Post on 14-Mar-2022
METHODSBACKGROUND
Systematic Collection of Biopsies and Quantification of Eosinophils in Multiple High-Power Fields is Required for Diagnosis of Eosinophilic Gastritis and/or Duodenitis
Kevin O. Turner DO1, Margaret H. Collins MD2, Marjorie M. Walker BMBS3, Maria A. Pletneva MD PhD4, Cory M. Mekelburg5, Amol P. Kamboj MD5, Henrik S. Rasmussen MD PhD5, Nicholas J. Talley MD PhD3, Robert M. Genta MD6
1UT Southwestern Medical Center, Dallas, TX; 2Cincinnati Children's Hospital Medical Center and University of Cincinnati College of Medicine, Cincinnati, Ohio; 3University of Newcastle, New South Wales, Australia; 4University of Utah, Salt Lake City, UT; 5Allakos Inc., Redwood City, CA; 6Baylor College of Medicine, Houston, TX
Figure 3. Biopsy and Histopathology Protocol and Diagnostic Criteria for EG and/or EoD Used in ENIGMA and Prevalence Studies
CONCLUSIONS/DISCUSSION• A systematic histopathology protocol with evaluation of gastric and duodenal
eosinophilia in patients with chronic, moderate–severe GI symptoms, in 2 prospective studies, revealed that about a third of patients without previous diagnoses of EG and/or EoD met histologic criteria for these disorders
• Results of Sydney and Marsh scoring suggest that low power evaluation of GI biopsies is not sufficient to detect EG and/or EoD
• Given the high diagnostic yield, a standardized biopsy and histopathology protocol should be used to evaluate patients for EG and/or EoD, so that they can receive an accurate diagnosis
Table 1. Patient Demographics
• Pathologic accumulation and over-activation of eosinophils and mast cells are implicated in chronic inflammatory diseases in the gastrointestinal (GI) tract, including eosinophilic esophagitis (EoE), gastritis (EG), duodenitis (EoD), and colitis—collectively termed eosinophilic gastrointestinal diseases (EGIDs)1,2
• Patients with EGIDs have decreased quality of life due to chronic debilitating and often nonspecific symptoms such as dysphagia, abdominal pain, abdominal cramping, bloating, early satiety, loss of appetite, nausea, vomiting, and diarrhea3
• ENIGMA was a randomized, controlled, phase 2 trial of adult patients with EG and/or EoD that established the therapeutic potential of lirentelimab, an investigational medicine, which is a monoclonal antibody against Siglec-8 that depletes eosinophils and inhibits mast cell activity4*
• Patients enrolled in ENIGMA were first screened for moderate–severe GI symptoms using a daily patient-reported outcome (PRO) questionnaire
• Patients who met the symptom criteria underwent esophagogastroduodenoscopy (EGD) with biopsy and histopathologic evaluation to confirm diagnoses of EG and/or EoD (≥30 eosinophils per high-power field [eos/hpf] in ≥5 hpfs in gastric biopsies and/or in ≥3 hpfs in duodenal biopsies)
• Among patients screened in ENIGMA, 45% had no previous diagnoses of EG and/or EoD; 29% of these patients were found to have EG and/or EoD
Min. of 5 hpfsevaluated per biopsy Systematic examination
and counting of eosinophils
POSITIVE GASTRIC BIOPSY
POSITIVE DUODENAL
BIOPSY
Min. of 12 biopsies collected per subject
during EGD4 gastric antrum4 gastric corpus
4 duodenum
Plus additional biopsies from areas of interest
BIOPSY PROTOCOL
HISTOPATHOLOGY PROTOCOL
≥30 eos in a hpf
POSITIVE HPF
≥30 eos/hpf in ≥5 hpfs in gastric
biopsiesand/or in
≥3 hpfs in duodenal biopsies
The requisite number of hpfs with ≥30 eos could be achieved within a single biopsy specimen or aggregated across multiple biopsy specimens
EG and/or EoDDIAGNOSTIC CRITERIA
Subjects with chronic, moderate to
severe GI symptoms
≥30 eos in ≥5 hpfs
≥30 eos in ≥3 hpfs
HISTOLOGIC FINDING DEFINITIONS
Figure 7. Detection Rate of EG and/or EoD Across ENIGMA and Prevalence Studies
RESULTS
1 Non-overlapping hpfs could be, but are not required to be, adjacent to the first hpf, depending on the distribution of eosinophils in the specimen2 If the size of a specimen is insufficient to evaluate 5 independent fields, eos are to be counted in as many non-overlapping fields as availableNote: a separate endoscopy study of healthy volunteers (controls) was conducted for comparison
Patients with moderate-severeEG and/or EoD symptoms
Evaluate at low-power magnification(40X and 100X)
Able to rule out other conditions NOT relevant to eos?
