ALOX15 Immunohistochemistry Aids in the Diagnosis of Eosinophilic Esophagitis on Pauci-eosinophilic Biopsies in Children

Abstract

Histologically, esophageal biopsies should have ≥15 intraepithelial eosinophils (IEEs) per high power field (HPF) to support a clinicopathologic diagnosis of eosinophilic esophagitis (EoE). Children with clinically apparent EoE may show pauci-eosinophilic biopsies due to patchy involvement. Immunostaining (Immunohistochemistry) for arachidonate-15 lipooxygenase (ALOX15) has been demonstrated to be a sensitive marker for EoE. We retrospectively assessed the expression of ALOX15 in 48 biopsies from 21 patients with established diagnosis of EoE and with tissue fragments below the threshold of 15 IEEs/HPF. Fragments were classified into pauci-eosinophilic and non-pauci-eosinophilic groups using cutoffs of 10 and 15 IEEs/HPF. Controls included patients with reflux and normal biopsies. Sixty-five (43.9%) fragments showed <10 IEEs/HPF and 83 (56.1%) showed ≥10 IEEs/HPF. Using a cutoff of 15 IEEs/HPF, 87 (58.7%) fragments showed <15 IEEs/HPF while 61 fragments (41.2%) had ≥15 IEEs/HPF. ALOX15 was positive in 53/65 (81.5%) of fragments with <10 IEEs/HPF versus 82/83 (98.8%) of fragments with ≥10 IEEs/HPF (P < .001). For a cutoff of 15 IEEs/HPF, 75/87 (86.2%) of pauci-eosinophilic fragments were ALOX15-positive, while 60/61(98.4%) of biopsies meeting the threshold were positive (P < .001). In 3/21 (14.3%) patients with EoE, all of the fragments (n = 7) were pauci-eosinophilic and all of them were positive for ALOX15. Two of 24 patients with reflux (one with 9 and one with 14 IEEs/HPF) were also positive. Fragments from normal controls (0 IEEs/HPF) were negative. Our results support the utility of ALOX15 immunohistochemistry in supporting the diagnosis of EoE in rare situations with strong clinical suspicion where no fragments reach 15 IEEs/HPF.

Keywords

eosinophilic esophagitis reflux ALOX15

Background

Eosinophilic esophagitis (EoE) is a common chronic progressive inflammatory disorder in children and adults that may lead to morbid sequelae including stenosis and strictures. It is diagnostically characterized by a combination of clinical, endoscopic, and pathologic findings including: a history of food allergy, dysphagia, endoscopic findings of plaques, exudates, rings and furrows, and biopsies with ≥15 intraepithelial eosinophils (IEEs)/high power field (HPF), marked basal cell hyperplasia, and subepithelial (lamina propria) fibrosis. Compared to gastroesophageal reflux disease (GERD), the proximal esophagus in EoE is more commonly affected, but brisk pathology in proximal esophageal biopsies from GERD patients has been well documented.¹ Per consensus recommendations, a biopsy should contain a minimum of 15 IEEs/HPF with the eosinophils constituting the majority of inflammatory cells.² A diagnosis of EoE is supported by a normal pH study or unresponsiveness to a trial of proton-pump inhibitor (PPI) therapy. A significant proportion of children with EoE may present with non-specific symptoms such as heartburn, failure-to-thrive, abdominal pain, nausea, and chest pain.³ As a result, EoE in children may present a diagnostic challenge clinically.

Management of EoE differs from reflux esophagitis, which may present with similar symptoms, although patients may have both conditions concomitantly.³ GERD in children is typically treated with PPIs, H2 blockers, and antacids. For severe cases that are refractory to medical management, surgical intervention such as Nissen fundoplication is considered. EoE is typically managed with dietary modification and topical glucocorticoid therapy as evidenced by reduction in the number of IEEs/HPF.^4–6 Endoscopic interventions such as esophageal dilatation are considered for severe cases with strictures but are ineffective in treating the underlying disorder.^7,8 Therefore, it is important to distinguish this entity from normal esophageal mucosa and changes secondary to GERD.

Histopathologically, patients may have clinically apparent EoE but biopsies may not reach the threshold eosinophil count due to the patchy distribution of involvement. Ravi et al.⁹ have previously described a low-grade esophageal eosinophilia in adults that appears to be within the spectrum of EoE. Jakarte¹⁰ also reported in abstract form low IEE count in patients that clinically had EoE and used the terminology “pauci-eosinophilic variant of EoE.” This author used a threshold IEE count of 10 per HPF along with other morphologic features for their/his cases of “pauci-eosinophilic” variant of EoE. Despite the lack of the requisite 15 IEEs/HPF, these patients present with clinical findings of EoE and benefit from management accordingly.^11,9 In our practice, we have encountered examples of such cases in children with clinical and endoscopic features of EoE, but biopsies do not reach the diagnostic threshold of 15 IEEs/HPF.

