useful in

diagnosing or ruling out adverse reactions to food presenting with delayed,

chronic or subjective symptoms (132,136).

However, since the DBPCFC is time and labor consuming it is rarely used in

daily clinical practice (116). Also, even though several elaborate guidelines

have been published during recent years the procedure of DBPCFC has not

yet been fully standardized and there is still a lack of validated DBPCFC

recipes (114,136).

Complementary diagnostic tests and future prognostic biomarkers

The discovery of new food allergy phenotypes brings about the need of new

diagnostic methods (64). Some old diagnostic methods are being tested for

new and more specific purposes and new diagnostic methods are being

developed following deeper understanding of the background mechanisms of

food allergy (126,138). Also, since DBPCFCs are time and labor consuming

and puts the patient at risk of an allergic reaction, there is an ongoing search

for easily available objective prognostic markers that would make double-

blind challenges superfluous (131).

In conclusion:

Even though a thorough clinical history and the easily available IgE-tests

might be helpful, they are usually not sufficient to diagnose or rule out a

food allergy. An oral challenge is usually necessary to establish a food

allergy diagnosis and the double-blind placebo-controlled food challenge

(DBPCFC) is considered gold standard. Since DBPCFCs are costly and

time consuming they are rarely used in daily clinical practice. Also, the

procedure of DBPCFC has not yet been fully standardized and there is still

a lack of validated DBPCFC-recipes.

20

Endoscopy

Since many years endoscopy has been used to diagnose inflammatory

diseases in the gastrointestinal tract. The possibility to take biopsies and

visualize the site of inflammation has contributed to the understanding of

different phenotypes of gastrointestinal allergies, their distribution and

immunological mechanisms (92,111). For some types of food allergy i.e.

eosinophilic esophagitis, endoscopy is essential for the diagnosis (137). In

the future, endoscopy will probably be more commonly used for

measurement of diagnostic markers on site (6,64,122,138).

However, diagnosis by endoscopic methods are invasive and especially for

children, not easily available. Establishing a food allergy diagnosis through

biopsies from the intestinal mucosa also requires access to experienced

pathologists. So far, the fairly extensive resources necessary for endoscopic

investigations has limited its use to the larger University hospitals (6,92).

Inflammatory and regulatory makers in faeces

Using endoscopic methods, signs of allergic inflammation can often be

detected where the vulnerable intestinal mucosa has been exposed to the

culprit food allergen (111). A less invasive upcoming diagnostic method is to

determine decomposing products from an intestinal allergic inflammation by

analyzing stool samples (112). Previous studies have shown that eosinophils

are activated in food allergic intestinal inflammation (139) and that

eosinophil-derived neurotoxion (EDN) also known as eosinophilic protein X

(EPX), one of the major proteins released by eosinophils, might be used as a

marker of food allergic reactions (139,140).

It has also been shown that measurement of the cytosolic,

immunomodulatory protein calprotectin in faeces can be a helpful tool in the

follow-up of response to treatment and determination of reoccurrence of

cow’s milk allergy (141). In addition, the antibiotic peptide human beta-

defensin 2 (HBD2) and secretory IgA have been investigated in stool samples

as possible markers of food-induced inflammation and tolerance in the gut

mucosa (142,143).

Atopy patch tests

Patch tests are commonly used to identify delayed-type immune reactions to

allergens (138). In a patch test the allergen is placed in a small chamber,

which is then applied to the skin of the patient for 48 hours. The test reaction

is measured after 48 and 72 hours respectively.

21

In atopy patch testing, the same method is performed using food antigens,

usually fresh foods (144). Atopy patch testing was first used in studies of

patients with eczema, where it was found to be a useful tool in diagnosing

food allergy (145). The test has also been shown to be helpful in identifying

food triggers in children with gastrointestinal allergy i.e. FPIES and

eosinophilic esophagitis (146,147). Nevertheless, the roles of atopy patch

testing in food allergy diagnosis is conflicting and large-scale clinical studies

are lacking. Also, the use of fresh foods as allergen extracts in patch testing

has not been standardized and neither has the method of reading the test

results, making it dependent on the skills of the investigator (148).

Serological tests

As previously discussed, the available tests measuring specific IgE to

allergenic molecules in foods are rapidly increasing (126). Studies of patients

allergic to nuts and peanuts have showed that IgE-sensitization to storage

proteins more often correlate to the presence of allergy symptoms, while

cross-sensitizations to pollen allergens are usually asymptomatic or causes

local, benign symptoms (120). With a more widespread availability, the use

of molecular allergen analyses to determine the relevance of food

sensitization will probably increase, as will our understanding of the

relationships between food allergen IgE-sensitization and symptoms (149).

The occurrence of food-specific IgG is thought to indicate previous exposure

to the food and the presence of tolerance, though the role of specific IgG in

the mechanism of tolerance development is not yet fully understood (88).

During oral and sublingual desensitization therapy, clinical improvement

has been associated with a decrease in specific IgE and an increase in

allergen specific IgG4 (150). In addition, specific IgG4 has been shown to

inhibit peanut-induced basophil and mast cell activation in children that are

sensitized but clinically tolerant to major peanut allergens (151). However,

the level of increased specific IgG4 does not correlate well with the degree of

improvement or severity of the allergic disease which limits the use of

specific IgG4 as a sole marker of tolerance development (5).

In addition to being detectable in stool samples, the immunomodulatory

protein calprotectin can also be measured in sera. Calprotectin is a cursor of

neutrophil inflammation and elevated serum calprotectin levels have been

shown to be a useful marker of some rheumatic diseases (152). Recently it

has also been suggested as a possible biomarker of neonatal sepsis (153). To

our knowledge, no studies have so far been published investigating the levels

of serum calprotectin in children with suspected food allergy.

22

Basophil activation tests

In cellular allergy testing, an allergic response is induced in vitro (126). The

most common cellular allergy test is the basophil activation test (BAT), an

upcoming tool for the diagnosis of IgE-mediated food allergy (154,155).

During a BAT, allergen-induced changes in the expression of proteins on the

surface of basophils are detected using flow cytometry (126). Studies have

shown that BAT might be superior to IgE-tests and clinical history in

distinguishing a symptomatic peanut allergy from tolerance, especially in

ambiguous cases (156,157). The BAT has also been shown to complement

more established allergy tests in improving the diagnostic sensitivity (158).

In addition, measurement of basophilic sensitivity as activation at graded

dilutions of allergens (CD-sens) offers the possibility of monitoring

treatment with anti-IgE and oral tolerance induction in children with severe

food allergy (154,159). Up to now performing BATs requires a fairly rapid

and costly processing of the collected samples and expertise in performing

and interpreting the tests. Increased availability and standardization of

methods are necessary for a more wide-spread use of BATs (155).

In conclusion:

New diagnostic test and prognostic markers are emerging, hopefully

facilitating future food allergy diagnostics. Analyses of IgE-sensitization

to molecular allergens and cellular allergen tests appear promising in

improving the diagnosis of symptomatic IgE-mediated allergies while

an increased use of endoscopy brings about the possibility of sampling

diagnostic markers on the site of inflammation. In addition,

decomposing products emanating from local allergic inflammation in

the intestinal mucosa may be detectable in stool samples from food

allergic children.

23

Cytokines

Food allergic children have a cytokine response pattern with increased

expression of Th2-related cytokines e.g. IL-4, IL-5 and IL-13 and reduced

production of the Th1-related cytokine IFN-γ (106,110,160). These, and other

cytokines including the regulatory cytokine IL-10, have been examined as

possible prognostic markers using in vitro stimulation of peripheral blood

mononuclear cells (PBMC) with relevant food allergens and more seldom in

vivo with oral food challenges (103,161,162).

After stimulation of PBMCs with specific food allergens, the expression of

the Th2 cytokine pattern is enhanced and the expressed levels of the

regulatory cytokine IL-10 diminished in children with symptomatic IgE-

mediated allergy compared to tolerant children (107,161). Due to its active

role in tolerance development, IL-10 has been discussed as a possible

biomarker for symptomatic IgE-mediated food allergy (101) and so has the

Th2-related cytokine IL-13 (163,164).

In conclusion:

Following stimulation with relevant food allergens, decreased

expression of regulatory cytokines and increased levels of Th2-related

and pro-inflammatory cytokines have been detected in blood samples

from children with symptomatic food allergy compared to tolerant

children. Some cytokines, including the regulatory cytokine IL-10 and

the Th2-related cytokine IL-13 have been suggested as possible

prognostic markers of a symptomatic IgE-mediated food allergy.

24

25

Objectives

The main objective of the project in this thesis was to examine the clinical

epidemiology of food hypersensitivity in a population-based cohort of

schoolchildren in Northern Sweden followed from age 7-8 to 11-12 years of

age.

The specific aims were:

- To validate new recipes for double-blind placebo-controlled

challenges with cow’s milk, hen’s egg, soy, cod and wheat (Paper I).

- To investigate the incidence and remission of parentally reported

food hypersensitivity in a population based cohort followed from 8

to 12 years of age (Paper II).

- To assess the prevalence of allergy to cow’s milk, hen’s egg, cod and

wheat among 11-12-year old Swedish children using reported data,

clinical investigations and double-blind placebo-controlled food

challenges (Paper III).

- To describe the food hypersensitivity phenotypes among Swedish 11-

12-year old children reporting hypersensitivity to cow’s milk and to

analyze how the different phenotypes correlated to body mass index

(BMI), living conditions, allergic sensitization, allergic heredity and

physician diagnosed asthma, rhinitis and eczema (Paper IV).

- To investigate the spontaneous cytokine mRNA expression in

peripheral mononuclear cells from children with possible food

allergy and healthy controls before and after a double-blind placebo-

controlled food challenge series. Also, to investigate inflammatory

markers in stool samples from children with possible food allergy

before and after the completed food challenge and in relation to the

challenge outcome (Paper V).

26

27

Materials and methods

The first part of this study emanated from a clinical need of improving the

methods for diagnosing or ruling out food allergy in our pediatric patients.

At the time of the study, there were very few validated recipes for DBPCFC.

To allow a wider spread of our developed recipes for double-blind challenges

with the foods; cow’s milk, hen’s egg, soy, cod and wheat and to be able to

use them in research protocols, we decided to perform a study with the aim

of validating these recipes.

1. Method: Paper I

Study population

Whole school classes in the town of Umeå were invited to participate in our

test panels. Apart from validating our DBPCFC recipes we also wanted to

investigate whether the taste and texture of the challenge substances were

acceptable for children of different ages. We therefore invited children from

two different age groups; classes 2 and 3 (8-10 years of age) and class 8 (14-

15 years of age). Children were excluded from the study if they had a

hypersensitivity to the food that was tested.

We aimed to invite around 50 children in each age group (165). During the

study, the recipes for milk and wheat had to be improved due to detectable

differences between the active and placebo substances. New school classes

were thus invited as test panels for the sensorial testing of the improved milk

and wheat DBPCFC recipes. In all, 275 children (81 % of invited) participated

in the testing of at least one DBPCFC recipe.

Sensorial testing

Triangle tests were used for the sensorial testing (165,166). In a Triangle test

the panelists are served three samples in coded covered mugs/canisters with

a straw, whereof two samples are identical and one is odd (Figure 4). The

task of the panelists is to pick out the odd sample. The Triangle test is a

forced-choice procedure, meaning that the test panelist has to choose a

sample whether or not a perceivable difference is detected (165).

28

Statistical analysis

To determine whether there was a statistically significant detectable

difference between our active and placebo substances the numbers of correct

responses were calculated and compared to the total number of participants.

Further, p-values were calculated with a standard binomial distribution and

a p-value <0.05 was considered statistically significant. One-sided testing

was used, since the outcome of interest was whether the odd sample could be

detected with significant frequency. The minimum numbers of correct

responses to conclude a difference were calculated according to the formula:

[x= (n/3) +1.64√ (2n/9)] (165).

Figure 4. Arrangement of covered, coded cups with test substances and a

worksheet for the performance of a Triangle test.

29

Method: The OLIN studies

The four remaining papers in this thesis are the results from a collaboration

project between the Department of Clinical Sciences, Pediatrics, Umeå

University and the Obstructive Lung disease In Northern Sweden (OLIN)

studies. For 30 years, the OLIN-studies have pursued epidemiological,

population based research with focus on obstructive lung disease and

allergies in adults and children in Norrbotten, in the north of Sweden and

the studies has so far resulted in more than 250 original scientific papers and

14 doctoral theses.

Study population

This project is based on the second pediatric OLIN cohort recruited in 2006.

When recruited, the cohort consisted of all children in classes 1 and 2 (aged

7-8 years, medium age 8 years) in 3 municipalities in Northern Sweden;

Luleå and Piteå located by the coast of the Baltic Sea and Kiruna located in

the innermost parts of the Norrbotten County (Figure 5). The children

invited to this cohort study composed almost 60% of all children in this age

group in Norrbotten County in 2006. In 2010, there was a study follow-up,

using the same methods. This time, all children in classes 5 and 6 (aged 11-12

years, medium age 12 years) in the same three municipalities were invited to

participate in the study (27).

Fig. 5. Study area

30

Questionnaires

Procedure

The parental questionnaire included the International Study of Asthma and

Allergies in Childhood (ISAAC) core questions (167) with added questions

about symptoms, physician diagnosis, medication and possible determinants

of asthma, rhinitis and eczema. This questionnaire has been used in the

OLIN cohort studies since 1996 (27,168-170). In 2006, queries about allergy

/hypersensitivity to foods were added to the questionnaire (36). The parental

questionnaires were distributed by school personnel.

Participants

When the cohort was recruited in 2006, the parents of 2585 children, 96% of

invited, participated in the questionnaire. At the 2010 study follow-up, the

parents of 2612 children, 96% of invited, answered the questionnaire,

whereof whom 2378 (89% of the original cohort) also had participated at

study start (Figure 6).

Skin Prick Tests

Procedure

The SPT followed the European Academy of Allergology and Clinical

Immunology (EACCI) recommendations (171). The Solu-Prick panel

included: birch, timothy, mugwort, cat, dog, horse, two house dust mite

extracts (Dermatophagoides pteronyssius and farinae) and two mold extracts

(Claudosporium and Alternaria) (ALK, Denmark).

The potency of the extracts was 10 histamine equivalent prick (HEP) except

for the two molds, which were 1:20 w/v. Histamine 10 mg/ml and glycerol

were used as the positive and negative controls. A positive test was defined

as at least one wheal ≥3mm in diameter, recorded after 15 minutes.

Participants

In 2006, the 1895 children from the two municipalities Luleå and Kiruna

were invited to the SPT and 1700 (90%) participated. At the 2010 study

follow-up, 1657 children (86% of invited) participated in the SPT, whereof

1450 (77 % of the children invited in 2006) also had participated in the skin

prick testing at study start in 2006 (Figure 6).

31

Child interview

Procedure

At the 2010 follow-up, all children in the two municipalities Luleå and

Kiruna were also invited to a short interview, hence called Child interview,

including questions about perceived allergy/hypersensitivity and avoidance

of the basic foods; milk, egg, cod and wheat. The interviews were performed

at the children’s schools along with the skin prick tests and assessment of

BMI.

Participants

A total of 1824 11-12 year old children (98% of invited) participated in the

Child interview (Figure 6).

Serological tests

Procedure

Serum samples were analyzed using an fx5 ImmunoCAP food screening test

(ThermoFisher Diagnostics, Uppsala, Sweden) including cow’s milk, hen’s

egg, cod, soy, wheat, and peanut. If the screening test was positive, specific

IgE was analyzed separately for all of the included foods. An ImmunoCAP

inhalant allergen screening was also conducted (ThermoFisher Diagnostics,

Uppsala, Sweden) and included the following allergens: birch, timothy,

mugwort, cat, dog, horse, dust mites (Dermatophagoides pteronyssius and

farinae), and Claudosporium.