Diagnose for other conditions
Yes
No
Survey for areas with highest eos density at medium-power magnification (200X)
Count eosinophils per representative high-power field (hpf, 0.237 mm2) (400X)
EG and/or EoD symptoms include abdominal pain/cramping, loss of appetite, early satiety, bloating, etc.
Evaluate for proper orientation and for the presence of lesions or conditions: e.g., H. pylori infection, celiac disease, neoplasia
Survey all levels to identify areas with the highest eos density
• Select first hpf from the highest density area and remaining from non-overlapping areas1
• Count a minimum of 5 hpfs, if possible2
• Count with a systematic, consistent approach
Figure 4. Histopathologic Evaluation Process: Steps for EG and/or EoD
Figure 5. Ideal Biopsy Specimen and Countable Eosinophils
Figure 1. Pathogenesis of EGIDs
Stomach
Duodenum
Biopsy Protocol
• 4 biopsies from the duodenum, 2 each from the descending and horizontal parts
• GASTRIC ANTRUM: 4 biopsies (2-5 cm proximal to the pylorus)
• GASTRIC CORPUS: 4 biopsies(2 from the proximal lesser curvature and 2 from the greater curvature)
Reference: (1) Caldwell JM, et al. J Allergy Clin Immunol. 2014.; (2) Youngblood BA, et al. Gastroenterology. 2019.; (3) Chehade M, et al. JACI in Practice 2020.; (4) Dellon ES, et al. NEJM. 2020.*Lirentelimab is an investigational medicine, its efficacy and safety profile have not been established, and it has not been approved by the FDA
Figure 2. New Diagnoses of EG and/or EoD in ENIGMA
• This high discovery rate of EG and/or EoD, along with other studies reporting underdiagnosis of EG and/or EoD, prompted further evaluation of the screening protocol
• We therefore conducted a prospective study of the prevalence of EG and/or EoD in patients with chronic unexplained gastrointestinal symptoms
• We used a systematic histopathology protocol in ENIGMA and this prevalence study to determine the discovery rate of EG and/or EoD
Figure 6. Three Systematic Approaches to Counting Eosinophils
Lawnmower
Scan up 1 column and down the next repeating across the field
aModerate–severe symptoms, defined as an average daily symptom score of ≥3 (scale 0-10) over 7 days for abdominal pain, diarrhea, and/or nausea on a PRO questionnaire for ≥2 weeksbModerate–severe symptoms, defined as an average daily symptom score of ≥3 (scale 0-10) for at least 2 of 3 weeks for abdominal pain, abdominal cramping, nausea, vomiting, diarrhea, bloating, or early satiety on a PRO questionnaire and average Total Symptom Score (TSS) ≥3 [scale 0-80]
38 (67%) met histologic criteria for
EG±EoD(≥30 eos/hpf in 5 hpf
in stomach)
50 (88%) met histologic criteria
for EoD±EG(≥30 eos/hpf in 3 hpf in duodenum)
7 (47%) met histologic criteria for
EG±EoD(≥30 eos/hpf in 5 hpf
in stomach)
12 (80%) met histologic criteria
for EoD±EG(≥30 eos/hpf in 3 hpf in duodenum)
59 (33%) met histologic criteria for
EG±EoD(≥30 eos/hpf in 5 hpf
in stomach)
165 (91%) met histologic criteria
for EoD±EG(≥30 eos/hpf in 3 hpf in duodenum)
62 patients screened who hadhistory of EG and/or EoD
51 patients screened who hadno prior history of EG and/or EoD
556 patients screened who hadno prior history of EG and/or EoD
62 met symptom criteriaa and underwent EGD with gastric and
duodenal biopsies
26 met symptom criteriaa and underwent EGD with gastric and
duodenal biopsies
405 met symptom criteriab and underwent EGD with gastric and
duodenal biopsies
57 met histologic criteria forEG and/or EoD
15 met histologic criteria forEG and/or EoD
181 met histologic criteria forEG and/or EoD
7 (12%) subjects EG only
19 (33%) subjectsEoD only
31 (54%) subjects EG+EoD
(concurrent)
3 (20%) subjects EG only
8 (53%) subjectsEoD only
4 (27%) subjects EG+EoD
(concurrent)
122 (67%)
subjectsEoD only
43 (24%) subjects EG+EoD
(concurrent)
16 (9%) subjects EG only