Arachidonate-15 lipooxygenase (ALOX15) is a non-heme iron-based enzyme involved in di-oxygenation of polyunsaturated fatty acids to form lipid-based cytokines.¹² It is strongly expressed in circulating blood eosinophils and is upregulated in inflammatory conditions such as asthma, which has been associated with EoE. Through gene expression microarray analysis, ALOX15 was found to be upregulated in EoE.¹³ Immunohistochemistry (IHC) for ALOX15 has been demonstrated to be a sensitive and specific marker of EoE in children when biopsies unequivocally met histopathologic criteria for EoE. In contrast, ALOX15 is rarely positive in reflux esophagitis.¹⁴ ALOX15 has not been previously studied in pauci-eosinophilic biopsies of EoE patients. Therefore, we sought to assess the potential utility of ALOX15 IHC for supporting the diagnosis of pauci-eosinophilic cases of EoE in children.

Methods

With institutional review board approval, records from 1/1/2009 to 1/1/2013 from the Department of Pathology and Laboratory Medicine at Rhode Island Hospital were searched. Patients with ages from 1 to 17 years with established clinical diagnosis for EoE were selected and their initial (treatment naïve) biopsies were used for the study. Patients referred with prior history of treatment including use of PPIs were excluded. Clinico-pathologic follow-up of at least 12 months confirmed the diagnosis of EoE in all cases. All patients had at least one follow-up biopsy with 15 IEEs/HPF. The diagnosis of eosinophilic esophagitis was made as defined by the 2011 consensus guidelines.¹⁵ Specifically, patients were required to have symptoms of esophageal dysfunction, one or more esophageal biopsies with a minimum of 15 IEEs/HPF, other causes of esophageal eosinophilia excluded, a trial with PPIs without remission of symptoms, clinical history of other allergies and/or food impaction, and endoscopic findings more consistent with EoE (furrows, rings, and ridging). Clinical diagnosis of GERD included clinical presentation and histopathology consistent with reflux in patients with good symptomatic response to acid suppression. Eosinophil counts were performed using an Olympus BX41 microscope with a HPF diameter of 0.5 mm. Follow-up data including clinical symptoms, endoscopic findings, and history of allergies were collected.

Since most biopsies contained multiple fragments of tissue, the IEE count per HPF was assessed on multiple levels in areas with the most abundant eosinophils. This was performed for each of the fragments. Counts were averaged and correlated with ALOX15 staining for each fragment separately. Some fragments from the same biopsy showed both pauci-eosinophilic and non-pauci-eosinophilic areas depending on whether the pathologic criteria used was ≥15 IEEs/HPF or a more stringent criteria of ≥10 IEEs/HPF. The comparison groups included 24 patients with GERD and 15 patients with normal esophageal biopsies. Clinico-pathologic follow-up of the normal controls at least 12 months after biopsy demonstrated no evidence of GERD or EoE.

IHC assays were performed using anti-ALOX15 antibodies (11-K; Santa Cruz Biotechnology, Santa Cruz, CA; 1:100 dilution). Validated EoE cases in children and adults were used as positive controls. Normal serum was utilized for creating negative controls as previously described.¹⁴ ALOX15 staining was considered positive if greater than 10% of squamous cells demonstrated at least moderate cytoplasmic staining.¹⁴ Adjacent stroma was negative for staining. Statistical analysis comparing ALOX15 staining between the groups was performed using Student’s t-test and Fisher’s exact test using Excel (Microsoft; Redmond, WA) and SAS (SAS Institute, Cary, NC) software.

Results

Twenty-one patients meeting the study criteria were included in the cohort. All 21 patients had other biopsy features of EoE including marked basal cell hyperplasia, elongated papillae, and subepithelial fibrosis. None of the patients had received PPIs at the time of their initial biopsies. The majority of the cases were male (n = 19, 90.5%). Table 1 shows the distribution of clinical symptoms in the studied patients at the time of biopsy. All 21 patients had endoscopic features in keeping with EoE. Proximal and distal involvement was seen in 20 patients (95.2%). Furrows were the most common endoscopic finding (n = 13, 61.9%) followed by ridging (n = 8, 38.1%) and plaques (n = 8, 38.1%).

