Methods for Treating Eosinophilic Esophagitis

The present invention relates to methods for identifying an EoE endotype of a patient and treating the patient with one or more therapies targeted to the patient's disease endotype; and related methods for stratifying patients for clinical trials.

Skip to: Description  ·  Claims  · Patent History  ·  Patent History
Description
CROSS REFERENCE TO RELATED APPLICATIONS

This application is a continuation of U.S. patent application Ser. No. 16/281,750, filed Feb. 21, 2019, which claims the benefit of and priority under 35 U.S.C. § 119 to U.S. Provisional Application No. 62/634,446, filed Feb. 23, 2018, the entire content of which is incorporated herein by reference in its entirety.

STATEMENT REGARDING FEDERALLY-SPONSORED RESEARCH

This invention was made with US government support under Grant Nos. A1045898, A1070235, and AI117804 awarded by the National Institute of Health. The government has certain rights in the invention.

FIELD OF THE INVENTION

The disclosure relates to methods for treating eosinophilic esophagitis by determining the disease endotype.

BACKGROUND OF THE INVENTION

Eosinophilic esophagitis (EoE) is an emerging disease characterized by marked esophageal specific eosinophilia that is typically driven by allergic sensitization to a variety of common foods. Lucendo A J, et al. United Eur. Gastroent. J. 2017; 5(3): 335-58. The diagnosis is dependent upon quantitative assessment of esophageal levels of eosinophils (i.e., peak eosinophil count of ≥15 intraepithelial eosinophils in one high-power field [HPF]). Although the gold standard for diagnosing disease and monitoring disease activity is the esophageal eosinophil level, recent advances have identified the potential value of a deeper analysis based on a wide range of quantifiable molecular, endoscopic, and histologic parameters. Warners M J et al. Am J Gastroenterol 2017; 112(11): 1658-69; Wen T, Rothenberg M E. Front Med (Lausanne) 2017; 4:108. In particular, a unique EoE transcriptome, referred to as the EoE Diagnostic Panel (EDP), a set of esophageal transcripts that distinguishes EoE from control individuals including those with gastroesophageal reflux disease, correlates with distinct disease features and can identify EoE amongst ambiguous cases. Wen T, et al. Gastroenterology 2013; 145(6): 1289-99. In addition, a deeper histologic assessment called the histologic scoring system (HSS) has been described and takes into account disease stage and grade across eight different parameters beyond peak eosinophil levels. Collins et al. Dis Esophagus 2017; 30(3): 1-8. Finally, a broad panel of endoscopic features, as measured by EoE endoscopic reference scoring (EREFS), which takes into account five endoscopic features (i.e., concentric rings, longitudinal furrows, white plaques/exudates, edema, and strictures), has significance in terms of understanding clinical features and monitoring the effect of therapy in both children and adults. Wechsler J B, et al. Clin Gastroenterol Hepatol 2017.

An outstanding need in the EoE field is to define the relationships between these various clinical, endoscopic, and histologic features (especially the gold standard of the disease, the esophageal eosinophil level) and the degree of patient heterogeneity, as current therapy is not governed by specific disease features. Atkins D, et al. Pediatr Allergy Immunol 2017; 28(4): 312-9. Although a fibrostenotic phenotype has been associated with a subset of subjects with EoE, its molecular features, particularly in comparison to a non-fibrostenotic phenotype, has not yet been determined. At present, EoE is treated by food elimination trials, focused on the most highly allergenic foods, and topical glucocorticoid therapy.

An unmet need in the therapy of eosinophilic esophagitis (“EoE”) is the ability to identify clinically relevant disease endotypes in EoE patients in order to tailor each patient's treatment to that most likely to benefit the patient's specific form of disease. The present invention addresses this need.

SUMMARY OF THE INVENTION

The present invention is based, in part, on the discovery of three distinct EoE subtypes, referred to herein as disease “endotypes” and more particularly as EoEe1 or “mild”, EoEe2 or “intermediate”, and EoEe3 or “severe”, each of which has characteristic histological and endoscopic features and is associated with distinct clinical characteristics and phenotypes. Collectively, the endotypes described here are useful to stratify patients with EoE into clinically relevant subgroups of mild, intermediate, or severe disease, thereby providing a framework for a precision medicine approach to EoE therapy, as described in more detail below. Generally, the EoEe1 endotype is characterized as having the mildest histological and clinical phenotype, most closely resembling findings seen in healthy tissue of normal biopsies. The EoEe2 endotype is generally characterized by substantial inflammatory changes, type-2 immune responses, and evidence of refractoriness to steroids. The EoEe3 endotype is generally characterized by a strong association with the presence of a narrow-caliber esophagus, the highest degree of endoscopic and histologic severity, and the lowest expression of epithelial differentiation genes. Accordingly, the disclosure provides methods for identifying the EoE endotype of a patient in need of treatment for EoE, including for example a patient diagnosed with EoE, and treating the patient with one or more therapies targeted to the patient's disease endotype, and related methods for stratifying patients for clinical trials.