An IgE serum level > 0.35 kU/L was considered positive for all of the IgE

tests. Further, tissue transglutaminase A (tTGA) was measured in serum as a

marker of celiac disease using the EliATM CelikeyTM IgA test (ThermoFisher

Diagnostics, and a tTGA serum level < 7 U/L was considered negative.

Participants

At the 2010 follow-up all children in Kiruna and 25% randomly selected

children from Luleå were invited to donate blood samples for analyses of

specific IgE-tests and tTGA. Of 997 invited, 695 children (71 %) participated

in the blood sampling.

32

Statistical analyses

The statistical analyses were performed using the Statistical Package for

Social Science Software 21 (SPSS Inc, Chicago, IL, USA). In Papers II-V, the

Chi square test, and Fisher’s exact test when appropriate, or bivariate logistic

regression was used for bivariate comparison of categorical variables.

Multivariate regression models were used in Paper II and IV when adjusting

for possible confounding factors. The multivariate regression models

included the variables that were significant or borderline significant in the

bivariate analyses. When comparing continuous variables between groups,

the Student’s T-test or ANOVA was used if the variable was normally

distributed and the Mann-Whitney test or the Kruskal-Wallis 1-way ANOVA

if not. In all papers, a 95% confidence interval (CI) was used and a p-value <

0.05 was considered statistically significant.

33

Definitions

Paper II:

▫Any FHS - was defined as questionnaire-reported symptoms to at least one

of the specific foods: cow’s milk, hen’s egg, fish, wheat, soy, kiwi, orange,

apple, raw carrots, banana, nuts, peanuts and almonds.

▫Non-milk FHS - was defined as questionnaire-reported symptoms to one or

more of the specific foods; hen’s egg, fish, wheat, soy, kiwi, orange, apple,

raw carrots, banana, nuts, peanuts and almonds, cow’s milk excluded.

▫Milk FHS - was defined as questionnaire-reported symptoms to cow’s milk

only.

Paper III:

▫Reported food allergy - was defined as reported allergy or hypersensitivity

to and complete avoidance of one or more of the foods; cow’s milk, hen’s egg,

fish, or wheat in the Child interview (children in Luleå and Kiruna) or in the

parental questionnaire (children in Piteå). Children were not included in this

definition if they had a questionnaire-reported physician diagnosed celiac

disease.

Paper III-V:

▫FHS phenotypes – were based on data collected from the structured

interviews and analyses of food specific IgE and tTGA. The food

hypersensitivity phenotypes were determined by the following preset

criteria:

▫IgE-mediated food/milk allergy phenotype – was defined as the fulfilled

mandatory criteria:

A positive food specific IgE-test – provided exposure to the culprit

food within the last two years

Symptoms triggered by less than 100 ml of milk or less than a

portion size of egg, cod or wheat products

…and at least two fulfilled secondary criteria:

Onset before five years of age – provided that the food had been

introduced in the child’s diet prior to that age

First symptom within 15 minutes of exposure

Symptoms from more than one organ system

Symptoms triggered by treasure amounts of the culprit food

34

Symptoms triggered by skin exposure

Symptoms triggered by airborne exposure

Anaphylaxis / exercise induced anaphylaxis

▫Non-IgE mediated food/milk allergy phenotype– was defined as the

fulfilled mandatory criteria:

A negative food specific IgE-test

Symptoms triggered by less than 100 ml of milk or less than a

portion size of egg, cod or wheat products

No celiac disease

…and at least two fulfilled secondary criteria:

Onset before five years of age – provided that the food had been

introduced in the child’s diet prior to that age

First symptom more than 1 hour after exposure

Symptoms from more than one organ system

Symptoms triggered by treasure amounts of the culprit food

▫Suspected food allergy – was defined as all but one fulfilled criteria for the

IgE-mediated or non-IgE mediated allergy phenotypes.

▫Outgrown food/milk allergy phenotype – was defined as a convincing

history of an IgE-mediated or non-IgE mediated allergy, and continued

avoidance of the culprit food despite tolerance of at least 100 ml of milk or a

portion size of egg, cod or wheat products without symptoms.

▫Probable lactose intolerance phenotype – was defined as fulfilled

mandatory criteria:

Symptom onset after five years of age

Symptoms limited to flatulence, stomach-ache and/or diarrhea

Symptoms triggered by 100 ml of milk or more

Symptom-free on a lactose free/reduced diet

No milk allergy

No Celiac disease

▫Suspected lactose intolerance – was defined as all but one fulfilled

mandatory criteria for the probable lactose intolerance phenotype.

▫Celiac disease phenotype - was defined as a reported physician diagnosis of

Celiac disease and/or a positive tTGA-test.

35

▫Non-definable phenotype – was defined as avoidance of milk, egg, cod or

wheat but did not fulfil the criteria for any of the listed phenotypes or

declined donation of food specific IgE and/or tTGA, which were essential for

the FHS phenotyping.

▫Discontinued elimination diet – was defined as reported food allergy at the

Child interview (children in Luleå and Kiruna) or in the parental

questionnaire (children in Piteå) and cancellation of the food avoidance prior

to the structured interview.

Paper II-V:

▫Allergic heredity – was defined as questionnaire-reported parental asthma,

rhinitis and/or eczema.

▫FHS heredity – was defined as questionnaire-reported parental food

hypersensitivity

▫Allergic sensitization – was defined as at least one positive SPT or positive

IgE screen test.

36

Figure 6. Overview of the study participation: total number of children

(and percentage of invited) in the different parts of investigation of the

pediatric OLIN-cohort that was recruited in 2006 and followed-up in 2010.

37

2. Method: Paper II

Study population

The main objective of Paper II was to investigate the incidence and

remission of parentally reported food hypersensitivity between 7-8 and 11-12

years of age. For this longitudinal study we included the 2378 children (89%

of invited) from the OLIN cohort who participated both at study start and at

the 2010 follow-up (Figure 6).

Questionnaires and Skin prick tests

Questionnaire data regarding parentally reported hypersensitivity to foods at

age 7-8 years and 11-12 years were collected. Variables from the parental

questionnaire and SPT results analyzed as possible risk factors for incidence

and remission of reported hypersensitivity to foods were collected from the

2006 investigation when the children were 7-8 years of age.

Serological tests

Among the serological analyses collected from the random population

sample, results from the fx5 ImmunoCAP food screen test (ThermoFisher

Diagnostics, Uppsala, Sweden) were available for 652 children (71% of

invited) of those who had participated in the questionnaire both at study

start and study follow-up. The presence of a positive IgE to any of the foods;

milk, egg, cod, wheat, soy and peanut was analyzed in relation to parentally

reported hypersensitivity to the corresponding food.

38

Figure 7: Study overview paper III

(Winberg el al. PlosOne 2015)

39

3. Method: Paper III

Study population

The main objective of Paper III was to assess the prevalence of allergy to

cow’s milk, hen’s egg, cod and wheat at 11-12 years of age. In this cross-

sectional study we included all of the 2612 children who participated in the

2010 questionnaire (Figure 6). A study overview is presented in Figure 7.

Questionnaires and Child interviews

The inclusion criteria for invitation to the clinical investigation was a

reported allergy to the basic foods; milk, egg, cod or wheat, defined as:

- Reported allergy/hypersensitivity to milk, egg, cod or wheat

- Complete avoidance of the culprit food

- No physician diagnosed celiac disease

The children from the two municipalities Luleå and Kiruna were included

according to their answer to the Child interview question:

“To what extent do you avoid the following foods; cow’s milk, hen’s egg,

fish, or wheat, due to allergy/ hypersensitivity?” The possible answer

alternatives for each food were; not at all, partially or completely.

Since the children in the municipality of Piteå were not invited to the Child

interviews and SPTs, these children were included according to the answer to

the following question in the parental questionnaire:

“To what extent does your child avoid the following foods; cow’s milk, hen’s

egg, fish, or wheat, due to allergy/ hypersensitivity?” The possible answer

alternatives for each food were; not at all, partially or completely.

Data on celiac disease was collected from the parental questionnaire query:

“Has your child been diagnosed by a physician as having celiac disease?”

40

Clinical examinations

The children with reported allergy were invited to a clinical examination,

including:

- A structured interview (Supplemental file)

- A physical examination

- Blood samples for analysis of tTGA , food specific IgE, an IgE food

screen test and an IgE inhalant screen test

The clinical examination included measurement of length (cm) and weight

(kg) and a clinical investigation including inspection of ears, mouth and skin,

auskultaion of heart and lungs and palpation of abdomen, thyroid and

nodular glands. All clinical examinations were performed by the same

pediatric allergist (AW).

Food hypersensitivity phenotypes

According to the results from the clinical examinations, children were sorted

into the following food hypersensitivity phenotypes according to preset

criteria (see Definitions):

- Current food allergy

o IgE-mediated allergy

o Non-IgE-mediated allergy

o Suspected allergy

- Outgrown food allergy

- Lactose intolerance

- Celiac disease

- Non-definable cases/ Non-avoidance diet

Double-blind placebo-controlled food challenges

The DBPCFC was performed as a 3-session challenge series, the sessions set

one week apart (Figure 9). At two of the challenge sessions the test substance

contained the challenge food and at one session the test substance was free

of the challenge food (placebo).

The serving order was drawn by lot and unknown to both the patients and

the personal undertaking the challenges (AW and ÅS). The test substances

were prepared immediately prior to the challenge session and the same

dietician (LN) was responsible for the preparation during all the challenge

series. The DBPCFC recipes used were those previously validated in Paper I.

41

Due to safety reasons, the DBPCFC series were performed in a hospital

setting. The test substance was served in increasing doses 30 minutes apart

and the child remained for observation for 2 hours after the completed

challenge. The test session was aborted at the dose step where the child

developed clear, objective symptoms or when all dose steps had been

completed. If the symptoms were severe, the whole challenge series was

terminated. Following each challenge session, the child performed a 3-day

symptom diary.

The DBPCFC code was broken at a follow-up visit one week after the

completed challenge series. The same pediatric allergist (AW) and study

nurse (ÅS) were responsible at all challenge-sessions and follow-up visits. A

DBPCFC series was considered positive if the child during or following the

challenge sessions with active substance developed symptoms that were

objective and/or repeatable.

42

Figure 8. Study overview paper IV

43

4. Method: Paper IV

Study population

The low prevalence of challenge-proven allergy to basic foods initiated

further examination of the children reporting partial milk avoidance, as we

had only included children reporting complete milk avoidance in Paper III.

For this cross-sectional study we included all 1824 children (98% of invited)

from the 2010 follow-up who had participated in the Child interview (Figure

6). The main aim of this study was to investigate the milk hypersensitivity

phenotypes among children reporting milk avoidance due to perceived

hypersensitivity and whether the phenotypes were different between

children reporting complete or partial milk avoidance.

Child interviews, structured interviews and serological tests

The inclusion criterion was a report of milk avoidance, partially or

completely, due to perceived milk hypersensitivity. The inclusion was based

on the following question from the Child interview:

“To what extent do you avoid the following foods; cow’s milk, hen’s egg,

fish, or wheat, due to allergy/ hypersensitivity?” The possible answer

alternatives for each food were; not at all, partially or completely.

Children with physician-diagnosed celiac disease were not invited to the

structured interview. Data on celiac disease was collected from the parental

questionnaire query:

“Has your child been diagnosed by a physician as having celiac disease?”

Children participating in the child interview were also invited to assessment

of BMI and 1808 children (97% of invited) participated in measurement of

both length and weight.

Data regarding living conditions, allergic sensitization, allergic heredity and

physician-diagnosed asthma, rhinitis and eczema were collected from the

2010 parental questionnaire.

44

The structured interviews were performed 6-12 months after the Child

interview.

- Children reporting complete milk avoidance were invited to the

Clinical examination described in Paper III, which included a

structured face-to-face interview.

- Children reporting partial milk avoidance were invited to the same

structured interview but over telephone.

Children, who at the time of the telephone interview still avoided milk, were

invited to donate blood for analyses of tTGA, milk specific IgE, an IgE food

screen test and an IgE inhalant screen test (ThermoFisher Diagnostics,

Uppsala, Sweden). Results from analyses from the same tests were already

available for the children reporting complete milk avoidance who

participated in the Paper III study.

Food hypersensitivity phenotypes

According to the results from the structured interviews and blood sample

analyses, the children were sorted into the following milk hypersensitivity

phenotypes according to preset criteria (see Definitions):

- Current milk allergy

o IgE-mediated allergy

o Non-IgE-mediated allergy

- Outgrown milk allergy

- Probable lactose intolerance

- Celiac disease

- Non-definable cases

- Discontinued milk avoidance

The prevalence of the different milk hypersensitivity phenotypes was

compared between children reporting complete and partial milk avoidance.

In addition, the association between the different milk hypersensitivity

phenotypes and BMI, living conditions, allergic sensitization, allergic

heredity and physician diagnosed asthma, rhinitis and eczema was analyzed.

45

5. Method: Paper V

Since DBPCFC are time and labor consuming and put the child at risk of an

allergic reaction, there is ongoing search of objective, prognostic markers

that could predict the food challenge otcome. We examined the spontaneous

cytokine mRNA expression of PMBCs, quantified for a panel of 15 cytokines

discriminating between a humoral Th2-, cytotoxic Th1-, regulatory Th3/Tr1-

and inflammatory responses in children with suspected food allergy and

helathy controls before and one week after a completed food challenge-

series. In addition, inflammatory and regulatory biomarkers in faeces were

analyzed in children with suspected food allergy. The main objective was to

investigate whether the included biomarkers had the potential of

determining the challenge outcome.

Study population

We included all 18 children with suspected food allergy who participated in

the DBPCFC series in Paper III. Complete analyzable blood samples were

available from 17 of these children.

In addition, 7 children without any allergies randomly selected from the

OLIN cohort participants living in Luleå, were invited to participate in a

DBPCFC series with egg. All of the invited controls participated in the

challenge series and complete blood samples were available from 6 of these

children.

Serum sampling and serological tests

Blood samples for the analyses of expressed cytokine mRNA patterns were

collected immediately before the first challenge session (baseline samples)

and one week after the last session (post challenge samples) (Figure 9). Sera

for analysis of food specific IgE and IgG4 (ThermoFisher Diagnostics,

Uppsala, Sweden) were collected from all children at baseline and from

children with suspected food allergy post challenge.

Cytokine expression in PBMCs

Peripheral blood mononuclear cells (PBMC) were isolated using Ficoll-paque

gradient centrifugation. Then, using quantitative real time polymerase chain

reaction (qRT-PCR), a cytokine gene expression analysis was made.

46

The messenger ribonucleic acid (mRNA) expression of the 12 included

cytokines were analyzed as markers of T-cell response; IFN-γ as a marker of

Th1-response, IL-4, IL-5 and IL-13 as markers of Th2-response, TGF-β and

IL-10 as markers of T-regulatory response, IL17-A as a marker of Th-17

response, IL-2 and IL-15 as markers of T-cell proliferation and TNF α, IL-1β,

IL-6 and IL-8 as markers of pro-inflammatory response. The expression of

the different cytokines at baseline and post challenge was then analyzed

according to DBPCFC outcomes and in relation to the control group.

Figure 9. Study overview of paper 5 and analyzed cytokine mRNA

expression in peripheral blood mononuclear cells including hallmark

cytokines for the humoral allergy-promoting T helper (Th) 2 response,

cellular cytotoxicity-promoting Th1 response, inflammatory-, and T

regulatory responses.

47

Biomarkers in stool samples

The 18 children with suspected food allergy were also invited to donate stool

samples at baseline and post challenge. From the complete stool samples

available from 12 of these children, five fecal markers where quantified using

enzyme-linked immunosorbent assay (ELISA) methods (Immundiagnostik,

Bensheim, Germany): PhiCal®Calprotectin, EDN (Eosinophil-derived

neurotoxin), sIgA (Secretory IgA), HBD-2 (human β-Defensin-2) and α1-

Antitrypsin. The levels of the different fecal markers at baseline and post

challenge were then analyzed according to DBPCFC outcomes.