113 patients screened in ENIGMA 556 patients screened forPrevalence Study
aHistory of asthma, allergic rhinitis, atopic dermatitis, and/or food allergybIrritable bowel syndrome, GERD, chronic gastritis/duodenitis, or functional dyspepsia
Quadrant
Scan the field in a conventional order, such as from left to right
Spiral
Scan the field in a spiral fashion, either inwards or outwards
Ideal Specimen Countable Eosinophils
An eosinophil was considered countable when it had 1 of the following: intact with a bilobed nucleus, fragmented with a partial nucleus, or a discrete cluster of eosinophil granules at least in part limited by a membrane, even if there is no clearly discernable nucleus
A biopsy that is oriented on edge and includes surface epithelium, mucosa, and muscularis mucosae of stomach and duodenum tissue
Gastric Antrum Gastric Corpus
Duodenum
Figure 9. TSS and Mean Eosinophil Counts in Patients vs Controls
Figure represents patients combined from ENIGMA and Prevalence studyaPatients and controls used the same patient-reported-outcome questionnaire and underwent identical biopsy protocols. Histologic evaluation for both groups were performed by the same central pathologists
51% (253/493) of patients and 6% (2/33) of controlsa met histologic criteria for EG and/or EoD (odds ratio, 16.34; 95% CI, 3.9–69.0; P=0.0001)
0
30
60
90
120
150
180
210
300
330✱✱✱✱
✱✱✱✱
0
20
40
60
80
✱✱✱✱
✱✱✱✱
✱✱✱✱
TSS
0
30
60
90
120
150
180
210✱✱✱✱
✱✱✱✱
Duodenum
Mea
n co
unts
/ 5
hpf
Stomach
Mea
n co
unts
/ 3
hpf
TSS
Mean Tissue Eosinophil Counts
Controls (n=33) EoD (n=149)EG+EoD (n=78)EG (n=26)
Intraepithelial Lymphocytosis
100%88%
64%48%
100% 90% 100% 95%
64%42%
7%33%
29%
5%
27%
29%
4% 3%19%
5% 2% 9%20%
1% 4% 1… 8%
0%
25%
50%
75%
100%
Perc
ent o
f Sub
ject
s
Gastric Features(Sydney System)
Active Inflammation
Chronic Inflammation
Intestinal Metaplasia Atrophy
Reactive Gastropathy
100% 93% 100% 99%
6%1%
Villus Architecture
Score=0 Score=1 Score=2 Score=3
Duodenal Features(Marsh Classification)
Figure 8. Sydney and Marsh Scores in Subjects with EG and/or EoD vs Controls
Acknowledgements: We thank the patients who participated in this study, investigators, and study staff. This study was funded by Allakos Inc.. and in part by Division of Intramural Research, NIAID, NIH
51 patients without history of EG and/or EoD entered ENIGMA screening
51% (26/51) met symptom criteria for endoscopy and biopsy
58% (15/26) had EG and/or EoD
• 29% (15/51) of the patients received a new diagnosis of EG and/or EoD
• Most patients without a previous diagnosis of EG and/or EoD came from general GI practices
• These patients had histories of chronic unexplained GI symptoms or diagnoses of functional disorders
Patient Characteristics
ENIGMA+PrevalenceHealthy Controls
N=33EG w/o EoD
N=26EG+EoD
N=78EoD w/o EG
N=149Mean age, years (range) 42 (18-72) 46 (18-76) 44 (19-78) 34 (18-51)Female sex, n (%) 21 (81%) 48 (62%) 106 (71%) 39%White, n (%) 24 (92%) 66 (85%) 130 (87%) 100%Weight, mean (range), kg 80 (47-136) 88 (50-180) 83 (45-138) 79 (46-113)Total Symptom Score (TSS) at baseline, mean ±SD 38 ±11 31 ±13 30 ±11 0.1 ±0.2
Atopya 17 (65%) 47 (60%) 79 (53%) 5 (15%)Prior history, n (%)
Eosinophilic gastritis and/or duodenitis (EG/EoD) 7 (27%) 31 (40%) 19 (13%) 0
Functional gastrointestinal disorderb 19 (73%) 49 (63%) 132 (89%) 0 GERD, acid reflux, or heartburn 12 (46%) 43 (55%) 102 (68%) 0 Peptic ulcer 4 (15%) 6 (8%) 9 (6%) 0Chronic gastritis/duodenitis 2 (8%) 4 (5%) 18 (12%) 0
Physician-guided treatment, n (%)Proton-pump inhibitor 8 (31%) 35 (45%) 54 (36%) 0Diet modification 3 (12%) 4 (5%) 12 (8%) 0
Presented at the American College of Gastroenterology (ACG), October 22nd – 27th, 2021
Figure represents patients combined from ENIGMA and Prevalence study; Controls (n=33); ENIGMA, EG (n=45), EoD (n=62); Prevalence, EG (n=38), EoD (n=142)