In embodiments, the disclosure provides methods for treating eosinophilic esophagitis (EoE) in a subject in need thereof, the method comprising subjecting a biological sample from the subject to a method for gene expression analysis, determining the expression of one or more genes or a panel of genes in the biological sample, determining the subject's EoE endotype based on the expression of the one or more genes or panel of genes, and treating the patient with an EoE therapy tailored to the patient's EoE endotype. In embodiments, the EoE endotype is selected from mild, intermediate, and severe.

In embodiments, treating the patient with an EoE therapy tailored to the patient's EoE endotype comprises one or more of the following,

    • where the EoE endotype is determined to be mild, treating the patient with one or both of proton pump inhibitor (PPI) therapy and dietary therapy;
    • where the EoE endotype is determined to be intermediate, treating the patient with one or more of an anti-cytokine therapy, an anti-TSLP therapy, and an anti-ALOX15 therapy; and
    • where the EoE endotype is determined to be severe, treating the patient with one or more of anti-ALOX15 therapy, esophageal dilation, anti-cytokine therapy, and glucocorticoid therapy.

In embodiments, the biological sample is an esophageal biopsy sample.

In embodiments, the determining the expression of one or more genes or a panel of genes in the biological sample is performed using a PCR-based method.

In embodiments, the determining the subject's EoE endotype based on the expression of the one or more genes or panel of genes is performed by a method comprising linear discriminant analysis. In embodiments, the linear discriminant analysis comprises determining a probability distance. In embodiments, the probability distance is the Mahalanobis distance.

In embodiments, the at least one gene or panel of genes comprises one or more of ANO1, CRYM, DSG1, PNLIPRP3, TNFAIP6, TSLP, UPK1A, and WDR36. In embodiments, the at least one gene or panel of genes further comprises one or more of the genes set forth in Table 2. In embodiments, the at least one gene or panel of genes comprises one or more of ALOX15, APOBEC3A, CDA, CRISP3, ACTG2, CCR3, FFAR3, IL4, RGS9BP, TSLP, CTNNAL1, EML1, FLG, PNLIPRP3 and TSPAN12.

In embodiments, the endotype is further characterized by one or more histologic, endoscopic, or clinical features. In embodiments, the one or more histologic features is selected from basal zone hyperplasia (BZH) and surface epithelial alteration (SEA). In embodiments, the one or more endoscopic features is selected from the occurrence of edema, exudates, and furrows. In embodiments, the one or more clinical features is selected from pediatric onset, adult onset, atopic, non-atopic, steroid sensitivity, steroid refractory, and fibrostenotic.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1A-C: Associations between peak esophageal eosinophil counts in eosinophilic esophagitis and diagnostic platforms. A linear correlation is seen between peak esophageal eosinophils per HPF and total score from EDP (A), HSS (B), and EREFS (C). EDP=eosinophilic esophagitis diagnostic panel. EREFS=eosinophilic esophagitis endoscopic reference score. HPF=high-power field. HSS=eosinophilic esophagitis histology scoring system.

FIG. 2A-B: Associations between the EDP and HSS domains and EREFS features Spearman correlation analysis between gene expression levels on EDP and HSS domains (A) and EREFS features (B), using the absolute value to account for differences in the direction of the effect across genes. p values calculated with the Kruskal-Wallis test and Dunn's post-hoc test. Datapoints represent the absolute median and error bars the IQR. BZH=basal zone hyperplasia. DEC=dyskeratotic epithelial cells. DIS=dilated intercellular spaces. EA=eosinophilic abscess. EDP=eosinophilic esophagitis diagnostic panel. EI=eosinophilic inflammation. EREFS=eosinophilic esophagitis endoscopic reference score. ESL=eosinophilic surface layering. HSS=eosinophilic esophagitis histology scoring system. LPF=lamina propria fibres. SEA=surface epithelial alteration. *Dunn's post-hoc test, p<0.0001 vs EA, ESL, DEC, DIS, SEA, and LPF. †Dunn's post-hoc test, p=0.0324 vs oedema, p=0.0034 vs exudates, and p<0.0001 vs rings and stricture.

FIG. 3A-E: Clustering analysis of the active eosinophilic esophagitis group in the discovery cohort. (A) Consensus CDF with increasing number of clusters (k2 to k12). (B) Unsupervised consensus clustering of active eosinophilic esophagitis showed optimum partitioning to three clusters (endotypes). (C) Comparison of esophageal transcriptomes by endotype. Color range is based on log 2 normalized intensity value. (D) Three-dimensional plot containing sample points from the three endotypes, derived from principal component analysis of the entities shown in the heat map to visualize the geometric distance between any given samples. (E) Venn diagrams were generated based on 92 EDP transcripts that met criteria for differentially expressed genes. Venn diagrams compare the number of genes identified as differentially expressed genes (adjusted p<0.05 and two-fold change) that characterize the three endotypes. CDF=cumulative distribution function. EDP=eosinophilic esophagitis diagnostic panel. EoEe=eosinophilic esophagitis endotype.