48

49

Results

Validation of new DBPCFC recipes (Paper I)

In this study we aimed to validate new recipes for low-dose double-blind

placebo-controlled food challenges in school children, by investigating if

there were any sensory differences between the active materials containing

cow’s milk, hen’s egg, soy, wheat or cod, and the placebo materials. The

challenge materials contained the same amino acid based product Elemental

028 extra liquid orange/pineapple flavour (Nutricia, Liverpool, England),

with or without added food allergens. Further, to match the sensorial

properties of the challenge foods, supplementary ingrediants had been added

to the different test materials. The test panels consisted of 275 school

children, aged 8-10 and 14-15 years respectively, from five Swedish schools.

Each participant tested at least one recipe.

Standardized blinded triangle tests were performed to investigate if any

sensory differences could be detected between the active and placebo

materials. In our final recipes, no significant differences could be detected

between the active and placebo materials for any challenge food (p>0.05).

These results remained after stratification for age and gender. The taste of

challenge materials was acceptable to the children in both age groups and no

unfavourable side effects related to test materials were observed.

Main results:

New recipes for low-dose double-blind placebo-controlled food

challenges for milk, egg, cod, wheat and soy were validated regarding

detectable differences between the active and placebo substances. The

taste and texture of the test substances were accepted by the children in

the test panels, i.e. at 8-10 and 14-15 years of age.

50

Incidence and remission of food hypersensitivity (Paper II)

Prevalence, incidence and remission

In this longitudinal study, we investigated parentally reported FHS in a

population based cohort of schoolchildren, followed from 8 to 12 years of

age. The prevalence of reported hypersensitivity to any food increased from

21% at 8 years to 26% at 12 years of age (p< 0.001). During the same 4-year-

period, the prevalence of reported hypersensitivity to the basic foods: milk,

egg, cod and wheat increased from 10 % to 15 % (p<0.001) and the

prevalence of reported milk hypersensitivity from 9 % to 13 % (p<0.001)

(Figure 10). The cumulative incidence of reported FHS was high (15%), as

was remission (33%). This pattern was recognizable for hypersensitivity to

many of the 13 common foods investigated, but it was particularly evident

for hypersensitivity to cow’s milk.

Figure 10. Comparison of prevalence of food hypersensitivity, asthma, rhinitis

and eczema respectively, at the age of 7-8 and 11-12 years of age.

51

Factors associated with incidence and remission

Female sex, allergic heredity, current rhinitis and allergic sensitization were

associated with the incidence of any FHS. Allergic sensitization was

negatively associated with remission. While the factors associated with the

incidence and remission of non-milk FHS were similar to the factors

associated with any FHS, the risk factor patterns for milk hypersensitivity

were different. Female sex and allergic heredity were associated with the

incidence of milk FHS and living in Kiruna was negatively associated with its

remission, though no association was found to other atopy related conditions

(Figure 11).

Figure 11. Association of listed variables at 8 years of age to the incidence of

non-milk FHS and milk FHS presented as odds ratios (OR) and 95% confidence

intervals (CI).

52

Reported FHS and IgE- sensitization

Among the 652 children (71% of invited) who participated in the serological

tests, parentally reported hypersensitivity to the foods; milk, egg, cod, wheat,

soy and peanut were analyzed in relation to the presence of a positive IgE-

test to the culprit food. Generally, the agreement between reported food

hypersensitivity and IgE-sensitization to the implicated food was poor. This

was particularly true for the foods: milk, wheat and soy. However, even for

peanut, only one third of the peanut sensitized children reported symptoms

and only one third of the children with reported hypersensitivity to peanut

had a positive IgE-test.

Main results:

There was a high incidence as well as a high remission of reported food

hypersensitivity from age 8 to 12 years of age. Risk factors associated

with incidence and remission were different for milk hypersensitivity

and hypersensitivity to foods other than milk. The agreement between

reported symptoms to milk, egg, cod, wheat, soy and peanut and IgE-

sensitization to the culprit food was poor.

53

Assessment of allergy to milk, egg cod and wheat (Paper III)

In this cross-sectional study we investigated the prevalence of allergy to the

basic foods; milk, egg, cod and wheat among 11 to 12-year-old children in a

population based cohort using;

1. Reported data

2. Clinical investigations (structured interviews, physical examinations

and analyses of serum specific IgE and a celiac screening test)

3. A double-blind placebo-controlled food challenge series.

The prevalence of reported hypersensitivity to cow’s milk, hen’s egg, cod,

and/or wheat among the 11 to 12 year olds was 14.7% and the prevalence of

reported allergy to these foods according to our definition was 4.8% (Figure

12).

The 125 children with reported allergy to milk, egg, cod and wheat were

invited to the clinical examination and 94 (75%) participated. Based on the

results, children were categorized into the following FHS phenotypes:

current allergy (29%), outgrown allergy (19%), lactose intolerance (40%)

and unclear (12%). Among children categorized as having a current allergy,

positive food (63%) or inhalant (74%) allergen screen test were common, as

were reported physician diagnoses of asthma (52%), rhinitis (44%), and

eczema (67%). Assuming a similar distribution among the non-participants,

the prevalence of allergy to milk, egg, cod and wheat according to the clinical

examination was 1.4% (Figure 12).

The 27 children categorized as current food allergy were then considered for

further evaluation with a DBPCFC series. Two of these children fulfilled

preset exclusion criteria and were therefore not included. Of 25 invited, 18

children (72%) participated in a total of 20 challenge series with milk, egg or

cod. Thus, 2 children participated in 2 challenge series with different basic

foods.

Out of the 20 DBPCFCs, 9 were considered positive. All but one child with a

positive challenge outcome were IgE-sensitized to the culprit food. Four of

the nine children with positive challenge outcomes reacted with an

anaphylaxis. Only one of these children had previously been prescribed an

epinephrine auto-injector and none of these children were on daily

treatment for their concomitant asthma. Assuming a similar distribution of

54

positive DBPCFC outcomes among the non-participants, the food allergy

prevalence following DBPCFC was 0.6% (Figure 12).

Main results:

The prevalence of reported allergy to cow’s milk, hen’s egg, cod, or wheat

was 4.8%, whereas the prevalence according to preset diagnostic criteria

was 1.4%. This figure was further halved when possible food allergy was

evaluated with double-blind placebo-controlled food challenges. The

majority of children with reported allergy to basic foods were categorized

as other food hypersensitivity phenotypes, the most common being

probable lactose intolerance and outgrown food allergy.

Figure 12: Prevalence of reported food hypersensitivity and allergy to the basic

foods; milk, egg cod and wheat according to reported data, clinical investigation

and double-blind placebo-controlled food challenges.

55

Milk hypersensitivity phenotypes (Paper IV)

In this cross-sectional study we investigated the milk hypersensitivity

phenotypes among 11-12-year old children with self-reported milk avoidance

due to perceived milk hypersensitivity. Of the whole study sample, n=1824,

the prevalence of self-reported milk hypersensitivity was 14.5% (n=265). Of

these, 193 children (73%) reported partial milk avoidance and 72 children

(27%) reported complete avoidance of milk. Among the 265 children with

reported milk hypersensitivity, 236 (89%) participated in a structured

interview.

Only 3 % of the 11 to 12-year-olds with self-reported milk hypersensitivity

were categorized as current milk allergy; of whom all had reported complete

milk avoidance. The most common milk hypersensitivity phenotype was

probable lactose intolerance (40%) followed by outgrown milk allergy

(23%).

Figure 13. Distribution of the different milk hypersensitivity phenotypes

among children reporting complete and partial milk avoidance

56

Children with outgrown milk allergy had a convincing history of milk allergy

but tolerated at least 100 ml of milk without any symptoms at the time of the

evaluation. Despite this, they had remained on an elimination diet. At the

time of the structured interview, 30 % of the children had discontinued their

milk elimination diet. All children with discontinued elimination diet had

previously reported partial milk avoidance. Except for the phenotypes

current milk allergy and discontinued milk avoidance, the distribution of

the different milk hypersensitivity phenotypes was similar between children

reporting complete and partial milk avoidance (Figure 13).

Figure 14. Association of the listed variables to milk allergy, current

or outgrown expressed as odds ratios (OR) with 95% confidence

intervals (CI).

57

Children with current or outgrown milk allergy had a lower BMI (OR 0.82,

95% CI 0.80-0.98) compared to children with no milk avoidance. Also,

among children with milk allergy the variables female sex, physician

diagnosed eczema and heredity for food hypersensitivity were more common

compared to children not avoiding milk (Figure 14). In contrast, children

with probable lactose intolerance were less often sensitized to aeroallergens

and food allergens compared to children with no reported milk

hypersensitivity.

Only 9% of the children currently avoiding milk reported that they in

previous health care contacts had ever been referred to a dietician for

nutritional advice and only 2% reported that their milk hypersensitivity

diagnosis had been established by an oral challenge.

Main results:

Even though the prevalence of reported milk hypersensitivity among

the 11-12 year olds was as high as 14.5%, only 3% of them were

categorized as current milk allergy. The most common milk

hypersensitivity phenotypes were probable lactose intolerance and

outgrown milk allergy. Current and outgrown milk allergy was

associated with other atopic disorders and lower BMI. Only 9 % of the

children currently avoiding milk reported that they had ever been

referred to a dietician for nutritional advice and only 2% had a milk

hypersensitivity diagnosis established by an oral challenge.

58

If only immunology was this simple…

59

Biomarkers in blood and stool samples (Paper V)

Here, we analyzed the spontaneous cytokine mRNA expression pattern of

PMBCs before and one week after a completed DBPCFC series with milk, egg

or cod in 12-13 year old children with suspected food allergy and in age-

matched healthy controls. Blood samples were available from 17 children

with suspected food allergy; 9 children with negative challenge outcomes and

8 with positive challenge outcomes. Of the children with positive challenge

outcomes, 3 had reacted with anaphylaxis during challenge. In addition, 7

children from the OLIN cohort without any allergies participated in a

DBPCFC series with egg and blood samples were available from 6 of these

controls.

Further, inflammatory and regulatory biomarkers in faeces were analyzed in

children with suspected food allergy before and one week after the DBPCFC

series. Stool samples were available for 12 children, 6 with positive and 6

with negative challenge outcomes.

Figure 15. Median mRNA expression of the cytokines IL-13 and IL-10 at

baseline and post challenge in children with positive and negative

challenge outcomes and in controls.

60

At baseline, children with suspected food allergy expressed higher mRNA

levels of the Th2-related cytokine IL-13 compared to controls. The baseline

mRNA level of IL-13 was higher in children with a positive challenge

outcome compared to children with a negative challenge outcome. The

baseline levels of the regulatory cytokine IL-10 were similar in the controls

and in children with a negative challenge outcome, though significantly

higher among children with a positive challenge otcome, i.e. challenge-

proven food allergy (Figure 15).

Post challenge, children with suspected food allergy expressed higher mRNA

levels of the pro-inflammatory cytokines IL-1β and IL-6 compared to

controls but there were no significant differences detected in the post

challenge cytokine mRNA profiles between children with positive and

negative food challenges.

Figure 16. Median values of fecal markers at baseline and post challenge

among children with a positive (n=6) and negative (n=6) challenge outcome.

61

Three of the children in our study reacted with anaphylaxis during DBPCFC.

Compared to healthy controls, children with anaphylaxis expressed 10-fold

higher mRNA levels of the Th2-related cytokines IL-4 and IL-13 at baseline

and 30-100 fold higher levels of the pro-inflammatory cytokines IL-1β and

IL-6 post challenge. Children with anaphylaxis also expressed a 10-fold

higher post challenge mRNA level of the regulatory cytokine IL-10 compared

to controls. As opposed to this finding, the post challenge mRNA level of the

regulatory cytokine TGF-β1 was decreased among children with anaphylaxis

compared to healthy controls. Due to the low number of children with

anaphylaxis, no statistical analyses were performed on this subgroup.

The levels of calprotectin and EDN in feces were higher among the children

with a positive food challenge compared to children with a negative

challenge outcome both at baseline and post challenge, though the

differences did not reach statistical significance (Figure 16). There were no

detectable differences between children with challenge-proven allergy and

children with a negative challenge outcome for the fecal biomarkers: sIgA, β-

Defensin-2 and α1-Antitrypsin.

Main results:

At baseline, children with suspected food allergy had elevated

expression levels of the Th2-associated cytokine IL-13 compared to

controls. IL-13 mRNA and mRNA for the regulatory cytokine IL-10 were

higher in challenge-proven food allergy compared to the cytokine levels

seen in negative challenges. Post challenge, IL-1β and IL-6 mRNA were

elevated in the allergic children compared to controls. Fecal calprotectin

and EDN levels were higher in challenge proven food allergy compared

to a negative challenge although not statistically significantly.

62

63

Discussion

Discussion of methodology

Sensory testing

There are two types of sensorial differentiation tests; Overall difference tests

that are designated to show whether subjects can detect any difference

between samples and Attribute difference tests where different samples are

ranked according to one or more attributes i.e. sweetness (166). Since our

aim was to investigate whether children in different age groups could detect

any sensorial difference between the active and placebo materials of our

recipes, we choose to use the Overall difference test called Triangle test.

Compared to the alternative Overall difference test called Duo-trio-test, the

Triangle test is statistically more efficient (165,166).

It has been suggested that for sensorial testing, professional adult panelists

should be used (166,172). However, since a secondary aim of our validation

study was to examine whether the taste and texture of our test samples were

acceptable to children of different ages, we chose to include children aged 7-

10 years and 14-15 years in our test panels. To increase the validity of the

tests, we used large test panels. In a triangle test, a test panel of at least 25-

30 people is required to obtain the requisite sensitivity of the test (165,172).

We therefore aimed to invite around 50 children in each age group. Further,

before starting the study tests the participating children were educated in the

triangle test technique using two different kinds of fruit juices.

Epidemiologic study design

A major strength of our project including Papers II-V, is the longitudinal and

population-based design and the high participation-rates, supporting the

representativeness of the population i.e. the internal validity. The internal

validity describes to what extent the study outcome defines the true situation

in the study population, while external validity describes whether the results

of the study are applicable to other populations or settings (173).

Since the investigated cohort represent almost 60% of the children in

Norrbotten County, we have reason to believe that our findings are

representative for schoolchildren in the North of Sweden.

64

Supported by the findings of others (16,32,35,38), we also think that our

results might be applicable to children in other parts of Sweden and perhaps

even to children with similar socio-economic conditions in other parts of the

Western world, even though the very high percentage of Caucasians in our

cohort could affect the prevalence of FHS (32,65).

The relatively large study cohort and the commonness of the investigated

conditions, support the reliability of our studies. The pediatric cohort was

originally recruited for investigating prevalence of asthma, rhinitis, eczema

and allergic sensitization and for this the power of the study was excellent.

The calculated power for investigating incidence and remission of these

conditions was also satisfactory.

In line with this, the power of our questionnaire studies investigating

prevalence, incidence and remission of reported food hypersensitivity was

good. Even though the participation in the structured interviews and clinical

examinations were somewhat lower, 89% and 75% respectively, the very high

participation (96-97% of invited) in the parental questionnaires allowed us

to perform non-responder analyses based on questionnaire data. In Paper III

and IV these analyses did not show any significant differences for variables

like: sex, living conditions, older siblings, parental smoking, allergic

heredity, allergic sensitization and physician diagnosed asthma, rhinitis and

eczema between participants and non-participants in the structured

interviews and clinical examinations. Thus the study participants were

representative of the entire cohort.

A possible limitation with the population-based design, was that despite a

large cohort, the high participation rate and a high prevalence of reported

hypersensitivity to the basic foods: milk, egg, cod and wheat in the entire

cohort, the final prevalence of current food allergy following clinical

examination was low. As a consequence, a small group of children were

finally invited to the DBPCFC series, affecting the power in the analyses of

the challenge outcomes in Paper III and more so in the statistical analyses of

biomarkers in Paper V. The findings of the latter study therefore needs to be

verified by larger studies, by either inviting a larger cohort or by inviting

study participants from a hospital material (173).