Table 1.

Clinical Characteristics of Patients.

	EoE (n = 21)	GERD (n = 24)	Normal Controls (n = 15)
Age (range)	10.29 (1–17 years)	10.25 (1–18 years)	10.52 (1–18 years)
Sex (male:female)	19:2	1:1	10:7
Symptoms
Abdominal pain	6 (28.6%)	13 (54.2%)	11 (73.3%)
Vomiting	4 (19.1%)	16 (66.7%)	3 (20.0%)
Dysphagia	3 (14.3%)	6 (25.0%)	1 (6.7%)
Food impaction	6 (28.6%)	3 (12.5%)	0 (0%)
Heartburn	1 (4.76%)	6 (25.0%)	3 (20%)
Failure to thrive	4 (19.1%)	3 (12.5%)	1 (6.7%)
Endoscopy
Normal	0 (0%)	10 (41.7%)	14 (93.3%)
Erythema	5 (23.8%)	4 (16.7%)	0 (0%)
Rings	3 (14.3%)	0 (0%)	1 (6.7%)
Ridging	8 (38.1%)	2 (8.3%)	1 (6.7%)
Furrows	13 (61.9%)	2 (8.3%)	0 (0%)
White plaques	8 (38.1%)	0 (0%)	0 (0%)
Allergies
Food allergy	5 (23.8%)	2 (8.3%)	1 (6.7%)
Asthma	3 (14.3%)	3 (12.5%)	1 (6.7%)
Rhinitis	3 (14.3%)	3 (12.5%)	1 (6.7%)
Dermatitis	3 (14.3%)	0 (0%)	0 (0%)
Involvement
Proximal	20 (95.2%)	0 (0%)	0 (0%)
Distal	21 (100%)	24 (100%)	0 (0%)

Proximal or/and mid, and distal biopsies were included for a total of 48 biopsies from the study cohort, each with 1 to 9 fragments (mean = 3.25 fragments, median = 3 fragments). The fragments were classified into pauci-eosinophilic (<10 IEEs/HPF) and non-pauci-eosinophilic (≥10 IEEs/HPF) groups. Additionally, they were divided into those meeting the EoE criteria of ≥15 IEEs/HPF versus those below this threshold. In total, 148 fragments were assessed. ALOX15 immunostaining was detected in the cytoplasm of squamous cells and eosinophils. Underlying stroma was consistently negative (Figure 1). Sixty-five (43.9%) fragments showed <10 IEEs/HPF, with an average number of 5.7 IEEs/HPF. Eighty-three (56.1%) of the 148 fragments showed ≥10 IEEs/HPF, with a mean of 39.3 IEEs/HPF. Using a cutoff of 15 IEEs/HPF, 87 (58.7%) fragments showed <15 IEEs/HPF while 61 fragments (40.5%) had ≥15 IEEs/HPF.

ALOX15 was positive in 53/65 (81.5%) of fragments with <10 IEEs/HPF versus 82/83 (98.8%) of fragments with >10 IEEs/HPF (P < .001). For a cutoff of 15 IEEs/HPF, 75/87 (86.2%) of the pauci-eosinophilic fragments were ALOX15-positive, while 60/61(98.4%) of the biopsies meeting the threshold were positive (P < .001). In 3/21 patients (14.3%), none of the fragments (n = 7) met EoE criteria of ≥ 15 IEEs/HPF but were positive for ALOX15. Only 2 of 24 cases with GERD had a rare fragment staining positive for ALOX15, one had 9 IEEs/HPF and the other had 14 IEEs/HPF (P < .0001). Biopsies for all normal controls were negative with eosinophil counts of 0 IEE/HPF (P < .0001). Table 2 summarizes the results.

Table 2.

Summary of Results.

	≥10 IEEs/HPF (n = 83 biopsy fragments)	<10 IEEs/HPF (n = 65 biopsy fragments)	≥15 IEEs/HPF (n = 61 biopsy fragments)	<15 IEEs/HPF (n = 87 biopsy fragments)	Reflux (n = 24 patients)	Normal Control (n = 15 patients)
Endoscopic procedures^a	21/21 (100%)	0/21 (0%)	18/21 (86%)	3/21 (14%)	N/A	N/A
ALOX15(+)Biopsy fragments/total fragments in group	82/83 (98.8%)	53/65 (81.5%)	60/61 (98.4%)	75/87 (86.2%)	2/24 (8.3%)	0/15 (0%)
ALOX15(−) Biopsy fragments/total fragments in group	1/83 (1.2%)	12/65 (18.5%)	1/61 (1.6%)	12/87 (13.8%)	22/24 (91.7%)	15/15 (100%)

IEE, intraepithelial eosinophils; HPF, high power field.