FIG. 4A-D: Clinical features of each eosinophilic esophagitis endotype. (A) Comparison of each eosinophilic esophagitis endotype by diagnostic platform, peak esophageal eosinophil count (upper left), EDP score (upper right), HSS score (lower left), and EREFS score (lower right). Centre line represents the median and error bars represent the IQR. Every dot represents an individual participant. p values at the top of each graph are calculated by the Kruskal-Wallis test, and p values indicated by symbols are calculated with Dunn's post-hoc test and are versus EoEe1. (B) Comparison of each HSS domain in each eosinophilic esophagitis endotype (upper) and each EREFS feature in each eosinophilic esophagitis endotype (lower). Data are mean (SE). p values at the top of each graph (for each feature) are calculated by the Kruskal-Wallis test, and p values indicated by symbols are calculated by Dunn's post-hoc test and are versus EoEe1. (C) Summary of significant associations for each endotype. (D) Multiple correspondence analysis of the relations between clinical phenotypes and endotypes. Dimension 1 represents the summation of the major variations, whereas Dimension 2 represents the minor variation for each data point. Distance between variables (phenotype and endotype) indicates the approximate relation between variables. The distance between variables is inversely proportional to the strength of the relation. Circles have been added to emphasize the proximity between points. BZH=basal zone hyperplasia. DEC=dyskeratotic epithelial cells. DIS=dilated intercellular spaces. EA=eosinophilic abscess. EDP=eosinophilic esophagitis diagnostic panel. EI=eosinophilic inflammation. EoEe=eosinophilic esophagitis endotype. EREFS=eosinophilic esophagitis endoscopic reference score. ESL=eosinophilic surface layering. HPF=high-power field. HSS=eosinophilic esophagitis histology scoring system. LPF=lamina propria fibres. RR=risk ratio. SEA=surface epithelial alteration. *p<0.0001. †p=0.0003. ‡p=0.0308. § p=0.0007. ¶p=0.0004. ∥p=0.0178. **p=0.0096. ††p=0.0017. ‡‡p=0.0317. §§ p=0.0163.

FIG. 5A-J: Eosinophilic esophagitis endotype prediction based on machine learning with high accuracy. (A) Stepwise discriminant analysis shows the eight strongest discriminatory genes for cluster assignment (red bars). (B) Canonical plot in which participants are plotted in a two-dimensional space. Every dot represents an individual participant. Canonical 1 represents the summation of major variations, whereas Canonical 2 represents the minor variation for each data point. A 95% CI ellipse (inner) and an ellipse denoting a 50% contour (outer) are plotted for each group. The flow of the analysis is plotted in the lower diagram. Diagnostic accuracy is summarized in the right-hand table. EDP=eosinophilic esophagitis diagnostic panel. EoEe=eosinophilic esophagitis endotype. NPV=negative predictive value. PPV=positive predictive value. (C-J) Gene expression measured as delta Ct in each endotype for PNLIPRP3 (C), CRYM (D), WDR36 (E), DSG1 (F), TSLP (G), TNFAIP6 (H), ANO1 (I), and UPK1A (J). Comparison of continuous variables between diagnostic groups was performed by Wilcoxon/Kruskal-Wallis tests (rank sums), using nonparametric comparison for all pairs using Dunn method for joint ranking.

DETAILED DESCRIPTION OF THE INVENTION

The present invention is based, in part, on the discovery of three distinct EoE endotypes, EoEe1, EoEe2, and EoEe3, which can be determined based on their differential gene expression patterns, as described infra. Each of the EoE endotypes described here has a pattern of gene expression as well as distinct histological, endoscopic, and clinical features. The disclosure provides methods for stratifying EoE patients into one of these endotypes and personalizing patient therapy based on same. In this context, the term ‘stratify’ and ‘classify’ are used interchangeably to refer to the grouping of EoE patients into one of at least three distinct EoE endotypes using the methods described here. An “endotype” refers to an EoE disease subtype defined by distinct molecular and cellular markers relevant to disease pathology. Each of the at least three EoE endotypes provided by the present disclosure is described in more detail in the following sections.

Generally, the histological features of each endotype are defined by a histological assessment, particularly the eosinophilic esophagitis histology scoring system (HSS), which takes into account disease stage and grade across eight different variables beyond peak eosinophil levels. See e.g., Collins et al. Newly developed and validated eosinophilic esophagitis histology scoring system and evidence that it outperforms peak eosinophil count for disease diagnosis and monitoring. Dis Esophagus 2017; 30: 1-8. The HSS assesses eight histological features: eosinophilic inflammation, basal zone hyperplasia, eosinophilic abscess, eosinophilic surface layering, dilated intercellular spaces, surface epithelial alteration, dyskeratotic epithelial cells, and lamina propria fibers. Each feature is scored on a 4-point scale for severity (grade) or extent (stage) of the abnormality, with 0 representing normal features and 3 denoting most severe or extensive features. A final HSS score (grade or stage) is the sum of the assigned scores for each feature assessed divided by the maximum possible score for that biopsy specimen.

As discussed in more detail below, two HSS features, basal zone hyperplasia (BZH) and surface epithelial alteration (SEA), demonstrated significant association with the EoE endotypes described here. BZH was significantly higher in EoEe2 compared to EoEe1, and SEA was significantly higher in EoEe3 compared to EoEe1.

The endoscopic features of each endotype are defined by the eosinophilic esophagitis endoscopic reference score (EREFS). The EREFS takes into account five endoscopic features, edema, concentric rings, white plaques or exudates, longitudinal furrows, and strictures, and has relevance in terms of understanding the clinical features and monitoring the effect of treatment in both children and adults.