65

In Paper II we investigated the cumulative incidence of reported food

hypersensitivity i.e. the onset of new cases with time, and the remission i.e.

the number of study participating recovering with time (173), between the

ages 7-8 and 11-12 years. A longitudinal study design, where a cohort is

followed over time, is preferable when examining incidence or remission of a

condition. In Papers III and IV, we used cross-sectional studies to

investigate the prevalence of allergy to basic foods and other food

hypersensitivity phenotypes at 11-12 years of age. Prevalence is a measure of

the percentage of individuals affected by a condition or exposure at a certain

point of time. Combining longitudinal and cross-sectional data on FHS

rendered a broader picture and a better understanding of the investigated

condition and was a major strength of our study.

Reported data

A significant part of this work is based on reported data, making it

susceptible to information bias (175). Bias stands for a systematic error that

causes distortion of the study, which might affect the study outcome

(173,174). Information bias is caused by measurement errors and can arise

when some people are more or less inclined to report the presence of a

disease or an exposure, defined as reporting bias, or if certain people are

more or less prone to remember conditions and exposures of the past,

defined as recall bias (175). Information bias may also occur if different

questionnaires or study methods are used at different study occasions, when

using study methods that are highly dependent on the investigator or when

the same reported data are collected from different individuals, e.g. children

and either of their parents.

In our study we used a comprehensive questionnaire, which was almost

identical at study start and follow-up. The questionnaire has also been used

in previous OLIN-studies since 1996 (27,168-170). Further, the clinical

examinations in Paper III were all performed by the same pediatric allergist

(AW). We did not, however, have access to the children’s medical journals to

verify reported data on food hypersensitivity diagnoses and previous

diagnostic tests, wich could have added further information and reduced the

risk of report bias.

In Paper III, the inclusion criteria for the children in the two municipalities

Luleå and Kiruna and the municipality of Piteå differed. The children in

Luleå and Kiruna were included according to self-reported FHS at the Child

interview, while children in Piteå were included according to parentally

reported FHS.

66

Of the children in Luleå and Kiruna who participated in the parental

questionnaire, 98% also participated in the Child interview, which allowed

comparison of parentally reported FHS and FHS reported by the child. This

comparison showed a good agreement between parental and children

reports of allergy/hypersensitivity to milk (Table 1) as well as FHS to egg,

cod and wheat. The agreement was highest for complete avoidance of these

foods. This is in line with previous study findings from another OLIN cohort

including children of the same ages as in the current cohort, where a very

high agreement between parental and child reports of allergic diseases was

found (176).

Table 1. Comparison between milk avoidance, complete or partially, reported by the

child and parentally reported milk avoidance of the child.

Milk avoidance reported by child

Parentally reported

milk avoidance

Non-avoidance

(n=1559)

n (%)

Milk avoidance

(n=265)

n (%)

Non-avoidance 1507 (97 %) 63 (24%)

Milk avoidance 52 (3 %) 202 (76 %)

Analyses of associated risk factors

In the analyses of possible risk factors in Papers II-IV, we included available

variables, which in previous studies have been associated with increased or

decreased risk of food hypersensitivity or food allergy (39,56,57,65). Further,

variables with significant or borderline significant differences in the bivariate

analyses, were included in multiple logistic regression analyses to adjust for

possible confounding factors (173,177).

67

Confounding factors are sometimes regarded as sources of bias (173,174). A

confounding factor is an extraneous variable that is correlated with the

investigated variable of interest as well as with the outcome. If an existing

confounding factor is not regarded in the analyses, a false association

between the investigated variable and the outcome might appear. Despite the

use of multiple logistic regressions, it is often difficult to rule out all possible

confounding factors, since many confounding factors are still unknown

(173).

It has been suggested, though not established, that when analyzing many

possible risk factors, the p-value should be diminished in order to reduce the

risk of significant associations appearing by chance (173,177). However, the

tested variables in our study were not included by chance but selected

according to previous study findings (39,56,57,65), suggesting these

variables as possibly associated to our study outcomes.

IgE-tests

The current project was part of a larger longitudinal population-based study,

investigating prevalence and associated risk factors for atopy related

conditions i.e. asthma, rhinitis, eczema and allergic sensitization. To

determine the prevalence of allergic sensitization, the children in the two

municipalities Kiruna and Luleå were invited to a SPT with 10 common

airborne allergens. Further, all children in Kiruna and 25% randomly

selected from Luleå were invited to donate blood for serological analyses and

695 children (71% of invited) participated. To facilitate participation, the

SPTs were performed at the children’s schools.

For the benefit of our project, it would have been preferable to add common

food allergens, especially milk, egg, cod and wheat to the SPT panel and this

was also discussed prior to study start. However, even though SPT with food

allergens are considered to be a safe procedure in recent management

guidelines (178), anaphylactic reactions during SPT with food allergens have

been described. During my 15 years practice as a pediatric allergist I have

witnessed two cases of anaphylaxis caused by SPT with food allergens, one

triggered by hen’s egg and the other by cod. Since severe reactions due to

SPT with food allergens are rare and when early recognized, fairly easily

treated, we would not have considered this a problem if the study SPT had

been performed in a health care facility. However, in this study SPTs were

made on 1700 children and performed at the children’s schools, some

located very far from a Health Care Center. During these circumstances we

could not guarantee the tests to be completely safe for the study participants.

68

In a previous pediatric OLIN cohort study the correlation between a positive

SPT and a specific IgE > 0.35 KU/L to aeroallergens included in the SPT was

validated and found to be excellent (179). Therefore, instead of including

foods in our SPT panels, food-specific IgE-analyses were added to the

serological tests in the random population sample (n=695).

With the exception of Paper V, specific IgE was only analyzed to the crude

food allergens. Before study start, analyses of specific IgE to allergen

molecules were also discussed for Papers II-IV. We did not, however,

prioritize this since by that time, only a few IgE-tests for molecular allergens

in foods were commercially available and the knowledge about their

usefulness in food allergy diagnostics was limited (126). There is however a

need for population based studies investigating the prevalence of IgE-

sensitization to these newly discovered allergenic molecules and how they

relate to the presence of allergic symptoms, for food allergens as well as

inhalant allergens (180). There is also a need for clinical research

investigating whether patterns and levels of IgE-sensitization to different

molecular food allergens might be useful in predicting food challenge

outcomes (126).

Validation of reported data

Another major strength of this project is the validation of reported data on

allergy to milk, egg, cod and wheat, first with the more easily available

clinical examinations and serological tests and then further with DBPCFCs

(135,178). The FHS phenotyping following the clinical examination and

serological tests, was useful for identifying the children with current food

allergy who were invited for further evaluation with a DBPCFC series. The

categorization also gave us a good overview of the distribution of FHS

phenotypes underlying reported hypersensitivity to milk, egg, cod and wheat

in the examined cohort.

Previous studies have reported that the prevalence of perceived food allergy

exceeds that of challenge-proven allergy (9,15,16) and the use of DBPCFC

renders the lowest prevalence (9,135). However, even though considered

gold standard for diagnosing food allergy, DBPCFC are rarely used in daily

clinical practice and until recently, few population based studies have

validated reported food allergy with double-blind challenges (15,16,181).

In Paper III, we used a DBPCFC series to validate reported allergy to milk,

egg, cod and wheat. Since our resources were limited, we chose to focus on

these four basic foods due to the high reported prevalence of FHS to milk,

egg, cod and wheat in this cohort (36), the possible negative nutritional

69

effects of elimination of these foods (182-184) and the lack of knowledge

about the phenotypes underlying perceived FHS to milk, egg, cod, and wheat

in schoolchildren (4,12). The omitted validation of the other FHS phenotypes

could be considered a study limitation, as could the exclusion of allergies to

other foods. However, validating all FHS phenotypes or including

hypersensitivity to more foods was not possible in our project. Instead, this

will be an eligible task for future population based studies on food

hypersensitivity.

The DBPCFC methodology

When we planned this study, very few validated recipes for DBPCFC in

children were available, whereof the wheat recipe required baking and no

validated recipe at all was available for cod (185). Therefore we developed

our own DBPCFC recipes, which were successfully validated regarding

detectable sensorial differences between the active and placebo substances in

a separate cohort of healthy schoolchildren (n=275) in Paper I.

The recipes all contained the same liquid amino acid based product with

supplementary ingredients to disguise the taste and texture of the challenge

food. The use of a liquid test substance made it easy to prepare and easy to

dose. The limitation was the difficulty to hide a large enough amount of

challenge food in a drinkable amount of test substance. Therefore, the doses

of allergenic food given during the DBPFC sessions were low, ranging from

2.0-2.4 g of food protein. The challenge foods, except for cod, was used in

raw form in order to keep their allergenicity (186) and to increase the

sensitivity of the test, the DBPCFC was performed as a 3-session challenge

series including two challenge occasions with active substance (114,187).

Cytokine mRNA expression in PBMC

In Paper V, we chose to measure cytokine mRNA expression profiles in

PBMCs instead of measuring levels of biologically active cytokine proteins. The rationale for this decision was the high sensitivity and specificity of qRT-

PCR as a method and access to a well-established qRT-PCR for measurement

of cytokine mRNA expression. Importantly, a previous study by our group,

showed an excellent correlation between the levels of cytokine mRNA

measured by this method and detected levels of corresponding cytokines at

the protein levels (188).

The strengths of the study in Paper V, were the broad palette of cytokines

and fecal markers analyzed before and after in vivo stimulation with relevant

food allergens and the well-defined study participants. A study weakness was

70

that the study groups were small and the children heterogeneous according

to type of food allergy, symptom severity, challenge food and the serving

order of test samples during the DBPCFC series. The limited number of

participants was due to a low prevalence of allergy to basic foods, even

though we initially invited all children with reported allergy to milk, egg, cod

and wheat from a large population-based cohort with a high prevalence of

reported FHS to these foods. The difference in the serving order of active and

placebo substances was a consequence of serving orders being drawn by lot,

which was necessary to achieve a double-blind food challenge series (135).

The transport time was an important issue in our study, since the food

challenges were performed in hospitals 250-500 km from the laboratory that

performed the cytokine analyses. Thus, for practical and logistic reasons, the

post challenge samples in our study were collected one week after the

completed challenge series. The optimal time for measuring post challenge

biomarkers is a matter of discussion. The mRNA expression of different

cytokines peaks at different time intervals, which are different depending on

the detection method used. A Polish study investigating cytokine protein

levels in sera, showed a significant increase in the levels of IL-4 and IL-5

following double-blind challenges with active substance but not with placebo

in children with asthma and IgE-mediated food allergy. The maximum

secretion for these cytokines were seen after 24 hours (162). In contrary, the

levels of fecal inflammatory markers and specific IgE and IgG4 are likely to

peak weeks after allergen exposure (87,90,141,142).

Analyses of fecal biomarkers

For the analyses of inflammatory and tolerogenic markers fecal markers we

used commercially available ELISA methods (Immundiagnostik, Bensheim,

Germany). The ELISA method is the conventional method used for detecting

fecal calpotectin levels in inflammatory bowel disease, though other, more

repid tests are emerging on the market. A previous study showed that the

sensitivity and specificity is somewhat different in different methods of

measuring fecal calprotetin and that the same detection method therefore

should be used when comparing fecal levels of calprotectin at different time

intervals or between individuals (189).

71

Discussion of main results

Epidemiology of food hypersensitivity

The prevalence of parentally reported FHS in this population-based cohort

in Northern Sweden was high, 21%, at 7-8 years of age and even higher, 26%,

at 11-12 years of age. Although the majority of children with allergies to basic

foods i.e. milk and egg develop tolerance before school age (5,15) and lactose

intolerance is thought to be uncommon among Swedish schoolchildren (31),

the prevalence of reported hypersensitivity to milk, egg, cod and wheat

increased from 10% at 7-8 years of age to almost 15% at 11-12 years of age.

The prevalence of reported FHS to basic foods found in this study was high

compared to other studies on older schoolchildren and adolescents (9,11,16).

According to our definition, the prevalence of reported allergy to the basic

foods: milk, egg, cod and wheat at 11-12 years of age was 4.8%. When

validated with a clinical examination and analyses of food specific IgE and a

celiac screen test, the prevalence of current allergy to these foods was 1.4%.

This prevalence was further halved following a DBPCFC series. The

prevalence of challenge-proven allergy to milk, egg, cod and wheat of 0.6%

was consistent or somewhat lower compared to the few other studies using a

similar study approach (9,16,181) and 25-fold lower than the prevalence of

reported hypersensitivity to these foods in our study cohort.

In Papers III and IV we showed that current food allergy was an uncommon

cause of reported hypersensitivity to milk, egg, cod and wheat. The most

common FHS phenotypes were probable lactose intolerance followed by

outgrown allergy. At the time of the evaluation, children with outgrown

allergy could tolerate a portion size of the culprit food without any

symptoms, but they had nevertheless remained on an elimination diet. This

is consistent with previous studies, showing that many children with food

allergy keep avoiding the culprit food even after they have become tolerant

(133).

In Paper IV we also showed that milk allergy, current or outgrown, was

associated with a lower BMI compared to children not avoiding milk. A

possible reason for this finding is that children with milk allergy had

eliminated dairy products since very early life and still were avoiding them,

even though children with outgrown milk allergy probably did not have to.

This finding is in line with other studies showing that a milk eliminated diet

due to allergy may lead to impaired growth and lower intake of important

nutrients like vitamin D (182-184, 190).

72

Of the children in Paper IV still avoiding milk, 43% reported that their milk

hypersensitivity had been diagnosed by a physician and a previous physician

diagnosis was more common among children who were categorized as

having milk allergy. Only 2 % of the children reported that their milk

hypersensitivity had been established by an oral challenge and 9% of the

children still avoiding milk, reported that they had ever been referred to a

dietician for nutritional advice. Together, these data supports previous

findings that food hypersensitivity diagnoses given early in life may have

dietary and nutritional consequences in adolescence (189,191) and pinpoints

the importance of following up and properly evaluating medical advice given

regarding elimination of basic foods (5,192,193).

In Paper III, four out of nine children with a positive challenge outcome

reacted with anaphylaxis. The food-specific IgE among these children were

moderately elevated, ranging from 3.3-13.1 kU/L. This is consistent with the

findings of others, showing that food allergy severity cannot be determined

by the level of specific IgE (90,122,128). None of the children who reacted

with anaphylaxis had reported previous anaphylactic reactions to the

challenge food and only one of them had been prescribed an epinephrine

auto-injector. None of these children were on daily medication for their

concomitant asthma. These findings highlight the importance of recurrent

evaluations of children with suspected food allergy (5,116).

By the time of the structured interviews in Paper IV, 30% of the children

reporting milk avoidance due to perceived hypersensitivity at the Child

interview, had discontinued their milk-eliminated diet. All of these children

had previously reported partial milk avoidance. Stated reasons for

discontinuing the milk elimination diet were that the children had tried to

eliminate milk during a period, seeking alleviation for various symptoms, but

had found that the elimination diet did not result in any improvement or

that they were symptom-free independently of diet. This high prevalence of

discontinued milk elimination was in line with our findings in Paper II,

where we found a high incidence (15%) as well as a high remission (33%) of

parentally reported FHS between 7-8 years to 11-12 years of age. This pattern

was recognizable for hypersensitivity to many of the investigated foods, but

was most obvious for reported hypersensitivity to milk.

Reported hypersensitivity to milk had different symptom patterns as well as

different associated risk factors compared to non-milk FHS. In Paper II,

incidence of non-milk FHS was associated with allergic heredity as well as

current rhinitis and allergic sensitization, while asthma and allergic

sensitization were negatively associated with the remission of non-milk FHS.

73

Similar relationships between atopy related conditions and prevalence of

FHS have been found by others (35,37), although there is a lack of

population-based studies investigating risk factors associated with incidence

and remission of FHS in older schoolchildren and adolescents (15,19).

Neither incidence nor remission of milk FHS were associated with any other

atopy-related conditions. A possible explanation for this finding could be

that the current allergy phenotype was more common among children with

non-milk FHS compared to children with milk FHS (20,194). In Paper IV,

where children avoiding milk due to perceived hypersensitivity were sorted

into different milk hypersensitivity phenotypes, only 3% of these children

were categorized as current milk allergy and 23% as outgrown milk allergy.