^aThis represents the total number of endoscopies with all biopsy fragments meeting the group’s inclusion criteria.

Discussion

In this study, we assessed the use of ALOX15 IHC in the diagnosis of pauci-eosinophilic EoE in children. This was achieved through comparison of biopsies from cohort of patients with confirmed EoE versus pediatric patients with GERD and normal controls. Specifically, we identified tissue fragments from biopsies from pediatric patients with EoE that did not reach the histopathologic criteria of the minimum number of 15 IEEs/HPF and compared them against fragments meeting criteria. Stringent criteria for pauci-eosinophilic EoE of <10 IEEs/HPF was used to assess fragments in addition to the established criteria of 15 IEEs/HPF as the minimum threshold to establish a diagnosis of EoE.

We have previously reported ALOX15 IHC achieving 95% positivity in EoE cases.^13,14 In this study, we observed that ALOX15 was still positive in 81.5% of the pauci-eosinophilic fragments including 3 cases where none of the fragments met histopathologic threshold of 15 IEEs/HPF.

Presently, multiple biopsy fragments are taken from each suspicious section of the esophagus. In the majority of esophageal biopsies from our EoE cohort, at least one tissue fragment contained ≥15 IEEs/HPF. These findings support the current practice of taking multiple biopsy fragments in areas suspicious for EoE for optimal diagnostic yield. An optimal number of 4–6 biopsy fragments from the proximal or mid esophagus has been established.¹⁶ The mean number of biopsies taken in our study was 3.25 with a median of 3. Our results suggest that the need for repeat biopsies may be reduced using ALOX15 IHC in cases with few biopsy fragments, none reaching the minimum diagnostic threshold.^17,18

Various clinical criteria and indices have been reported to help predict EoE in patients. However, use of clinical findings alone for EoE is controversial. Safroneeva et al.¹⁹ demonstrated that scoring of clinical symptoms alone could only modestly predict clinical and pathologic remission of EoE with 67.7% accuracy. Also unclear is the relationship between clinical versus pathologic remission status and risk of EoE sequelae. Our data suggest that biopsies from patients presenting with strong clinical signs of EoE may benefit from adjunct ALOX15 IHC when histopathologic morphology alone fails to make a definitive diagnosis.

Analysis of individual biopsy fragments was beneficial for this study. For most cases, at least one fragment met histopathologic criteria of ≥15 IEEs/HPF for comparison. The GERD and normal control biopsies were not analyzed as fragments. The vast majority of the fragments in these cases were negative for ALOX15 with only 2 GERD patients showing positive staining. These 2 GERD cases that were positive for ALOX15 had focally 14 IEEs/HPF. Since these cases were clinically classified as GERD based on their response to therapy with PPI and the lack of sufficient eosinophilia in their esophageal biopsies, it is likely that they belong to the so-called group of “PPI responsive esophageal eosinophilia.” This category is being increasingly recognized in the medical community as a group of patients with overlapping features of GERD and EoE.^2,20,21 A prospective study of patients within this clinical category would be necessary to establish whether ALOX15 IHC could be used to predict the likelihood of that patient to respond favorably to antireflux therapy or to antigen elimination/topical steroids treatment.

Correlation between eosinophil markers such as eosinophil peroxidase and symptom severity has been studied by Martin et al.²² In that study, symptomatology was collected using a standardized questionnaire. Symptom data suggesting EoE in our group was collected from either legal guardians or patients where appropriate in a nonstandardized fashion. As a result, correlation between severity of symptoms and ALOX15 staining could not be assessed. However, clinical follow-up in all cases confirmed the diagnosis of EoE.

In summary, ALOX15 IHC could be helpful in supporting the diagnosis of EoE in the appropriate clinical setting. It may be especially useful in rare situations in which none of the tissue fragments reach the diagnostic threshold of 15 IEEs/HPF. ALOX15 staining may be positive in pauci-eosinophilic EoE, although sensitivity is slightly reduced compared to conventional EoE.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was partially supported by the Department of Pathology and Laboratory Medicine of Lifespan Academic Medical Center.

References

Molina-Infante

Ferrando-Lamana

Mateos-Rodriguez

Perez-Gallardo

Prieto-Bermejo

. Overlap of reflux and eosinophilic esophagitis in two patients requiring different therapies: a review of the literature. World J Gastroenterol 2008; 14(9): 1463–1466.