As discussed in more detail below, the EREFS features of edema, exudates, and furrows showed significant association with the EoE endotypes described herein. Edema was higher in both EoEe2 and EoEe3 compared to EoEe1. The occurrence of exudates was significantly higher in EoEe2 compared to EoEe1, and the occurrence of furrows was significantly higher in EoEe3 compared to EoEe1.

Each endotype described here is also associated with specific clinical characteristics and phenotypes. For example, EoEe1 is strongly associated with normal appearing esophagus and the clinical phenotypes designated atopic, pediatric-onset, steroid sensitive, and normal endoscopic appearance. For this reason, the EoEe1 endotype may also be referred to herein as the ‘mild’ endotype.

EoEe2 is strongly associated with steroid refractory disease and the clinical phenotypes designated inflammatory and steroid-refractory. For this reason, the EoEe2 endotype may also be referred to herein as the ‘intermediate’, ‘inflammatory’ or ‘steroid-refractory’ endotype.

EoEe3 is strongly associated with the presence of narrow-caliber esophagus and the clinical phenotypes designated non-atopic, adult onset, and fibrostenotic. In the context of the present disclosure, the endotypes may also be referred to with reference to their associated clinical phenotypes. For this reason, the EoEe3 endotype may also be referred to herein as the ‘severe’ endotype.

In accordance with the methods described here, a patient's EoE disease is determined to be of a mild, intermediate, or severe EoE endotype. The methods comprise determining the gene expression of one or more genes, or the gene expression profile of a panel of genes, whose expression is associated with a mild, intermediate, or severe endotype. The disclosure further provides for the targeted treatment of a patient based on EoE endotype. In accordance with the methods described here, the patient may be treated with one or more EoE therapies, including, for example, proton pump inhibitor therapy, dietary therapy, anti-cytokine therapy, anti-ALOX15 therapy, anti-TSLP therapy, glucocorticoid therapy, and esophageal dilation. In the context of the present disclosure, the terms “treatment”, “treating”, or “treat” describe the management and care of a patient for the purpose of combating EoE and may include the administration of a therapeutic agent as well as the administration of a therapy such as a restricted diet, including for example elemental and elimination diets, or a medical procedure such as esophageal dilation, to alleviate one or more symptoms or complications of EoE, or to eliminate one or more symptoms or complications of EoE, thereby treating the EoE.

Therapeutic agents may include small molecules, such as proton pump inhibitors and glucocorticoids, or biologic agents, such as therapeutic antibodies or nucleic acids, including interfering RNAs.

Proton pump inhibitor (PPI) therapy may include treatment with a PPI such as dexlansoprazole, esomeprazole, lansoprazole, omeprazole, pantoprazole, and rebeprazole.

Dietary therapy may include, for example, elemental and elimination diets.

Anti-cytokine therapy may include, for example, a biologic agent targeted to inhibit cytokine signaling by one or more cytokines via their cognate receptors. In embodiments, the anti-cytokine therapy is an anti-T helper type 2 (Th2) therapy. A Th2 immune response is generally characterized by the production of interleukin-4 (IL-4), interleukin-5 (IL-5), and interleukin-13 (IL-13). Accordingly, an anti-Th2 therapy encompasses a therapy targeting one or more of IL-4, IL-5, and IL-13, and/or their receptors in order to inhibit IL-4, IL-5, and/or IL-13 mediated signal transduction. The most common biologics for anti-cytokine therapy are antibodies, preferably monoclonal antibodies, and most preferably fully human or humanized monoclonal antibodies. In embodiments of the methods described here, the anti-cytokine therapy is an anti-T helper type 2 (Th2) therapy selected from one or more of a therapy targeting the IL-4 and/or IL-13 signaling pathway, and a therapy targeting the IL-5 signaling pathway.

Interleukin-4 and interleukin-13 both mediate inflammation through their receptors, with IL-13 also binding to type 2 IL-4 receptors. IL-4 and IL-13 signaling pathways thus overlap and therapies envisioned by the methods described here may target one or both of these signaling pathways. Therapies targeting IL-4 signaling include monoclonal antibodies such as dupilumab, which targets the IL-4 receptor alpha (IL-4Ra). Therapies targeting IL-13 signaling include monoclonal antibodies such as RPC4046 or tralokinumab, both of which target IL-13.

Interleukin-5 (IL-5, CD125) is an eosinophil growth, activation, and survival factor. Humanized anti-IL-5 antibodies have been shown to be effective in treating asthma patients with the severe eosinophilic form of the disease, as discussed in Rothenberg M E. Humanized Anti-IL-5 Antibody Therapy. Cell 2016; 165(3): 509. Therapies targeting the IL-5 signaling pathway include, for example, therapies targeting IL-5 and its receptor, also known as CD125. Such therapies include monoclonal antibodies such as mepolizumab and reslizumab, which target IL-5, and monoclonal antibodies such as benralizumab, which target the IL-5 receptor.

Anti-ALOX15 therapy is therapy directed at suppressing the expression or activity of the ALOX15 gene product, arachidonate 15-lipoxygenase. Examples of ALOX15 inhibitors include PD146176.