The most common phenotype was probable lactose intolerance. While milk

allergy, current or outgrown, was associated with a number of atopy related

variables, children in the probable lactose intolerance phenotype were less

sensitized to aeroallergens and food allergens compared to children not

avoiding milk.

Female sex was associated with incidence of any FHS, non-milk FHS and

milk FHS. It was also associated with milk avoidance due to perceived

hypersensitivity as well as having a milk allergy phenotype at 11-12 years of

age. This is in line with the findings of others, showing that while IgE-

sensitization to food allergens and food allergy is more common among

younger boys than girls, food allergy is more common among adolescent

girls and women compared to men (59,66). The reason for this shift is

unknown, but thought to be related to hormonal factors (67). In addition,

previous studies have shown that women more often report symptoms of

lactose intolerance than men (70,72).

In Paper II, we also investigated the relationship between IgE-sensitization

to milk, egg, soy, cod, wheat and peanut and parentally reported symptoms

to the corresponding food in their child. Generally, there was a poor

agreement between reported symptoms and a positive specific IgE to the

culprit food and the agreement was weakest for milk, soy and wheat. This is

consistent with other studies (90, 195) as well as with our findings in Paper

III and IV showing that current IgE-mediated allergy was an uncommon

course of reported hypersensitivity to basic foods in this cohort of 11-12 year

olds.

Diagnosing food allergy

In this project, we successfully validated recipes for low-dose DBPCFC with

cow’s milk, hen’s egg, soy, cod and wheat in children. These recipes were

74

then used in DBPCFC series for validating reported allergy to basic foods,

but the validation of our recipes has also allowed a spread of these recipes in

order to facilitate the use of DBPCFC in daily clinical practice. It has been

shown in previous studies that the performance of an oral food challenge can

reduce the fear of future adverse reactions to the culprit food but may also

improve quality of life in families with children with suspected food allergy

(196). Even though DBPCFCs are considered gold standard for diagnosing or

ruling out food allergy, the methodology has not yet been fully standardized

(135,136). The access to validated DBPCFC recipes may facilitate such a

standardization, which will make results from studies using DBPCFCs easier

to compare.

Since DBPCFCs are time and resource consuming, there is a search of

objective prognostic markers that could predict challenge outcome and

thereby make DBPCFCs superfluous. In Paper V, we showed that the

baseline mRNA levels of the Th2-related cytokine IL-13 were higher in

children with challenge-proven food allergy compared to children with a

negative challenge outcome compared to healthy controls. It has been shown

in other studies that high levels of IL-13 are expressed after stimulation of

PBMCs with relevant food allergens in food allergic patients and that the IL-

13 level decreases after tolerance development (163,164). In addition,

Metcalfe et al recently showed that an elevated egg-specific Th2 cytokine

response – of IL-13 and IL-5 in particular – at 4 months of age could predict

egg allergy at 12 months of age (161), which could be in line with our

findings.

Further, we showed that the expressed mRNA level of the regulatory

cytokine IL-10 at baseline was higher among children with challenge-proven

food allergy compared to children with a negative challenge outcome. Higher

IL-10 levels have been shown following stimulation of PBMCs with cow’s

milk protein in children with IgE-mediated cow’s milk allergy compared to

children with non-IgE mediated milk allergy (163,197) and in children

during tolerance development compared to children with symptomatic food

allergy (164).Previous studies have also shown that after stimulation of

PBMCs with specific food allergens the expression of the Th2 cytokine

pattern is enhanced and the expressed levels of the regulatory cytokine IL-10

diminished in children with symptomatic IgE-mediated allergy compared to

children who have developed tolerance (107,161). Due to its active role in

tolerance development, IL-10 has been discussed as a possible biomarker for

symptomatic IgE-mediated food allergy (101). However, since data are not

univocal and study results vary with study population and investigated food

(108,164,198) it has been suggested that analyzing IL-10 as part of a broader

Th2-profile would be preferable (103,199,200).

75

In Paper V, we also investigated the fecal levels of EDN as a marker of

eosinophilic inflammation and calprotectin as a marker of neutrophilic

inflammation of the gastrointestinal tract, in relation to DBPCFC outcome.

The baseline levels of EDN and the baseline and post challenge levels of fecal

calprotectin were almost twice as high among children with a positive

compared to a negative challenge outcome, although none of these

differences reached statistical significance. This was probably due to a lack

of power caused by the low number of participants toghether with the large

natural variance in faecal calprotectin levels (141,189). Our findings are,

however, in line with the results from other studies showing that fecal levels

of EDN might be used as a marker of food allergic reactions (139,140) and

that measurement of fecal calprotectin can be a helpful tool in the follow-up

of response to treatment and determination of reoccurrence of cow’s milk

allergy (141,142).

Ethical considerations

The benefit of our study participants was the possibility to be diagnosed or

freed from a suspected allergy to basic foods by the use of gold standard

methods (5,116). To secure the safety of our study participants during

DBPCFC, the double-blind challenges were performed in a hospital setting

with drugs and equipment for the treatment of allergic symptoms, including

anaphylaxis, immediately available. Also, the same experienced allergist

(AW) and allergy nurse (AS) were responsible at all challenge occasions.

Before each challenge occasion, an intravenous line was set in case of need to

administer intravenous drugs or rehydration fluids. This intravenous line

was also used for blood sampling when appropriate, to avoid unnecessary

venous punctures. Children with positive tTGA-tests and no previous

diagnosis of celiac disease, were informed about the test result and referred

to a pediatric clinic for further investigation.

The children participating in the validation of our DBPCFC recipes and the

analyses of cytokine mRNA expressions, had no immediate personal benefits

of participating in the study, neither were they exposed to any health-related

risks caused by their participation, except for the discomfort of donating

blood samples in Paper 5. These children have, however, contributed to

research that will hopefully improve future diagnostics and care-taking of

children with suspected food allergy and consequently decrease unnecessary

food avoidance.

76

77

Personal point of closure

I have learned a lot during these years of doctoral studies, adding new

dimensions to my clinical work. A better understanding of food

hypersensitivity in schoolchildren, its prevalence, clinical expressions and

underlying phenotypes has increased the interest and engagement in our

allergy team to disentangle the cause of adverse reactions to foods in our

pediatric patients and to improve our diagnostic methods. The most

significant lessons I have learned during this project were that perceived

food hypersensitivity to basic foods was even more common, and that the

percentage of children with current allergy to these foods lower, than I first

thought. Further, that medical advice on food avoidance early in life, may

have nutritional consequences in adolescence.

Even though I find our attempts to diagnose or free children with suspected

food hypersensitivity important, this study has made me realize that to

achieve significant changes, we have to start much earlier on. Since most

children with allergies to basic foods develop tolerance before school age and

since the older the child the more difficult it is for them to accept new tastes

and textures, we must become better at evaluating diagnostic elimination

diets and at following-up children with suspected food hypersensitivity to

avoid unnecessary food avoidances. It is of course not necessary for everyone

to eat all kinds of foods, but since I believe it to be a good start of a healthy

relationship to food, I want as many children as possible to have the

opportunity to do so. Following-up and properly diagnosing children with

suspected food hypersensitivity will also diminish the risk of children with

severe food allergy being disbelieved and thereby liable to risks of

unnecessary exposure to the culprit food.

Our developed recipes for double-blind challenges in children have been

used frequently in our daily clinical practice and the validation of these

recipes has allowed a spread to other clinics in the country. Further, we have

gained knowledge about how schoolchildren experience food hyper-

sensitivity and food avoidance, which has facilitated improvement of our

patient follow-ups. The upcoming theses of my friend and colleague Åsa

Strinnholm, will contribute additional knowledge on quality of life in

adolescents with perceived food allergy. Also, continuous studies and new

research from the OLIN group and the department of Pediatrics, Umeå

University will contribute to future knowledge on epidemiology and risk

factors of food hypersensitivity – and I will probably not be able to keep my

fingers out of that jar…

78

79

Acknowledgements

Thank You…

To Our Study Participants

…to all of the families answering our Call

Without you this Work would be nothing at all

To the OLIN family

…for selflessly sharing your research skills

…for dragging and pushing me over the hills

To the OLIN “study personnel”

…you´re the true heroes – one must understand

Collectors of data from “Nowhereland”

To Annika and Anette

at the department of Food and Nutrition

…for successfully blindfolding class after class

…for the excellent taste of working with us

To Lucia, Olga, Ivan, Ignacio, Catharina and Agneta

at the department of Immunology

…for handling our samples – and doing it well

…for teaching me methods I hardly can spell

80

Thank You…

To Catarina, Yvonne and the other amazing ladies

at the “Pediatric Research Unit”

…for always having a hand to lend

…for being the pillars upon who we depend

To Ulla, Karin, Elisabeth, Carina, Tove and Marlene

…for unvaluable help and for keeping all

the “beg, steel and borrowing” under control

To my PhD-student collegues

…for the scanning of papers – for wrongs and for rights

…for fellowship, fun and “Falafel-fights”

To Helena, the best (looking) Statistician ever

…for making logaritms natural, p less bizarre

A statistical lifeboat – that´s what you R

To the best Co-authors ever

…for forbearance and patience, for taking your time

with all of my synonyms – preferably in rhyme

To Christina and Eva, the best Supervisors ever

…for friendship and wisdom – for letting me thrive

For parts of this journey you´ve kept me alive

To Åsa and Lisbeth, my Partners in crime:

…for being my “Sisters in Arms” through the years

rumbling the country – like three musketeers

81

Thank You…

To Per Erik, the best Boss ever

…for all of the hours of research I got

…for believing in me when I – frankly – did not

To my Colleagues at the department of Pediatrics

…for lightening the burdens, for fun of all sorts

…for vivid discussions and friendly supports

To the best Allergy team ever

…my Backbone and Strength – I depend on you all –

to keep everything strictly under control

To the best Pulmonary and Respiratory teams ever

…your cunning and skills reassure that I might

have a nightmare-free, pressure-less sleep every night

To my Mentors: Christian and Stig

…for the sake of deliberately befooling me

into the path of Allergology

To my Gurus: Olle, Solveig and Sten-Erik

…for the answering of questions – giant or small -

whenever I needed a friend to call

To the “Children´s choir”

(or “kväkare utan gränser” or “the unreal group”)

…for the music and also – for good and for worse –

the art of preforming without a rehearse

82

Thank You…

To old and new aquaintances

…to all of the fabulous friends that I´ve got

You´re the “treasure of life” – and I love you a lot

To Karl-Gunnar, Ing-Britt, Pentti, Leena, Per and Ina

…for all your support, all the strings you´ve unnerved

You are Relatives better than I have deserved

To my Brother Olle

…you´re a piece of me missing, distant but near

You´re a part of this too – even though you´re not here

To Mikael, Kalle, Emil and Gustav

…my Husband, my Children – my dear Family

Through all of these Chapters you´ve been there for me

Without your sweet loving – where would I be?

83

Funding

This thesis was funded by:

The Swedish Research Council

The Swedish Heart and Lung Foundation

VISARE Norr

The Swedish Asthma and Allergy Foundation

ALF - a regional agreement between Umeå University and Väster-botten and Norrbotten County Council

FoU - the State Government funding for Health Care research

Norrbotten County Council

The Swedish Society of Medicine

Insamlingsstiftelsen

The Sven Jerring Foundation

The Kempe Foundation

The Oskar foundation

The Anerska Foundation

The Kjerstin Hejdenberg Foundation

Stiftelsen Samariten

BLF – the section for Allergology

Nutricia AB is acknowledged for providing DBPCFC test material

ThermoFisher Diagnostics, Uppsala, Sweden provided funding for parts of the IgE tests and tTGA analyses

84

References

1.Prescott SL, Pawankar R, Allen KJ, Campbell DE, Sinn J, Fiocchi A, et al. A global survey of changing patterns of food allergy burden in children. World Allergy Organ J 2013;6(1):21.

2.Johansson SG, Bieber T, Dahl R, Friedmann PS, Lanier BQ, Lockey RF, et al. Revised nomenclature for allergy for global use: Report of the Nomenclature Review Committee of the World Allergy Organization, October 2003. J Allergy Clin Immunol 2004;113(5):832-6.

3.Sicherer SH, Sampson HA. Food allergy. J Allergy Clin Immunol 2010;125(2 Suppl 2):S116-25.

4.Nowak-Wegrzyn A, Sampson HA. Adverse reactions to foods. Med Clin North Am. 2006;90(1):97-127.

5.Burks AW, Tang M, Sicherer S, Muraro A, Eigenmann PA, Ebisawa M, et al. ICON: food allergy. J Allergy Clin Immunol 2012;129(4):906-20.

6.Nowak-Wegrzyn A, Katz Y, Mehr SS, Koletzko S. Non-IgE-mediated gastrointestinal food allergy. J Allergy Clin Immunol 2015;135(5):1114-24.

7.Wilt TJ, Shaukat A, Shamliyan T, Taylor BC, MacDonald R, Tacklind J, et al. Lactose intolerance and health. Evid Rep Technol Assess 2010(192):1-410.

8.Mills EN, Mackie AR, Burney P, Beyer K, Frewer L, Madsen C, et al. The prevalence, cost and basis of food allergy across Europe. Allergy. 2007;62(7):717-22.

9.Rona RJ, Keil T, Summers C, Gislason D, Zuidmeer L, Sodergren E, et al. The prevalence of food allergy: a meta-analysis. J Allergy Clin Immunol 2007;120(3):638-46.

10.Enroth S, Dahlbom I, Hansson T, Johansson A, Gyllensten U. Prevalence and sensitization of atopic allergy and coeliac disease in the Northern Sweden Population Health Study. Int J Circumpolar Health. 2013;5(72).

11.Osterballe M, Mortz CG, Hansen TK, Andersen KE, Bindslev-Jensen C. The prevalence of food hypersensitivity in young adults. Pediatr Allergy Immunol 2009;20(7):686-92.

12.Sicherer SH, Sampson HA. Food allergy: Epidemiology, pathogenesis, diagnosis, and treatment. J Allergy Clin Immunol 2014;133(2):291-307.

85

13.Prescott S, Allen KJ. Food allergy: riding the second wave of the allergy epidemic. Pediatr Allergy Immunol 2011;22(2):155-60.

14.Nwaru BI, Hickstein L, Panesar SS, Muraro A, Werfel T, Cardona V, et al. The epidemiology of food allergy in Europe: a systematic review and meta-analysis. Allergy 2014;69(1):62-75.

15.Schoemaker AA, Sprikkelman AB, Grimshaw KE, Roberts G, Grabenhenrich L, Rosenfeld L, et al. Incidence and natural history of challenge-proven cow's milk allergy in European children - EuroPrevall birth cohort. Allergy 2015;70(8):963-72.

16.Nwaru BI, Hickstein L, Panesar SS, Roberts G, Muraro A, Sheikh A. Prevalence of common food allergies in Europe: a systematic review and meta-analysis. Allergy 2014;69(8):992-1007.

17.Allen KJ, Koplin JJ. The epidemiology of IgE-mediated food allergy and anaphylaxis. Immunol Allergy Clin North Am 2012;32(1):35-50.

18.Eller E, Kjaer HF, Host A, Andersen KE, Bindslev-Jensen C. Food allergy and food sensitization in early childhood: results from the DARC cohort. Allergy 2009;64(7):1023-9.

19.Venter C, Pereira B, Voigt K, Grundy J, Clayton CB, Higgins B, et al. Prevalence and cumulative incidence of food hypersensitivity in the first 3 years of life. Allergy 2008;63(3):354-9.

20.Elizur A, Rajuan N, Goldberg MR, Leshno M, Cohen A, Katz Y. Natural course and risk factors for persistence of IgE-mediated cow's milk allergy. J Pediatr 2012;161(3):482-7.

