Hugh A' Sampson, MD

1
Hugh A. Sampson, MD
2
Relative Clinical Usefulness of In Vitro Versus
In Vivo Measurements of Food Allergy

• Hugh A. Sampson, M.D.
• Professor of Pediatric Immunobiology
• Jaffe Food Allergy Institute
• Mount Sinai School of Medicine
• New York, NY

3
Case Study
A 16 y/o boy has had 2 anaphylactic reactions
while eating pizza at a local pizzeria. He has
eaten there before regularly ingests milk, egg,
wheat, tomato including spaghetti with no
problem. The best first step would be to

• PST to milk, wheat, tomato, garlic fish
• RAST to milk, wheat, tomato, garlic fish
• Challenge the boy to garlic
• Have the boy run a mile challenge him to wheat

4
Case Study Cont
PSTs reveal a 3 mm wheal to milk all other skin
tests are negative. The boy has had milk with
every meal without symptoms. The best next step
would be to

• A. RAST to milk, wheat, tomato, garlic peanut
• B. Tell boy to avoid that pizzeria but he can
  eat other pizza without concern
• C. Obtain exact pizza perform prick prick to
  various components in pizza
• D. Review the history for other possible
  allergens

5
DIAGNOSING FOOD ALLERGY

• DBPCFC is the GOLD STANDARD
• History lt 30 - 40 of histories confirmed
• Skin Tests or RASTs lt30 - 40 confirmed
• Elimination Diets 0 - 40 of patients
  responding to elimination diet confirmed
• Open Challenge 20 - 60 confirmed

6
Prick Skin Testing
7
Paradigm Shift

• Tests viewed as positive or negative
• e.g., a 3 mm wheal is a positive test
• Tests viewed as probability of reaction

100
80
60
Probability of a reaction ()
40
20
0
Food-specific IgE Antibody Concentration or
Skin Test Wheal Size
8
Receiver Operator Curve (ROC)
9
Predictive Value of PSTs
Comparison of PST results outcome of oral
milk challenges - 120 challenges - 37
positive
Subgroup lt 2 yrs
10
PST and Clinical Reactivity
74 egg allergic children with egg-specific IgE lt
2.5 kU/L challenged to egg 29 passed
challenge, 45 failed
Knight et al. JACI 2006117842-7
11
Predictive Value of PSTs
Interpretation must take into account clinical
history
Knight et al. JACI 2006117842-7
12
Features Affecting Skin Tests

• Extract non-standardized lot-to-lot variation
• Device used for prick/puncture
• Operator pressure applied during
  application precision of measurement
• Location of skin test back gt volar aspect of
  arm mid- upper- back gt lower back proximal
  forearm gt distal forearm 3 cm/5 cm
• Means of measuring wheal size
• No added value for intradermal testing
  Bock et al. JACI 1978 8559-64

13
Atopy Patch Tests (APT)

• Reagents methods not standardized - outcome
  is very operator dependent
• Diagnostic value in suspected food allergy in
  children Mehl et al. JACI 2006 118923-29
  - 873 oral challenges compared to PST, APT
  and food-specific IgE milk, egg, wheat /or
  soy - APT is least sensitive (23 - 41), but
  most specific (86 - 95) - Using a 95
  predictive value, the combination of APT with
  PST or Ag-IgE eliminates the need for only 7 -
  16 of oral food challenges

14
Probability of Reacting to Egg
100
90
Egg white
80
70
60
50
Probability
Retrospective study n 300
40
30
Prospective study n 100
20
10
Logit model using log(kU
/L)
A
0
0.35
3.5
17.5
50
100
0.7
IgE Antibody Concentration (kUA/L)
15
95 Predictive Decision Levels

• Allergen Decision Pt PPV Sens.
  Spec.
• (kUA/L)____________________
  _____
  
• Egg 7 98 61 98
• (lt 2 yrs of age) 2
  95
• Milk 15 95 57 94
• (lt 1yr of age) 5
  95
• Peanut 14 100 57 100
• Soy 30 73 44 94
• Wheat 26 74 61 92
• Tree nuts 15 95 ---- ----

Boyano MT, et al. Clin Exp Allergy 2001
311464-9. Garcia-Ara C, et al. JACI 2001
107185-90. Clark AT, Ewan P. Clin Exp
Allergy 2003 331041-45.
Sampson JACI 2001 107891-96.
16
Consistency of Values Over Time
17 years
100
10
ImmunoCAP kUA/l, 2004
ImmunoCAP kUA/l, 1995
1
0.1
0.1
1
10
100
Pharmacia CAP System kUA/l, 1987
Paganelli R et al.. Allergy 1998 53 763-768
17
Not All Quantization is the Same
Some tests extrapolate kU/L from PRIST assay
- less bound Ag leads to
flattening of std curve at higher
values
18
Case Atopic Dermatitis
2 month old male with irritability,
eczematous rash - exclusively breast fed -
regular maternal diet - no apparent
correlation with mothers diet - mother
has mild AR
19
Evaluation of Infant with AD
Given the history, you should first -

• A. Have mother eliminate egg milk from her
  diet
• B. Discontinue breastfeeding start
  hypoallergenic formula
• C. PST to egg, milk, peanut, soy wheat
• D. Obtain food-specific IgE levels to egg,
  milk, peanut, soy and wheat