Anti-TSLP therapy or anti-ALOX15 therapy may also comprise the administration of a single or double stranded ribonucleic acid (RNA) agent that inhibits the expression of the TSLP gene or the ALOX15 gene, for example, by catalyzing the post-transcriptional cleavage of the target mRNA, or by inhibiting transcription or translation of the target mRNA. In embodiments, the RNA agent is a double stranded or single stranded RNA interference-based agent (RNAi). The RNAi agent may be based on a microRNA (miRNA), a short hairpin RNA (shRNA), or a small interfering RNA (siRNA). The RNAi agent comprises a region that is at least partially, and in some embodiments fully, complementary to the target RNA. Although perfect complementarity is not required, the correspondence should be sufficient to enable the RNAi agent, or its cleavage product in the case of double stranded siRNA or RNAi agents comprising cleavable linkers, to direct sequence specific silencing of the target mRNA, e.g., by RNAi-directed cleavage of the target mRNA.

Glucocorticoid therapy may comprise, for example, therapy with one or more glucocorticoids selected from fluticasone, prednisone and budesonide.

Previous work has suggested that EoE progresses from a chronic inflammatory phenotype, which may manifest on endoscopy with white exudates, edema, and linear furrows, to fibrostenosis, which is characterized by esophageal rings, strictures, and/or narrowing. Accordingly, the disclosure also provides methods for monitoring the progression of EoE in a patient diagnosed with EoE, and methods for monitoring the efficacy of therapy for an EoE patient. These methods may additional temporal assaying steps in order to monitor the changes in a patient's gene expression profile before and during the course of therapy, in order to determine the progression of the patient's EoE, for example from a mild endotype to an intermediate or severe endotype, and/or to monitor the effectiveness of the therapy in delaying the onset of a more serious endotype or ameliorating one or more symptoms of the patient's disease, including reverting to a less serious endotype.

Classification of patients into one of the EoE endotypes described here can also be used to stratify clinical trial participants, either prospectively or retrospectively, to identify subgroups with distinct responsiveness to therapy. Accordingly, the disclosure also provides methods for the stratification of clinical trial participants to identify participants whose disease can be characterized according to an EoE endotype as described here.

The present disclosure provides three defined EoE endotypes, which are described in more detail below, along with exemplary EoE therapies targeted to each. The disclosure also provides methods for differentiating among these three endotypes based on the expression of one or more genes or a panel genes. In embodiments, the one or more genes or panel of genes comprises one or more genes selected from the group consisting of ACTG2, ALOX15, ANO1, APOBEC3A, CCR3, CDA, CRISP3, CRYM, DSG1, FFAR3, IL4, PNLIPRP3, RGS9BP, TNFAIP6, TSLP, UPK1A, and WDR36. In embodiments, the expression of a panel of genes is assayed. In embodiments, the panel comprises or consists of two or more, or all eight, of the following genes: ANO1, CRYM, DSG1, PNLIPRP3, TNFAIP6, TSLP, UPK1A, and WDR36, which were determined by machine learning assisted methods to be sufficient for discriminating between the three endotypes, as described in more detail below. In embodiments, the panel comprises or consists of one or more of the genes disclosed in Table 2. In embodiments, the panel of genes assayed comprises or consists of one or more, or all, of ALOX15, APOBEC3A, CDA, CRISP3, ACTG2, CCR3, FFAR3, IL4, RGS9BP, TSLP; and PNLIPRP3; or ALOX15, APOBEC3A, CDA, CRISP3, ACTG2, CCR3, FFAR3, IL4, RGS9BP, TSLP, CTNNAL1, EML1, FLG, PNLIPRP3 and TSPAN12, each of which is differentially expressed among the three endotypes described here, as detailed in Table 2.

In accordance with the methods for determining a patient's EoE endotype described here, the methods may comprise determining the gene expression profile for a panel of genes selected comprising one or all of the genes in the following panels:

    • Panel 1: ANO1, CRYM, DSG1, PNLIPRP3, TNFAIP6, TSLP, UPK1A, and WDR36;
    • Panel 2: ALOX15, APOBEC3A, CDA, and CRISP3;
    • Panel 3: ACTG2, CCR3, FFAR3, IL4, RGS9BP, and TSLP; and
    • Panel 4: CTNNAL1, EML1, FLG, PNLIPRP3, and TSPAN12.

In this context, a gene expression profile represents the expression of multiple genes and may include genes whose expression is increased or decreased relative to a reference, for example as set forth in Table 2.

The disclosure provides methods of determining a patient's EoE endotype, the methods comprising assaying the expression of at least one gene or panel of genes selected from the genes set forth in Table 2.