21.Skripak JM, Matsui EC, Mudd K, Wood RA. The natural history of IgE-mediated cow's milk allergy. J Allergy Clin Immunol 2007;120(5):1172-7.

22.Kattan JD, Cocco RR, Jarvinen KM. Milk and soy allergy. Pediatr Clin North Am 2011;58(2):407-26.

23.Ho MH, Wong WH, Heine RG, Hosking CS, Hill DJ, Allen KJ. Early clinical predictors of remission of peanut allergy in children. J Allergy Clin Immunol 2008;121(3):731-6.

24.Hourihane JO. Peanut allergy. Pediatr Clin North Am 2011;58(2):445-58.

25.Geroldinger-Simic M, Zelniker T, Aberer W, Ebner C, Egger C, Greiderer A, et al. Birch pollen-related food allergy: clinical aspects and the role of allergen-specific IgE and IgG4 antibodies. J Allergy Clin Immunol 2011;127(3):616-22.

86

26.Warm K, Backman H, Lindberg A, Lundbäck B, Rönmark E. Low incidence and high remission of allergic sensitization among adults. J Allergy Clin Immunol 2012;129(1):136-42.

27.Rönmark E, Bjerg A, Perzanowski M, Platts-Mills T, Lundbäck B. Major increase in allergic sensitization in schoolchildren from 1996 to 2006 in northern Sweden. J Allergy Clin Immunol 2009;124(2):357-63, 63.

28.Allen KJ, Davidson GP, Day AS, Hill DJ, Kemp AS, Peake JE, et al. Management of cow's milk protein allergy in infants and young children: an expert panel perspective. J Paediatr Child Health. 2009;45(9):481-6.

29.Katz Y, Goldberg MR, Rajuan N, Cohen A, Leshno M. The prevalence and natural course of food protein-induced enterocolitis syndrome to cow's milk: a large-scale, prospective population-based study. J Allergy Clin Immunol 2011;127(3):647-53.

30.Ingram CJ, Mulcare CA, Itan Y, Thomas MG, Swallow DM. Lactose digestion and the evolutionary genetics of lactase persistence. Hum Genet 2009;124(6):579-91.

31.Vuorisalo T, Arjamaa O, Vasemägi A, Taavitsainen JP, Tourunen A, Saloniemi I. High lactose tolerance in North Europeans: a result of migration, not in situ milk consumption. Perspect Biol Med 2012;55:163-74.

32.Almon R, Engfeldt P, Tysk C, Sjöstrom M, Nilsson TK. Prevalence and trends in adult-type hypolactasia in different age cohorts in Central Sweden diagnosed by genotyping for the adult-type hypolactasia-linked LCT -13910C > T mutation. Scand J Gastroenterol 2007;42(2):165-70.

33.Carroccio A, Montalto G, Cavera G, Notarbatolo A. Lactose intolerance and self-reported milk intolerance: relationship with lactose maldigestion and nutrient intake. Lactase Deficiency Study Group. J Am Coll Nutr 1998;17(6):631-6.

34.Paajanen L, Korpela R, Tuure T, Honkanen J, Järvelä I, Ilonen J, et al. Cow milk is not responsible for most gastrointestinal immune-like syndromes--evidence from a population-based study. Am J Clin Nutr 2005;82(6):1327-35.

35.Östblom E, Lilja G, Pershagen G, van Hage M, Wickman M. Phenotypes of food hypersensitivity and development of allergic diseases during the first 8 years of life. Clin Exp Allergy 2008;38(8):1325-32.

87

36.Strinnholm Å, Winberg A, West C, Hedman L, Rönmark E. Food hypersensitivity is common in Swedish schoolchildren, especially oral reactions to fruit and gastrointestinal reactions to milk. Acta Paediatr 2014;103(12):1290-6.

37.Soost S, Leynaert B, Almqvist C, Edenharter G, Zuberbier T, Worm M. Risk factors of adverse reactions to food in German adults. Clin Exp Allergy 2009;39(7):1036-44.

38.Sandin A, Annus T, Björksten B, Nilsson L, Riikjarv MA, van Hage-Hamsten M, et al. Prevalence of self-reported food allergy and IgE antibodies to food allergens in Swedish and Estonian schoolchildren. Eur J Clin Nutr 2005;59(3):399-403.

39.Tan TH, Ellis JA, Saffery R, Allen KJ. The role of genetics and environment in the rise of childhood food allergy. Clin Exp Allergy 2012;42(1):20-9.

40.Fuertes E, Standl M, von Berg A, Lehmann I, Hoffmann B, Bauer CP, et al. Parental allergic disease before and after child birth poses similar risk for childhood allergies. Allergy 2015;70(7):873-6.

41.Hong X, Tsai HJ, Wang X. Genetics of food allergy. Curr Opin Pediatr 2009;21(6):770-6.

42.West CE, Renz H, Jenmalm MC, Kozyrskyj AL, Allen KJ, Vuillermin P, et al. The gut microbiota and inflammatory noncommunicable diseases: associations and potentials for gut microbiota therapies. J Allergy Clin Immunol 2015;135(1):3-13.

43.Wold AE. The hygiene hypothesis revised: is the rising frequency of allergy due to changes in the intestinal flora? Allergy. 1998;53:20-5.

44.Laursen MF, Zachariassen G, Bahl MI, Bergström A, Host A, Michaelsen KF, et al. Having older siblings is associated with gut microbiota development during early childhood. BMC Microbiol. 2015;15(154):015-0477.

45.Campbell BE, Lodge CJ, Lowe AJ, Burgess JA, Matheson MC, Dharmage SC. Exposure to 'farming' and objective markers of atopy: a systematic review and meta-analysis. Clin Exp Allergy 2015;45(4):744-57.

46.Amberbir A, Medhin G, Abegaz WE, Hanlon C, Robinson K, Fogarty A, et al. Exposure to Helicobacter pylori infection in early childhood and the risk of allergic disease and atopic sensitization: a longitudinal birth cohort study. Clin Exp Allergy 2014;44(4):563-71.

88

47.Lynch SV, Wood RA, Boushey H, Bacharier LB, Bloomberg GR, Kattan M, et al. Effects of early-life exposure to allergens and bacteria on recurrent wheeze and atopy in urban children. J Allergy Clin Immunol 2014;134(3):593-601.

48.Pyrhönen K, Näyhä S, Hiltunen L, Läärä E. Caesarean section and allergic manifestations: insufficient evidence of association found in population-based study of children aged 1 to 4 years. Acta Paediatr 2013;102(10):982-9.

49.West CE, Jenmalm MC, Prescott SL. The gut microbiota and its role in the development of allergic disease: a wider perspective. Clin Exp Allergy 2015;45(1):43-53.

50.Koplin JJ, Dharmage SC, Ponsonby AL, Tang ML, Lowe AJ, Gurrin LC, et al. Environmental and demographic risk factors for egg allergy in a population-based study of infants. Allergy. 2012.

51.Sharief S, Jariwala S, Kumar J, Muntner P, Melamed ML. Vitamin D levels and food and environmental allergies in the United States: results from the National Health and Nutrition Examination Survey 2005-2006. J Allergy Clin Immunol 2011;127(5):1195-202.

52.Allen KJ, Koplin JJ, Ponsonby AL, Gurrin LC, Wake M, Vuillermin P, et al. Vitamin D insufficiency is associated with challenge-proven food allergy in infants. J Allergy Clin Immunol 2013;131(4):1109-16.

53.Magnusson J, Kull I, Westman M, Håkansson N, Wolk A, Melen E, et al. Fish and polyunsaturated fat intake and development of allergic and nonallergic rhinitis. J Allergy Clin Immunol 2015. Epub 2015/07/15.

54.Lack G, Plaut M, Sayre PH. Peanut consumption in infants at risk for peanut allergy. N Engl J Med 2015;372(22):2165-6.

55.Palmer DJ, Metcalfe J, Makrides M, Gold MS, Quinn P, West CE, et al. Early regular egg exposure in infants with eczema: A randomized controlled trial. J Allergy Clin Immunol. 2013;132(2):387-92.

56.Lack G. Update on risk factors for food allergy. J Allergy Clin Immunol 2012;129(5):1187-97.

57.Ben-Shoshan M, Soller L, Harrington DW, Knoll M, La Vieille S, Fragapane J, et al. Eczema in early childhood, sociodemographic factors and lifestyle habits are associated with food allergy: a nested case-control study. Int Arch Allergy Immunol 2015;166(3):199-207.

58.Just J, Deslandes-Boutmy E, Amat F, Desseaux K, Nemni A, Bourrat E, et al. Natural history of allergic sensitization in infants with early-onset atopic

89

dermatitis: results from ORCA Study. Pediatr Allergy Immunol 2014;25(7):668-73.

59.Liu AH, Jaramillo R, Sicherer SH, Wood RA, Bock SA, Burks AW, et al. National prevalence and risk factors for food allergy and relationship to asthma: results from the National Health and Nutrition Examination Survey 2005-2006. J Allergy Clin Immunol 2010;126(4):798-806.

60.Pinart M, Benet M, Annesi-Maesano I, von Berg A, Berdel D, Carlsen KC, et al. Comorbidity of eczema, rhinitis, and asthma in IgE-sensitised and non-IgE-sensitised children in MeDALL: a population-based cohort study. The Lancet Respir Med 2014;2(2):131-40.

61.Santos A, Dias A, Pinheiro JA. Predictive factors for the persistence of cow's milk allergy. Pediatr Allergy Immunol. 2010;21(8):1127-34.

62.Pyrhönen K, Hiltunen L, Kaila M, Näyhä S, Läärä E. Heredity of food allergies in an unselected child population: an epidemiological survey from Finland. Pediatr Allergy Immunol 2011;22(1 Pt 2):e124-32.

63.Goldberg M, Eisenberg E, Elizur A, Rajuan N, Rachmiel M, Cohen A, et al. Role of parental atopy in cow's milk allergy: a population-based study. Ann Allergy Asthma Immunol 2013;110(4):279-83.

64.Ashley S, Dang T, Koplin J, Martino D, Prescott S. Food for thought: progress in understanding the causes and mechanisms of food allergy. Curr Opin Allergy Clin Immunol 2015;15(3):237-42.

65.Kumar R, Tsai HJ, Hong X, Liu X, Wang G, Pearson C, et al. Race, ancestry, and development of food-allergen sensitization in early childhood. Pediatrics 2011;128(4):e821-9.

66.DunnGalvin A, Hourihane JO, Frewer L, Knibb RC, Oude Elberink JN, Klinge I. Incorporating a gender dimension in food allergy research: a review. Allergy 2006;61(11):1336-43.

67.Chen W, Mempel M, Schober W, Behrendt H, Ring J. Gender difference, sex hormones, and immediate type hypersensitivity reactions. Allergy 2008;63(11):1418-27.

68.Allen KJ, Martin PE. Clinical aspects of pediatric food allergy and failed oral immune tolerance. J Clin Gastroenterol 2010;44(6):391-401.

69.Haberkorn BC, Ermens AA, Koeken A, Cobbaert CM, van Guldener C. Improving diagnosis of adult-type hypolactasia in patients with abdominal complaints. Clin Chem Lab Med 2011;50(1):119-23.

90

70.Baadkar SV, Mukherjee MS, Lele SS. Study on influence of age, gender and genetic variants on lactose intolerance and its impact on milk intake in adult Asian Indians. Ann Hum Biol 2014;41(6):548-53.

71.Rasinperä H, Saarinen K, Pelkonen A, Järvelä I, Savilahti E, Kolho KL. Molecularly defined adult-type hypolactasia in school-aged children with a previous history of cow's milk allergy. World J Gastroenterol 2006;12(14):2264-8.

72.Barr SI. Perceived lactose intolerance in adult Canadians: a national survey. Appl Physiol Nutr Metab 2013;38(8):830-5.

73.Kaiko GE, Horvat JC, Beagley KW, Hansbro PM. Immunological decision-making: how does the immune system decide to mount a helper T-cell response? Immunology 2008;123(3):326-38.

74.Houghteling PD, Walker WA. From Birth to "Immunohealth," Allergies and Enterocolitis. J Clin Gastroenterol 2015;49(1):S7-S12.

75.Annunziato F, Romagnani P, Cosmi L, Lazzeri E, Romagnani S. Chemokines and lymphopoiesis in human thymus. Trends Immunol 2001;22(5):277-81.

76.Rothenberg EV. Stepwise specification of lymphocyte developmental lineages. Curr Opin Genet Dev 2000;10(4):370-9.

77.Mosmann TR, Kobie JJ, Lee FE, Quataert SA. T helper cytokine patterns: defined subsets, random expression, and external modulation. Immunol Res 2009;45(2-3):173-84.

78.Akdis M, Verhagen J, Taylor A, Karamloo F, Karagiannidis C, Crameri R, et al. Immune responses in healthy and allergic individuals are characterized by a fine balance between allergen-specific T regulatory 1 and T helper 2 cells. J Exp Med 2004;199(11):1567-75.

79.Schmidt-Weber CB, Akdis M, Akdis CA. TH17 cells in the big picture of immunology. J Allergy Clin Immunol 2007;120(2):247-54.

80.Hinks TS, Zhou X, Staples KJ, Dimitrov BD, Manta A, Petrossian T, et al. Innate and adaptive T cells in asthmatic patients: Relationship to severity and disease mechanisms. J Allergy Clin Immunol 2015;136(2):323-33.

81.79.Akdis M, Blaser K, Akdis CA. T regulatory cells in allergy: novel concepts in the pathogenesis, prevention, and treatment of allergic diseases. J Allergy Clin Immunol 2005;116(5):961-8.

91

82.Palomares O, Martin-Fontecha M, Lauener R, Traidl-Hoffmann C, Cavkaytar O, Akdis M, et al. Regulatory T cells and immune regulation of allergic diseases: roles of IL-10 and TGF-beta. Genes Immun 2014;15(8):511-20.

83.Pandiyan P, Zhu J. Origin and functions of pro-inflammatory cytokine producing Foxp3(+) regulatory T cells. Cytokine 2015;76(1):13-24.

84.Huang W, August A. The signaling symphony: T cell receptor tunes cytokine-mediated T cell differentiation. J Leukoc Biol 2015;97(3):477-85.

85.Zhang H, Kong H, Zeng X, Guo L, Sun X, He S. Subsets of regulatory T cells and their roles in allergy. J Transl Med 2014;12(125):1479-5876.

86.Yao Y, Vent-Schmidt J, McGeough MD, Wong M, Hoffman HM, Steiner TS, et al. Tr1 Cells, but Not Foxp3+ Regulatory T Cells, Suppress NLRP3 Inflammasome Activation via an IL-10-Dependent Mechanism. J Immunol 2015;195(2):488-97.

87.Kumar S, Verma AK, Das M, Dwivedi PD. Molecular mechanisms of IgE mediated food allergy. Int Immunopharmacol 2012;13(4):432-9.

88.Kim EH, Burks W. Immunological basis of food allergy (IgE-mediated, non-IgE-mediated, and tolerance). Chem Immunol Allergy 2015;101:8-17.

89.Eigenmann PA. Mechanisms of food allergy. Pediatr Allergy Immunol 2009;20(1):5-11.

90.Sicherer SH, Wood RA. Allergy testing in childhood: using allergen-specific IgE tests. Pediatrics 2012;129(1):193-7.

91.Lorenz AR, Scheurer S, Vieths S. Food allergens: molecular and immunological aspects, allergen databases and cross-reactivity. Chem Immunol Allergy 2015;101:18-29.

92.Heine RG. Gastrointestinal food allergies. Chem Immunol Allergy 2015;101:171-80.

93.MacDonald TT, Di Sabatino A. The immunologic basis for gastrointestinal food allergy. Current opinion in gastroenterology. 2009;25(6):521-6.

94.Bergmann MM, Eigenmann PA. Food allergy in childhood (infancy to school age). Chem Immunol Allergy 2015;101:38-50.

95.O'Shea JJ, Gadina M, Kanno Y. Cytokine signaling: birth of a pathway. J Immunol 2011;187(11):5475-8.