Liacouras

Furuta

Hirano

et al.

Eosinophilic esophagitis: updated consensus recommendations for children and adults. J Allergy Clin Immunol 2011; 128(1): 3–20 e6. quiz 1–2.

Miehlke

. Clinical features of Eosinophilic esophagitis in children and adults. Best Pract Res Clin Gastroenterol 2015; 29(5): 739–748.

D’Alessandro

Esposito

Pesce

Cuomo

De Palma

Sarnelli

. Eosinophilic esophagitis: from pathophysiology to treatment. World J Gastrointest Pathophysiol 2015; 6(4): 150–158.

Gonsalves

. Distinct features in the clinical presentations of eosinophilic esophagitis in children and adults: is this the same disease? Dig Dis 2014; 32(1–2): 89–92.

Gonsalves

. Steroids versus dietary therapy for the treatment of eosinophilic esophagitis. Curr Opin Gastroenterol 2014; 30(4): 396–401.

Schoepfer

Gonsalves

Bussmann

et al.

Esophageal dilation in eosinophilic esophagitis: effectiveness, safety, and impact on the underlying inflammation. Am J Gastroenterol 2010; 105(5): 1062–1070.

Muir

Merves

Liacouras

. Role of endoscopy in diagnosis and management of pediatric eosinophilic esophagitis. Gastrointest Endosc Clin N Am 2016; 26(1): 187–200.

Ravi

Talley

Smyrk

et al.

Low grade esophageal eosinophilia in adults: an unrecognized part of the spectrum of eosinophilic esophagitis?

Dig Dis Sci 2011; 56(7): 1981–1986.

10.

Jakarte

. Pauci-eosinophilic esophagitis: an under-recognized variant of eosinophilic esophagitis. Mod Pathol 2011; 24(15): 153A.

11.

Doyle

Odze

. Eosinophilic esophagitis without abundant eosinophils? The expanding spectrum of a disease that is difficult to define. Dig Dis Sci 2011; 56(7): 1923–1925.

12.

Ivanov

Kuhn

Heydeck

. Structural and functional biology of arachidonic acid 15-lipoxygenase-1 (ALOX15). Gene 2015; 573(1): 1–32.

13.

Matoso

Mukkada

et al.

Expression microarray analysis identifies novel epithelial-derived protein markers in eosinophilic esophagitis. Mod Pathol 2013(5); 26: 665–676.

14.

Matoso

Allen

Herzlinger

et al.

Correlation of ALOX15 expression with eosinophilic or reflux esophagitis in a cohort of pediatric patients with esophageal eosinophilia. Hum Pathol 2014; 45(6): 1205–1212.

15.

Liacouras

. Clinical presentation and treatment of pediatric patients with eosinophilic esophagitis. Gastroenterol Hepatol (N Y) 2011; 7(4): 264–267.

16.

Nielsen

Lager

Lewin

Rendon

Roberts

. The optimal number of biopsy fragments to establish a morphologic diagnosis of eosinophilic esophagitis. Am J Gastroenterol 2014; 109(4): 515–520.

17.

Lipka

Kumar

Richter

. Impact of diagnostic delay and other risk factors on eosinophilic esophagitis phenotype and esophageal diameter. J Clin Gastroenterol 2016; 50(2): 134–140.

18.

Schroeder

Robinson

Masterson

et al.

Esophageal human beta-defensin expression in eosinophilic esophagitis. Pediatr Res 2013; 73(5): 647–654.

19.

Safroneeva

Straumann

Coslovsky

et al.

Symptoms have modest accuracy in detecting endoscopic and histologic remission in adults with eosinophilic esophagitis. Gastroenterology 2016; 150(3): 581–590 e4.

20.

Molina-Infante

Prados-Manzano

Gonzalez-Cordero

. The role of proton pump inhibitor therapy in the management of eosinophilic esophagitis. Expert Rev Clin Immunol 2016; 12(9): 945–952.

21.

Jiao

Ishimura

Maruyama

et al.

Similarities and differences among eosinophilic esophagitis, proton-pump inhibitor-responsive esophageal eosinophilia, and reflux esophagitis: comparisons of clinical, endoscopic, and histopathological findings in Japanese patients. J Gastroenterol 2016; 52(2): 203–210.

22.

Martin

Franciosi

Collins

et al.

Pediatric eosinophilic esophagitis symptom scores (PEESS v2.0) identify histologic and molecular correlates of the key clinical features of disease. J Allergy Clin Immunol 2015; 135(6): 1519–1528 e8.