In embodiments, the at least one gene or panel of genes is selected from the group consisting of ACTG2, ALOX15, ANO1, APOBEC3A, CCR3, CDA, CRISP3, CRYM, DSG1, FFAR3, IL4, PNLIPRP3, RGS9BP, TNFAIP6, TSLP, UPK1A, and WDR36. In embodiments, the at least one gene or panel of genes is selected from the group consisting of ALOX15, APOBEC3A, CDA, and CRISP3, wherein the expression of one or both of ALOX15 and APOBEC3A is decreased more than 2-fold and the expression of one or both of CDA and CRISP3 is increased more than 2-fold, compared to a reference, and the subject's EoE endotype is determined to be mild. In embodiments, the methods may also comprise assaying the expression of one or more of ALOX12, COL8A2, ENDOU, EPB41L3, GCNT3, HILPDA, IGFL1, PLAUR, SPINK7, UPK1A, and ZNF365. In embodiments, a subject determined to have a mild EoE endotype is treated with an EoE therapy targeted to the endotype, for example the subject is administered a therapy comprising one or both of proton pump inhibitor (PPI) therapy and dietary therapy.

In embodiments, the at least one gene or panel of genes is selected from the group consisting of ACTG2, CCR3, FFAR3, IL4, RGS9BP, and TSLP. In embodiments, the expression of one or more of ACTG2, CCR3, FFAR3, IL4, RGS9BP, and TSLP is increased more than 2-fold, more than 5-fold, or more than 10-fold, compared to a reference, and the subject's EoE endotype is determined to be intermediate. In embodiments, a subject determined to have an intermediate EoE endotype is treated with an EoE therapy targeted to the endotype, for example the subject is administered a therapy comprising one or more of anti-cytokine therapy, anti-TSLP therapy, and anti-ALOX15 therapy. In embodiments, a subject whose EoE endotype is determined to be intermediate as described herein is not administered glucocorticoid therapy.

In embodiments, the at least one gene or panel of genes is selected from the group consisting of ACPP, CITED2, CTNNAL1, EML1, FLG, GRPEL2, MT1M, PNLIPRP3, and TSPAN12. In embodiments, the expression of one or more of CTNNAL1, EML1, FLG, PNLIPRP3, and TSPAN12 is decreased at least 2-fold, at least 5-fold, or at least 10-fold, compared to a reference, and the subject's EoE endotype is determined to be severe. In embodiments, a subject determined to have a severe EoE endotype is treated with an EoE therapy targeted to the endotype, for example the subject is administered a therapy comprising one or more of anti-ALOX15 therapy, esophageal dilation, anti-cytokine therapy, and glucocorticoid therapy.

The methods of the present disclosure are preferably applicable to human subjects, also referred to as “patients”, but the methods may also be applied to other mammalian subjects. Accordingly, in embodiments a method described here may be performed on a “subject” which may include any mammal, for example a human, primate, mouse, rat, dog, cat, cow, horse, goat, camel, sheep or a pig. Preferably, the subject is a human. The term “patient” refers to a human subject.

In accordance with the methods described here, in some embodiments, a subject “in need of” treatment may be a subject who has already been diagnosed with EoE. In other embodiments, the subject “in need of” treatment may be one who is suffering from one or more symptoms such as difficulty swallowing, impaction of food in the esophagus, chest pain, heartburn, or upper abdominal pain.

The term “biological sample” as used herein may refer to a sample, including a biopsy sample, of a tissue, or other biological sample such as an exudate, saliva, serum, plasma, mucus, blood, or urine sample; or a swab such as an oral or a buccal swab. In some embodiments, the sample is a tissue sample, for example an esophageal tissue sample obtained at biopsy.

Endotype 1 “Mild or EoEe1”

The EoEe1 endotype is generally characterized by relatively small changes in epithelial differentiation genes, a pauci-inflammatory state, and a greater proportion of normal-appearing esophagus by endoscopy. Other clinical features associated with EoEe1 include pediatric onset, atopic, and steroid sensitivity.

EoEe1 is also characterized by markedly low expression of the ALOX15 gene. ALOX15 may also be referred to as arachidonate 15-lipoxygenase or 15-lipoxygenase-1. Other genes differentially expressed in EoEe1 include APOBEC3A, which is also underexpressed in EoEe1 compared to either EoEe2 or EoEe3; CDA, and CRISP3, each of which is overexpressed in EoEe1. Additional genes that are differentially expressed in EoEe1 include ALOX12, COL8A2, ENDOU, EPB41L3, GCNT3, HILPDA, IGFL1, PLAUR, SPINK7, UPK1A, and ZNF365.

The disclosure provides methods for treating EoE patients whose disease endotype is determined to be mild, or EoEe1. The methods comprise one or both of proton pump inhibitor (PPI) therapy and dietary therapy, which may include for example elemental and elimination diets. Suitable proton pump inhibitors for use in PPI therapy are described above.

Endotype 2 “Intermediate or EoEe2”

The EoEe2 endotype is generally characterized by particularly high type-2 immune response mechanisms and a steroid-refractory phenotype. Other clinical features associated with EoEe2 include pediatric onset.

EoEe2 is also characterized by high expression of the interleukin-4 (IL-4), thymic stromal lymphopoietin (TSLP), and the actin gamma smooth muscle 2 (ACTG2) genes. Other genes differentially expressed in EoEe2 include CCR3, FFAR3, and RGS9BP, each of which is overexpressed in EoEe2 relative to either EoEe2 or EoEe3. Additional genes that are differentially expressed in EoEe2 include CLEC16A, FKBP5, HPGDS, IL5RA, KRT23, LRRC32, MUC4, NTRK1, PTGFRN, RUNX2, SAMSN1, TGFB1, UPK1B and WDR36.