92

96.Galli SJ, Tsai M, Piliponsky AM. The development of allergic inflammation. Nature 2008;454(7203):445-54.

97.Akdis CA, Akdis M. Mechanisms of immune tolerance to allergens: role of IL-10 and Tregs. J Clin Invest 2014;124(11):4678-80

98.Wisniewski J, Agrawal R, Woodfolk JA. Mechanisms of tolerance induction in allergic disease: integrating current and emerging concepts. Clin Exp Allergy 2013;43(2):164-76.

99.Stanic B, van de Veen W, Wirz OF, Ruckert B, Morita H, Sollner S, et al. IL-10-overexpressing B cells regulate innate and adaptive immune responses. J Allergy Clin Immunol 2015;135(3):771-80.

100.Till S. Mechanisms of immunotherapy and surrogate markers. Allergy 2011;66 Suppl 95:25-7.

101.Alonso R, Pineda F, Enrique E, Tella R, Cistero-Bahima A. Usefulness of serum interleukin-10 in determining food tolerance. Allergy 2007;62:710-1.

102.Sommanus S, Kerddonfak S, Kamchaisatian W, Vilaiyuk S, Sasisakulporn C, Teawsomboonkit W, et al. Cow's milk protein allergy: immunological response in children with cow's milk protein tolerance. Asian Pac J Allergy Immunol 2014;32(2):171-7.

103.Tiemessen MM, Van Ieperen-Van Dijk AG, Bruijnzeel-Koomen CA, Garssen J, Knol EF, Van Hoffen E. Cow's milk-specific T-cell reactivity of children with and without persistent cow's milk allergy: key role for IL-10. J Allergy Clin Immunol 2004;113(5):932-9.

104.Lee JH, Noh J, Noh G, Choi WS, Cho S, Lee SS. Allergen-specific transforming growth factor-beta-producing CD19+CD5+ regulatory B-cell (Br3) responses in human late eczematous allergic reactions to cow's milk. J Interferon Cytokine Res 2011;31(5):441-9.

105.Jacob CM, Pastorino AC, Okay TS, Castro AP, Gushken AK, Watanabe LA, et al. Interleukin 10 (IL10) and transforming growth factor beta1 (TGFbeta1) gene polymorphisms in persistent IgE-mediated cow's milk allergy. Clinics 2013;68(7):1004-9.

106.Williams CM, Rahman S, Hubeau C, Ma HL. Cytokine pathways in allergic disease. Toxicol Pathol 2012;40(2):205-15.

107.Schade RP, Van Ieperen-Van Dijk AG, Van Reijsen FC, Versluis C, Kimpen JL, Knol EF, et al. Differences in antigen-specific T-cell responses between infants with atopic dermatitis with and without cow's milk allergy: relevance of TH2 cytokines. J Allergy Clin Immunol 2000;106(6):1155-62.

93

108.Scott-Taylor TH, Hourihane JB, Harper J, Strobel S. Patterns of food allergen-specific cytokine production by T lymphocytes of children with multiple allergies. Clin Exp Allergy 2005;35(11):1473-80.

109.Neaville WA, Tisler C, Bhattacharya A, Anklam K, Gilbertson-White S, Hamilton R, et al. Developmental cytokine response profiles and the clinical and immunologic expression of atopy during the first year of life. J Allergy Clin Immunol 2003;112(4):740-6.

110.Tomicic S, Fälth-Magnusson K, Böttcher MF. Dysregulated Th1 and Th2 responses in food-allergic children--does elimination diet contribute to the dysregulation? Pediatr Allergy Immunol 2010;21(4 Pt 1):649-55.

111.Sampson HA. Food allergy: immunology of the GI mucosa towards classification and understanding of GI hypersensitivities. Pediatr Allergy Immunol 2001;12 Suppl 14:7-9.

112.Saarinen KM, Sarnesto A, Savilahti E. Markers of inflammation in the feces of infants with cow's milk allergy. Pediatr Allergy Immunol 2002;13(3):188-94.

113.Gupta RS, Springston EE, Warrier MR, Smith B, Kumar R, Pongracic J, et al. The prevalence, severity, and distribution of childhood food allergy in the United States. Pediatrics 2011;128(1):e9-17.

114.Niggemann B, Beyer K. Pitfalls in double-blind, placebo-controlled oral food challenges. Allergy 2007;62(7):729-32.

115.Kleiman J, Ben-Shoshan M. Food-dependent exercise-induced anaphylaxis with negative allergy testing. BMJ Case Rep 2014;2014.

116.Nowak-Wegrzyn A. Food allergy guidelines and beyond. Immunol Allergy Clin North Am 2012;32(1):xv-xix.

117.Pesek RD, Jones SM. Anaphylaxis in food allergy. Chem Immunol Allergy 2015;101:191-8.

118.Panesar SS, Javad S, de Silva D, Nwaru BI, Hickstein L, Muraro A, et al. The epidemiology of anaphylaxis in Europe: a systematic review. Allergy 2013;68(11):1353-61.

119.Hofmann A, Burks AW. Pollen food syndrome: update on the allergens. Curr Allergy Asthma Rep 2008;8(5):413-7.

120.Asarnoj A, Nilsson C, Lidholm J, Glaumann S, Östblom E, Hedlin G, et al. Peanut component Ara h 8 sensitization and tolerance to peanut. J Allergy Clin Immunol 2012;130(2):468-72.

94

121.Järvinen KM, Nowak-Wegrzyn A. Food protein-induced enterocolitis syndrome (FPIES): current management strategies and review of the literature. J Allergy Clin Immunol 2013;1(4):317-22.

122.Soares-Weiser K, Takwoingi Y, Panesar SS, Muraro A, Werfel T, Hoffmann-Sommergruber K, et al. The diagnosis of food allergy: a systematic review and meta-analysis. Allergy 2014;69(1):76-86.

123.Koletzko S, Niggemann B, Arato A, Dias JA, Heuschkel R, Husby S, et al. Diagnostic approach and management of cow's-milk protein allergy in infants and children: ESPGHAN GI Committee practical guidelines J Pediatr Gastroenterol Nutr 2012;55(2):221-

124.Skypala IJ, Venter C, Meyer R, deJong NW, Fox AT, Groetch M, et al. The development of a standardised diet history tool to support the diagnosis of food allergy. Clin Transl Allergy 2015;5:7.

125.Mehl A, Niggemann B, Keil T, Wahn U, Beyer K. Skin prick test and specific serum IgE in the diagnostic evaluation of suspected cow's milk and hen's egg allergy in children: does one replace the other? Clin Exp Allergy 2012;42(8):1266-72.

126.Borres MP, Sato S, Ebisawa M. IgE-related examination in food allergy with focus on allergen components. Chem Immunol Allergy 2015;101:68-78.

127.Sampson HA, Ho DG. Relationship between food-specific IgE concentrations and the risk of positive food challenges in children and adolescents. J Allergy Clin Immunol 1997;100(4):444-51.

128.Bousquet J, Anto JM, Bachert C, Bousquet PJ, Colombo P, Crameri R, et al. Factors responsible for differences between asymptomatic subjects and patients presenting an IgE sensitization to allergens. A GA2LEN project. Allergy 2006;61(6):671-80.

129.Komata T, Söderström L, Borres MP, Tachimoto H, Ebisawa M. The predictive relationship of food-specific serum IgE concentrations to challenge outcomes for egg and milk varies by patient age. J Allergy Clin Immunol 2007;119(5):1272-4.

130.Shreffler WG. Microarrayed recombinant allergens for diagnostic testing. J Allergy Clin Immunol 2011;127(4):843-9.

131.Crameri R. The crux with a reliable in vitro and in vivo diagnosis of allergy. Allergy 2013;68(6):693-4.

132.Wood RA. Diagnostic elimination diets and oral food provocation. Chem Immunol Allergy 2015;101:87-95.

95

133.Eigenmann PA, Caubet JC, Zamora SA. Continuing food-avoidance diets after negative food challenges. Pediatr Allergy Immunol 2006;17(8):601-5.

134.Bird JA, Lack G, Perry TT. Clinical management of food allergy. J Allergy Clin Immunol Pract 2015;3(1):1-11.

135.Nowak-Wegrzyn A, Assa'ad AH, Bahna SL, Bock SA, Sicherer SH, Teuber SS. Work Group report: oral food challenge testing. J Allergy Clin Immunol 2009;123(6 Suppl):S365-83.

136.Venter C. Food hypersensitivity: diagnosing and managing food allergies and intolerances. J Allergy (Cairo) 2012;2012:576017.

137.Papadopoulou A, Koletzko S, Heuschkel R, Dias JA, Allen KJ, Murch SH, et al. Management guidelines of eosinophilic esophagitis in childhood. J Pediatr Gastroenterol Nutr 2014;58(1):107-18.

138.Matsumoto K. Non-IgE-related diagnostic methods (LST, patch test). Chem Immunol Allergy 2015;101:79-86.

139.Majamaa H, Laine S, Miettinen A. Eosinophil protein X and eosinophil cationic protein as indicators of intestinal inflammation in infants with atopic eczema and food allergy. Clin Exp Allergy 1999;29(11):1502-6.

140.van Odijk J, Peterson CG, Ahlstedt S, Bengtsson U, Borres MP, Hulthen L, et al. Measurements of eosinophil activation before and after food challenges in adults with food hypersensitivity. Int Arch Allergy Immunol 2006;140(4):334-41.

141.Beser OF, Sancak S, Erkan T, Kutlu T, Cokugras H, Cokugras FC. Can Fecal Calprotectin Level Be Used as a Markers of Inflammation in the Diagnosis and Follow-Up of Cow's Milk Protein Allergy? Allergy Asthma Immunol Res 2014;6(1):33-8.

142.Merras-Salmio L, Kolho KL, Pelkonen AS, Kuitunen M, Mäkelä MJ, Savilahti E. Markers of gut mucosal inflammation and cow's milk specific immunoglobulins in non-IgE cow's milk allergy. Clin Transl Allergy 2014;4(1):2045-7022.

143.Kim JS, Sampson HA. Food allergy: a glimpse into the inner workings of gut immunology. Curr Opin Gastroenterol 2012;28(2):99-103.

144.Turjanmaa K, Darsow U, Niggemann B, Rance F, Vanto T, Werfel T. EAACI/GA2LEN position paper: present status of the atopy patch test. Allergy 2006;61(12):1377-84.

96

145.Niggemann B, Reibel S, Wahn U. The atopy patch test (APT)-- a useful tool for the diagnosis of food allergy in children with atopic dermatitis. Allergy 2000;55(3):281-5.

146.Canani RB, Buongiovanni A, Nocerino R, Cosenza L, Troncone R. Toward a standardized reading of the atopy patch test in children with suspected cow's milk allergy-related gastrointestinal symptoms. Allergy 2011;66(11):1499-500.

147.Nocerino R, Granata V, Di Costanzo M, Pezzella V, Leone L, Passariello A, et al. Atopy patch tests are useful to predict oral tolerance in children with gastrointestinal symptoms related to non-IgE-mediated cow's milk allergy. Allergy 2013;68(2):246-8.

148.Mehl A, Rolinck-Werninghaus C, Staden U, Verstege A, Wahn U, Beyer K, et al. The atopy patch test in the diagnostic workup of suspected food-related symptoms in children. J Allergy Clin Immunol 2006;118(4):923-9.

149.Scala E, Alessandri C, Bernardi ML, Ferrara R, Palazzo P, Pomponi D, et al. Cross-sectional survey on immunoglobulin E reactivity in 23,077 subjects using an allergenic molecule-based microarray detection system. Clin Exp Allergy 2010;40(6):911-21.

150.Burton OT, Logsdon SL, Zhou JS, Medina-Tamayo J, Abdel-Gadir A, Noval Rivas M, et al. Oral immunotherapy induces IgG antibodies that act through FcgammaRIIb to suppress IgE-mediated hypersensitivity. J Allergy Clin Immunol 2014;134(6):1310-7.

151.Santos AF, James LK, Bahnson HT, Shamji MH, Couto-Francisco NC, Islam S, et al. IgG4 inhibits peanut-induced basophil and mast cell activation in peanut-tolerant children sensitized to peanut major allergens. J Allergy Clin Immunol 2015;135(5):1249-56.

152.Mariani A, Marsili M, Nozzi M, Faricelli R, Chiarelli F, Breda L. Serum calprotectin: review of its usefulness and validity in paediatric rheumatic diseases. Clin Exp Rheumatol 2015;33(1):109-14.

153.Decembrino L, De Amici M, Pozzi M, De Silvestri A, Stronati M. Serum Calprotectin: A Potential Biomarker for Neonatal Sepsis. J Immunol Res 2015;147973(10):26.

154.Hoffmann HJ, Santos AF, Mayorga C, Nopp A, Eberlein B, Ferrer M, et al. The clinical utility of basophil activation testing in diagnosis and monitoring of allergic disease. Allergy 2015. E-pub:2015/07/23

155.Patil SU, Shreffler WG. BATting above average: basophil activation testing for peanut allergy. J Allergy Clin Immunol 2014;134(3):653-4.

97

156.Glaumann S, Nopp A, Johansson SG, Borres MP, Nilsson C. Oral peanut challenge identifies an allergy but the peanut allergen threshold sensitivity is not reproducible. PloS One 2013;8(1):e53465.

157.Santos AF, Douiri A, Becares N, Wu SY, Stephens A, Radulovic S, et al. Basophil activation test discriminates between allergy and tolerance in peanut-sensitized children. J Allergy Clin Immunol 2014;134(3):645-52.

158.Nilsson N, Nilsson C, Hedlin G, Johansson SG, Borres MP, Nopp A. Combining analyses of basophil allergen threshold sensitivity, CD-sens, and IgE antibodies to hydrolyzed wheat, omega-5 gliadin and timothy grass enhances the prediction of wheat challenge outcome. Int Arch Allergy Immunol 2013;162(1):50-7.

159.Nopp A, Johansson SG, Ankerst J, Palmqvist M, Öman H. CD-sens and clinical changes during withdrawal of Xolair after 6 years of treatment. Allergy 2007;62(10):1175-81.

160.Sicherer SH, Wood RA, Stablein D, Burks AW, Liu AH, Jones SM, et al. Immunologic features of infants with milk or egg allergy enrolled in an observational study (Consortium of Food Allergy Research) of food allergy. J Allergy Clin Immunol 2010;125(5):1077-83.

161.Metcalfe JR, D'Vaz N, Makrides M, Gold MS, Quinn P, West CE, et al. Elevated IL-5 and IL-13 responses to egg proteins predate the introduction of egg in solid foods in infants with eczema. Clin Exp Allergy 2015;6(10):12608.

162.Krogulska A, Wasowska-Krolikowska K, Polakowska E, Chrul S. Cytokine profile in children with asthma undergoing food challenges. J Investig Allergol Clin Immunol 2009;19(1):43-8.

163.Vocca I, Canani RB, Camarca A, Ruotolo S, Nocerino R, Radano G, et al. Peripheral blood immune response elicited by beta-lactoglobulin in childhood cow's milk allergy. Pediatr Res 2011;70(6):549-54.

164.Dang TD, Tang ML, Koplin JJ, Licciardi PV, Eckert JK, Tan T, et al. Characterization of plasma cytokines in an infant population cohort of challenge-proven food allergy. Allergy 2013;68(10):1233-40.

165.International Standardization Organization. Report No. ISO 4120:2004. Sensory analysis. Methodology. Triangular test. Second edition 2004/06/15

166.Meilgaard MC, Civille, GV, Carr BT. Sensory evaluation techniques. 4th edition: CRC Press Inc: Boca Raton, FL.: 2007.

98

167.Asher MI, Keil U, Anderson HR, Beasley R, Crane J, Martinez F, et al. International Study of Asthma and Allergies in Childhood (ISAAC): rationale and methods. Eur Resp J 1995;8(3):483-91.

168.Bjerg A, Sandström T, Lundbäck B, Rönmark E. Time trends in asthma and wheeze in Swedish children 1996-2006: prevalence and risk factors by sex. Allergy 2010;65(1):48-55.