20
Results of Studies

• Prick Skin Tests - milk 8/12 - egg 2/4 -
  peanut 0/2 - soy negative - wheat 4/15 -
  histamine 5/10 - saline 0/2

• Food-IgE kU/L - milk 6.1 - egg 0.8 -
  peanut lt0.35 - soy lt0.35 - wheat 4.1

• Your next step
• A. Eliminate egg, milk wheat from mother
• B. Eliminate milk wheat from mother
• Eliminate milk from mother
• D. Start hypoallergenic formula

21
AD and Milk Allergy
Milk PST 8/12 mm Milk-IgE 6.1 kUA/L
2 weeks after initiating a milk-free diet
1/3 of infants with atopic dermatitis have
food allergies
22
Cross-reactivity in Testing
23
The Protein Families Contained in the Pfam
Database

• Pfam database contains 8,296 protein families.
• (Database release 20.0, May. 2006)
• 22 of these 8,296 (0.27) Pfam families contain
  all known plant food allergens - 4 families of
  these 22 account for more than 65 of all
  known plant food allergens.
• The most frequently decribed animal food
  allergens are found in 11 of the 8,296 (0.13)
  Pfam families
• Jenkins J. et al., J JACI
  2005115164

24
Plant Food Allergens
Total
No. of Allergens
Short Pfam Name
Pfam Acc. No.
30
36
Prolamin
PF00234
44
18
Cupin
PF00190
57
16
Profilin
PF00235
67
12
PR-10/Bet v 1 homologues
PF00407
71
5
Papain-like
PF00112
75
5
Chitin-binding
PF00187
79
4
Thaumatin-like
PF00314
82
4
Chitinase class I
PF00182
25
Food Proteins Cupin Family
26
Allergen Epitope Structure
Sequential Conformational
27
Ovomucoid IgE Epitopes
Transient egg allergy
Persistent egg allergy
IgE-binding Sites
N L D
N L D
N L D
Sampson Cooke. J Immunol 1997 1592026-2032.
28
IgE-Epitope Recognition as a Diagnostic Marker
Ovomucoid
Conformational Epitope
Sampson Cooke. J Immunol 1997 1592026-2032.
29
Epitope Analysis with Peptide Microarray

• 210 overlapping peptides covering Ara h1, Ara h2
  and Ara h3 were spotted on glass slides

Sum log Intensity

• Immunolabeled with sera from 90 pts

Shreffler et al. JACI 2004 113776-782
30
Predicting Reactivity Severity

• Reactivity binding to specific epitopes is
  associated with clinical reactivity and
  persistence of food allergy
• Severity - History patients with asthma or
  a previous severe reaction are more
  likely to experience a severe reaction -
  PSTs and Food-specific IgE levels - No
  correlation with PST size or serum IgE level
  - Diversity of epitope recognition

31
Epitope Diversity Reactivity
Pooled
Pt 1
Greater epitope diversity relatively more
peanut-specific binding sites present on
mast cells ? greater releasibility
Pt 2
Pt 3
Pt 4
Pt 5
Greater epitope diversity more severe
reactions
Pt 6
Pt 7
Shreffler et al. JACI 2004 113776-782
32
Clinical Reactivity to Peanut

• From 77 peanut-allergic subjects, reviewed the
  clinical reactions of 38 patients - clinical
  score 0 40 determined for each
  reaction Sampson HA. Pediatrics 2003
  1111601-1608.
• Median age 7 years
• Reactions
• - 14 cutaneous only, 24 multi-system
• 18 controls with no history of reaction
• Peanut-IgE 1.97 to gt100 kU/L (median gt 100)

Shreffler et al. JACI 2004 113776-782
33
Epitope Diversity Associated with Severity of
Reaction
38 peanut allergic patients scored for severity
of reaction
Shreffler et al. JACI 2004 113776-782
34
Clinically Reactive Patients
Epitope Number
Peanut-specific IgE
Shreffler et al. JACI 2004 113776-782
35
Natural History

• Dependent on food immunopathogenesis
• 80 CM, egg, wheat, soy allergy remit by 4 5
  yrs
• Declining/low levels of specific-IgE predictive
• IgE binding to conformational epitopes predictive
• Allergy to peanut, nuts, seafood typically
  persist - 15 - 20 of young children outgrow
  reactivity
• Non-IgE-associated GI allergy
• Infant forms resolve 1-3 years
• Toddler/adult forms more persistent

36
Outgrowing Food Allergy

• Weigh history, PN-specific IgE and PST results
• Predicting passage of peanut challenge HX lt
  0.35 0.36 - lt 2 2 4.9 gt5 kU/L
  76 44 40 0 ? 88
  71 33 77 Perry et al JACI
  2004 114144-149.
• PST gt 8mm and PN-IgE lt 2 kU/L lt 40 pass
• More likely to pass if PN-IgE never gt 5 kU/L, PST
  lt 3 mm no reaction in past 3 yrs, isolated
  allergy
• No correlation with severity of 1st reaction

37
Conclusions

• PSTs and allergen-specific IgE can be useful in
  the diagnosis management of IgE- mediated
  food allergy
• When interpreting results, must consider several
  factors - predictive value of test result
  - strength of history - age of patient
  potential cross-reactivities
• When considering OFC, consider benefit of
  adding food probability of passing