The disclosure provides methods for treating EoE patients whose disease endotype is determined to be intermediate, or EoEe2. The methods comprise anti-cytokine therapy, including anti-Th2 immune therapies, such as therapies targeting the IL-4 and/or IL-13 signaling pathway, and therapies targeting the IL-5 signaling pathway. In embodiments, the anti-cytokine therapy is an anti-Th2 therapy. In embodiments, the anti-Th2 therapy is a monoclonal antibody targeted against one or more of IL-4, IL-13, and IL-5, or their receptors. In embodiments, the therapy is a monoclonal antibody targeted against IL-5 or its receptor. In embodiments, the monoclonal antibody targeted against IL-5 or its receptor is selected from mepolizumab, reslizumab, and benralizumab. In embodiments, the therapy is a monoclonal antibody targeted against IL-4 or its receptor. In embodiments, the monoclonal antibody targeted against IL-4 or its receptor is dupilumab. In embodiments, the therapy is a monoclonal antibody targeted against IL-13 or its receptor. In embodiments, the monoclonal antibody targeted against IL-13 or its receptor is RPC4046 or tralokinumab.

The methods may also comprise anti-TSLP therapy, such as monoclonal antibodies against TSLP including tezepelumab (MED19929, AMG 157), and therapies targeted to suppress ALOX15.

Endotype 3 “Severe or EoEe3”

The EoEe3 endotype is generally characterized by low expression of epithelial differentiation genes and a greater frequency of narrow-caliber esophagus. Other clinical features associated with EoEe3 include adult onset, non-atopic, and fibrostenotic.

EoEe3 is also characterized by low expression of the EML1, PNLIPPR3, and TSPAN12 genes. Other genes differentially expressed in EoEe3 include ACPP, CITED2, CTNNALI, FLG, GRPEL2, and MT1M.

The disclosure provides methods for treating EoE patients whose disease endotype is determined to be severe, or EoEe3. The methods comprise therapies targeted to suppress ALOX15, esophageal dilation, anti-cytokine therapy, for example anti-IL-13 and anti-IL-4 receptor alpha (IL-4Ra) therapies including monoclonal antibodies, for example RPC4046 (anti-IL-13) and dupilumab (anti-IL4Ra), and glucocorticoid therapy, for example therapy with fluticasone, prednisone or budesonide.

Methods of Measuring Gene Expression

In accordance with the methods described here, the expression of one or more genes, or of a panel of genes, is determined in a biological sample, such as an esophageal biopsy sample, obtained from a patient in need of treatment as described herein.

Gene expression may be determined, for example, using a method for detecting and quantitating mRNA expression. Such methods include PCR-based methods such as reverse transcription followed by a polymerase chain reaction (PCR), including a quantitative PCR (qPCR) reaction. The steps may comprise generating a single stranded complementary DNA (cDNA) template from mRNA of the biological sample, e.g., through the performance of a reverse transcription (RT) reaction. Additional steps may include amplification of the cDNA and performance of a method for determining the amount of amplified DNA, for example through the use of labeled probes or DNA intercalating dyes. Additional methods include quantitative PCR performed with a low density array or high density microarray based technique. In embodiments, the methods described here may further comprise one or more steps of converting mRNA to cDNA, converting cDNA to labelled cRNA, e.g., biotinylated cRNA, and hybridizing the labelled cRNA to an oligonucleotide-based DNA microarray chip.

The term “microarray” refers to arrays of probe molecules that can be used to detect analyte molecules, e.g., oligonucleotide probe arrays to measure gene expression. The terms “array,” “slide,” and “chip” may be used interchangeably to refer to oligonucleotide probe arrays. Such arrays may comprise oligonucleotide probes that are synthesized in silico on the array substrate, sometimes referred to as ‘high density’ arrays, or the arrays may be spotted arrays, which tend to have lower densities.

The term “gene expression” refers to the transcription of DNA sequences into RNA molecules. The expression level of a given gene measured at the nucleotide level refers to the amount of RNA transcribed from the gene measured on a relevant or absolute quantitative scale. The measurement can be, for example, an optic density value of a fluorescent signal on a microarray image. Differential expression means that the expression levels of certain genes, as measured at the nucleotide level, are different in different states, tissues, or type of cells, relative to the amount or level of gene expression of a reference gene. In the context of the methods described here, certain genes are differentially expressed in different EoE disease endotypes, for example as set forth in Table 2. The terms ‘level’ and ‘amount’ when used in the context of gene expression are used interchangeably to refer to the amount of gene transcripts in a cell or tissue sample. Where the amount is a relative amount, it is relative to the expression of a reference gene or the expression of a reference set of genes, or the amount of one or more reference oligonucleotides which are exogenously added to a sample. In some embodiments, the reference is an endogenous gene, an exogenously added reference oligonucleotide including an artificial RNA or DNA, a reference gene index, or a target gene index. A reference gene index may be comprised of multiple averaged endogenous control genes such as multiple housekeeping genes. A target gene index may be comprised of multiple averaged genes of interest, such as multiple genes described herein as differentially expressed in an EoE endotype. In some embodiments, more than one reference may be used, for example multiple exogenous control oligonucleotides and multiple endogenous housekeeping genes may be used in the same assay. In embodiments, the reference may be a computed average expression value for one or more genes expressed in a target endotype relative to the expression of the one or more genes in each of the other endotypes, for example EoEe1 relative to EoEe2 and/or EoEe3.