169.Rönmark E, Jonsson E, Platts-Mills T, Lundbäck B. Different pattern of risk factors for atopic and nonatopic asthma among children--report from the Obstructive Lung Disease in Northern Sweden Study. Allergy 1999;54(9):926-35.

170.Rönmark E, Lundbäck B, Jonsson E, Platts-Mills T. Asthma, type-1 allergy and related conditions in 7- and 8-year-old children in northern Sweden: prevalence rates and risk factor pattern. Resp Med 1998;92(2):316-24.

171.Dreborg S. The skin prick test in the diagnosis of atopic allergy. J Am Acad Dermatol 1989;21(4 Pt 2):820-1.

172.Calle-Alonso F, Pérez CJ. A Statistical Agreement-Based Approach for Difference Testing. J Sensory Studies 2013;28(5):358-69.

173.Greenberg R. Medical epidemiology. 4C edition: New York: Lange Medical Books/McGraw-Hill 2005.

174.Howe CJ, Cain LE, Hogan JW. Are all biases missing data problems? Curr Epidemiol Rep 2015;2(3):162-71.

175.Mortz CG, Andersen KE, Bindslev-Jensen C. Recall Bias in Childhood Atopic Diseases Among Adults in The Odense Adolescence Cohort Study. Acta Derm Venereol 2015;28(10):00015555-2128.

176.Hedman L, Bjerg A, Perzanowski M, Rönmark E. Good agreement between parental and self-completed questionnaires about allergic diseases and environmental factors in teenagers. J Clin Epidemiol 2010;63(7):783-9.

177.Pagano M, Gauvreau K. Principles of biostatistics. 2th edition: Pacific Grove: Duxbury; 2000.

178.Boyce JA, Assa'ad A, Burks AW, Jones SM, Sampson HA, Wood RA, et al. Guidelines for the Diagnosis and Management of Food Allergy in the United States: Summary of the NIAID-Sponsored Expert Panel Report. J Allergy Clin Immunol 2010;126(6):1105-18.

99

179.Rönmark E, Perzanowski M, Platts-Mills T, Lundbäck B. Four-year incidence of allergic sensitization among schoolchildren in a community where allergy to cat and dog dominates sensitization: report from the Obstructive Lung Disease in Northern Sweden Study Group. J Allergy Clin Immunol 2003;112(4):747-54.

180.Bjerg A, Winberg A, Berthold M, Mattsson L, Borres MP, Rönmark E. A population-based study of animal component sensitization, asthma, and rhinitis in schoolchildren. Pediatr Allergy Immunol 2015;26(6):557-63.

181.Kaya A, Erkocoglu M, Civelek E, Cakir B, Kocabas CN. Prevalence of confirmed IgE-mediated food allergy among adolescents in Turkey. Pediatr Allergy Immunol 2013;24(5):456-62.

182.Persson K, Öhlund I, Nordström L, Winberg A, Rönmark E, West CE. Vitamin D deficiency at the Arctic Circle - a study in food-allergic adolescents and controls. Acta Paediatr. 2013;102(6):644-9.

183.Meyer R, De Koker C, Dziubak R, Venter C, Dominguez-Ortega G, Cutts R, et al. Malnutrition in children with food allergies in the UK. J Hum Nutr Diet 2013. E-pub: 2013/08/14

184.Mehta H, Ramesh M, Feuille E, Groetch M, Wang J. Growth comparison in children with and without food allergies in 2 different demographic populations. J Pediatr 2014;165(4):842-8.

185.Vlieg-Boerstra BJ, Bijleveld CM, van der Heide S, Beusekamp BJ, Wolt-Plompen SA, Kukler J, et al. Development and validation of challenge materials for double-blind, placebo-controlled food challenges in children. J Allergy Clin Immunol 2004;113(2):341-6.

186.Nowak-Wegrzyn A, Fiocchi A. Rare, medium, or well done? The effect of heating and food matrix on food protein allergenicity. Curr Opin Allergy Clin Immunol 2009;9(3):234-7.

187.Vlieg-Boerstra BJ, van der Heide S, Bijleveld CM, Kukler J, Duiverman EJ, Dubois AE. Placebo reactions in double-blind, placebo-controlled food challenges in children. Allergy 2007;62(8):905-12.

188.West CE, Hernell O, Andersson Y, Sjöstedt M, Hammarström ML. Probiotic effects on T-cell maturation in infants during weaning. Clin Exp Allergy 2012;42(4):540-9.

189.Kolho KL, Turner D, Veereman-Wauters G, Sladek M, de Ridder L, Shaoul R, et al. Rapid test for fecal calprotectin levels in children with Crohn disease. J Pediatr Gastroenterol Nutr 2012;55(4):436-9.

100

190.Robbins KA, Wood RA, Keet CA. Milk allergy is associated with decreased growth in US children. J Allergy Clin Immunol 2014. Epub 2014/10/15.

191.Nachshon L, Goldberg MR, Schwartz N, Sinai T, Amitzur-Levy R, Elizur A, et al. Decreased bone mineral density in young adult IgE-mediated cow's milk-allergic patients. J Allergy Clin Immunol 2014;134(5):1108-13.

192.Meyer R, Venter C, Fox AT, Shah N. Practical dietary management of protein energy malnutrition in young children with cow's milk protein allergy. Pediatr Allergy Immunol 2012;23(4):307-14.

193.Nowak-Wegrzyn A, Groetch M. Nutritional aspects and diets in food allergy. Chem Immunol Allergy 2015;101:209-20.

194.Kokkonen J, Haapalahti M, Tikkanen S, Karttunen R, Savilahti E. Gastrointestinal complaints and diagnosis in children: a population-based study. Acta Paediatr 2004;93(7):880-6.

195.Peters RL, Gurrin LC, Allen KJ. The predictive value of skin prick testing for challenge-proven food allergy: a systematic review. Pediatr Allergy Immunol 2012;23(4):347-52.

196.Kemp AS, Allen CW, Campbell DE. Parental perceptions in egg allergy: does egg challenge make a difference? Pediatr Allergy Immunol 2009;20(7):648-53.

197.Sutas Y, Kekki OM, Isolauri E. Late onset reactions to oral food challenge are linked to low serum interleukin-10 concentrations in patients with atopic dermatitis and food allergy. Clin Exp Allergy 2000;30(8):1121-8.

198.Lotoski LC, Simons FE, Chooniedass R, Liem J, Ostopowich I, Becker AB, et al. Are plasma IL-10 levels a useful marker of human clinical tolerance in peanut allergy? PLoS One 2010;5(6):0011192.

199.Krogulska A, Borowiec M, Polakowska E, Dynowski J, Mlynarski W, Wasowska-Krolikowska K. FOXP3, IL-10, and TGF-beta genes expression in children with IgE-dependent food allergy. J Clin Immunol 2011;31(2):205-15.

200.Savilahti EM, Karinen S, Salo HM, Klemetti P, Saarinen KM, Klemola T, et al. Combined T regulatory cell and Th2 expression profile identifies children with cow's milk allergy. Clin Immunol 2010;136(1):16-20.

101

Supplement

Questions from structured interviews considered in Paper III and/or Paper IV.

1. Was your child breastfed?

□ No

□ Yes

If yes, for how long?

□ < 3 months □ 3-6 months

□ 6-12 months □ >12 months

2. Did your child suffer from any symptoms during breastfeeding?

□ No

□ Yes

If yes, what kind of symptom/s?

□ Colic □ Gastrointestinal

□ Eczema □ Urticaria

□ Respiratory □ other: _____________

3. At what age was your child introduced to taste portions?

□ 3-6 months □ 6-12 months

□ >12 months

4. Did your child suffer from any symptoms during introduction of

taste portions?

□ No

□ Yes

If yes, what kind of symptom/s?

□ Colic □ Gastrointestinal

□ Eczema □ Urticaria

□ Respiratory □ Other: ______________

5. Has your child had a growth deviation?

□ No

□ Yes

If yes, which type of growth deviation?

□ Length □ Weight

□ Head circumference

102

6. Parental allergy/atopy related conditions?

□ No

□ Yes, mother

If yes, what type of allergy/atopy related condition in mother?

□ Asthma

□ Rhino-conjunctivitis

□ Eczema

□ Gastrointestinal

□ Allergy to airborne allergen/s, which?

________________________________

□ Food hypersensitivity;

□ During childhood

□ In adulthood

To which food/foods? -

______________________

□ Yes, father

If yes, what type of allergy/atopy related condition in father?

□ Asthma

□ Rhino-conjunctivitis

□ Eczema

□ Gastrointestinal

□ Allergy to airborne allergen/s, which?-

________________________________

□ Food hypersensitivity;

□ During childhood

□ In adulthood

To which food/foods? -

______________________

7. Other parental (mother or father) diseases/conditions?

a. Celiac disease

□ Yes □ No

b. Diabetes Mellitus type 1

□ Yes □ No

c. Inflammatory bowel disease

□ Yes □ No

d. Thyroid disease

□ Yes □ No

e. Gastritis/Esophagitis

□ Yes □ No

f. Irritable bowel disease

□ Yes □ No

103

8. Atopy related diseases/conditions in child (except for food

hypersensitivity)

a. Asthma

□ Yes, now □ Yes, before □ No, never

b. Rhino-conjunctivitis

□ Yes, now □ Yes, before □ No, never

c. Eczema

□ Yes, now □ Yes, before □ No, never

d. Allergy

□ Yes, now □ Yes, before □ No, never

If yes, what kind of allergy?

□ Birch pollen

□ Grass pollen

□ Ragweed

□ Furred animals

□ Bee/Wasp

□ Drugs, type: ______________________

9. Other diseases/conditions in child?

a. Celiac disease

□ Yes □ No

b. Diabetes Mellitus type 1

□ Yes □ No

c. Inflammatory bowel disease

□ Yes □ No

d. Thyroid disease

□ Yes □ No

e. Gastritis/Esophagitis

□ Yes □ No

f. Irritable bowel disease

□ Yes □ No

10. Has the child ever had anaphylaxis caused by food?

□ No

□ Yes

If yes, by what food trigger/s?

__________________________________________

At what age/s?

______________________________________

104

Questions on symptoms caused by the specific foods: milk, egg,

cod and wheat.

If the child reported hypersensitivity to more than one of these foods, the

questions were asked separately for the different foods.

11. Current symptoms caused by intake of milk/egg/cod/wheat:

□ Airways: □ Rhinitis □ Asthma □ Laryngospasm

□ Heart/circulation (decreased blood pressure/ circulatory failure)

□ Eyes (conjunctivitis/kerato-conjunctivitis)

□ Oral: □ Itching □ Aphtae □ Lip swelling

□ Itching/bloated feeling of the throat

□ Skin: □ Eczema □ Urticaria □ Angioedema

□ Gastrointestinal: □ Vomiting □ Diarrhea □ Constipation

□ Flatulence □ Stomach ache

□ Affected growth

□ Other/s:

_______________________________________________

_______________________________________________

_______________________________________________

12. Symptoms caused by intake of milk/egg/cod/wheat at symptom

onset:

□ Airways: □ Rhinitis □ Asthma □ Laryngospasm

□ Heart/circulation (decreased blood pressure/ circulatory failure)

□ Eyes (conjunctivitis/kerato-conjunctivitis)

□ Oral: □ Itching □ Aphtae □ Lip swelling

□ Itching/bloated feeling of the throat

□ Skin: □ Eczema □ Urticaria □ Angioedema

□ Gastrointestinal: □ Vomiting □ Diarrhea □ Constipation

□ Flatulence □ Stomach ache

□ Affected growth

□ Other/s:

_______________________________________________

_______________________________________________

_______________________________________________

105

13. Has the child had symptoms upon

a. Airborne exposure of the culprit food?

□ Yes □ No

b. Skin exposure of the culprit food?

□ Yes □ No

14. Are co-factors necessary for the development of food

hypersensitivity symptoms?

□ No

□ Yes

If yes, what kind of co-factors?

□ Exercise □ Infection

□ Pollen season □ Coldness

□ Drugs

□ Other: __________________________

15. Childs age at symptom onset?

□ 0-2 years □ 2-5 years

□ 5-8 years □ 9-12 years

□ >12 years

16. Time to symptom onset after intake of the culprit food?

□ 0-15 minutes □ 15-60 minutes

□ 1-3 hours □ 4-24 hours

□ 1-3 days

17. Symptom duration after intake of the culprit food?

□ 0-15 minutes □ 15-60 minutes

□ 1-3 hours □ 4-24 hours

□ 1-3 days □ > 3 days

18. Current elimination of the culprit food from child’s diet?

□ Not at all

□ Partially □ Completely

106

19. If hypersensitivity to milk: does lactose-free products trigger

symptoms?

□ No

□ Yes

If yes, at what dose?

__________________________________

20. If hypersensitivity to milk: What replacement products does the

child eat/drink?

□ Lactose free/reduced products

□ < 1 dl/day □ 1-3 dl/day □ >1 dl/day

□ Soy products

□ < 1 dl/day □ 1-3 dl/day □ >1 dl/day

□ Oat products

□ < 1 dl/day □ 1-3 dl/day □ >1 dl/day

□ Rice products

□ < 1 dl/day □ 1-3 dl/day □ >1 dl/day

□ Calcium supplement

□ Other:

_________________________________________

21. If hypersensitivity to fish, what kind of fish causes symptoms?

□ Cod fish □ Salmon

□ Shell fish

22. Physician diagnosis of celiac disease?

□ No

□ Yes

If yes, at what age?

__________________________________________

23. If hypersensitivity to wheat: does other grains cause symptoms?

□ No

□ Yes

If yes, what kind of grains?

□ Barley □ Rye □ Oats

107

24. At what age was the culprit food introduced in the child’s diet?

□ < 3 months □ Through breastmilk □ Own diet

□ 3-6 months □ Through breastmilk □ Own diet

□ 6-12 months □ Through breastmilk □ Own diet

□ > 12 months □ Through breastmilk □ Own diet

25. Amount and type of trigger food necessary to cause symptoms:

a. Milk □ Baked milk:

□ Treasure amount □ Table spoon

□ < 1 dl □ >1 dl

□ Raw (pasteurized) milk:

□ Treasure amount □ Table spoon

□ < 1 dl □ >1 dl

b. Egg □ Baked egg:

□ Treasure amount

□ Tablespoon of muffin

□ Whole muffin

□ Pancake

□ Boiled/fried egg

□ Raw egg

□ Treasure amount

□ Tablespoon of sponge

c. Cod □ Cooked cod:

□ Treasure amount □ Tea spoon

□ Table spoon □ Portion size

□ Raw cod (sushi, pickled herring):

□ Treasure amount □ Tea spoon

□ Table spoon □ Portion size

d. Wheat □ Baked wheat:

□ Treasure amount

□ Tea spoon of bread

□ Table spoon of bread

□ Portion size

□ Repeated exposure

□ Raw wheat

108

26. When was the child last exposed to the culprit food?

□ < 3 months □ 3-6 months

□ 6-12 months □ 1-2 years

□ >2 years

27. How long ago did the child last have a reaction due to intake of the

culprit food?

□ < 3 months □ 3-6 months

□ 6-12 months □ 1-2 years

□ >2 years

28. Has the child been diagnosed with hypersensitivity to the culprit

food through a previous health care contact?

□ No

□ Yes

If yes, by whom?

□ By a physician

□ By a child health care nurse

□ Other:__________________________

29. Has the child performed any investigations/diagnostic test before

the food hypersensitivity diagnosis?

□ No

□ Yes

If yes, what kind of investigation/test?

□ Clinical history

□ Positive □ Negative

□ Specific IgE/Skin prick test

□ Positive □ Negative

□ Oral challenge

□ Positive □ Negative

□ Lactose challenge

□ Positive □ Negative

□ Gene test (lactase down-regulating gene)

□ Positive □ Negative

□ Other: ___________________________

30. Has the child ever been referred to a dietician for nutritional advice

due to the food hypersensitivity?

□ No

□ Yes