For qPCR based methods, the gene expression may be presented as a delta cycle threshold (Ct) value. The Ct value is defined as the number of PCR cycles required for the fluorescent signal of an amplified product to exceed a background or threshold level. The Ct value is therefore inversely proportional to the amount of the target nucleic acid in the sample. The delta Ct value represents the difference in expression between a target gene and a reference gene calculated as a difference in the Ct values of the target and reference genes in the sample.

In some embodiments, gene expression may further be compared to a second relative parameter such as a nontreated control, a time point (e.g., time zero), or healthy cells, tissues or subjects. Generally, normal healthy esophageal tissue is defined histologically as having zero eosinophils per high power field and no basal layer expansion.

In embodiments, the methods described here may further comprise one or more additional steps of extracting RNA from a biological sample obtained from the patient, for example, an esophageal biopsy sample. The steps may include isolating total RNA and/or mRNA from the biological sample, converting mRNA to cDNA, and performing a PCR-based amplification step. mRNA may be isolated from total RNA, for example using a commercially available kit, such as the RNeasy™ Mini kit (Qiagen), followed by enriching for mRNA using a suitable method, such as oligo(dT) magnetic beads. The mRNA may also be fragmented into short fragments of about 200 base pairs (bp) using a suitable fragmentation buffer. cDNA may be produced from the mRNA, for example, using the fragmented mRNA as a template with random hexamer primers for first-strand cDNA synthesis followed by second-strand cDNA synthesis and purification of the short double-stranded cDNA fragments using standard protocols or a commercially available kit, for example a QIAquick™ PCR purification kit (Qiagen).

Methods of Assigning Subjects to Endotypes

The present disclosure provide methods for assigning a subject to an EoE endotype for treating EoE in the subject. The methods comprise determining the subject's EoE endotype based on the expression of one or more genes or a panel of genes in a biological sample from the subject, and may optionally further comprise determining the expression of the one or more genes or a panel of genes, for example using a method as described above.

The methods for determining a subject's endotype are based on a linear discriminant analysis of the expression of the one or more genes or panel of genes in the biological sample from the subject. In embodiments, the analysis comprises or consists of one or more of ANO1, CRYM, DSG1, PNLIPRP3, TNFAIP6, TSLP, UPK1A, and WDR36; or a panel comprising or consisting of the foregoing genes. In this analysis, group membership, i.e., endotype, is predicted by the continuous variables, i.e., gene expression, which may also be referred to as covariates.

The methods generally relate to analyzing gene expression data taking into account the multidimensional structure of the data. The endotypes identified by the present disclosure are represented such that an input vector comprising the expression of the one or more genes or panel of genes can be assigned to an endotype based on the calculation of a similarity metric between the input vector and each of endotypes defined herein. This analysis determines which endotype is the nearest to, or most similar to, the input vector. In accordance with the present disclosure, the methods comprise utilizing a probability distance metric, e.g., the Mahalanobis distance, and calculating a probability distance between an input vector and the endotypes.

EXAMPLES

The following describes the identification and characterization of three EoE endotypes for use in the treatment of EoE. As discussed in more detail below, each of the three endotypes has characteristic histological and endoscopic features and is associated with distinct clinical characteristics and phenotypes. In addition, we provide eight genes with strong discriminatory power for assignment into one of the three endotypes, which may be used in clinical practice.

Gene expression profiles from a total of 285 esophageal biopsy samples were analyzed, including the discovery cohort of 185 individual subjects from 10 clinical sites and the validation cohort of 100 individual subjects from a single site (Cincinnati cohort) Basic demographic characteristics are detailed in Table 1.

TABLE 1 Basic characteristics of subjects in the discovery and validation cohorts Discovery Validation Cohort (n = 185) Cohort (n = 100) Demographics Age at biopsy (years) 18.3 (8.8-37.1) 10.2 (6.3-15.2) Sex Male 125 (68%) 80 (80%) Female 60 (32%) 20 (20%) Ethnic origin White 170 (92%) 95 (95%) Other 15 (8%) 5 (5%) History of eosinophil gastrointestinal disease Eosinophil esophagitis 185 (100%) 100 (100%) Eosinophil gastritis 4 (2%) 6 (6%) Eosinophil colitis 4 (2%) 0 (0%) Treatment at biopsy Current PPI treatment 62 (34%) 84 (84%) Current topical steroid treatment 95 (51%) 58 (58%) Ongoing diet therapy 97 (52%) 63 (63%) Disease variables at biopsy Peak eosinophil count (per HPF) 15 (1-46) 69 (34-144) EDP total score 242 (67-352) 126 (67-249) HSS total score 0.5 (0.2-0.9) 0.8 (0.5-1.0) EREFS total score 2.0 (0.0-6.0) 1.0 (1.0-2.0)* Data are number (%) or median (IQR). EDP = eosinophil esophagitis diagnostic panel. EREFS = esophagitis endoscopic reference score. ESS = endoscopic severity score. HPF = high-power field. HSS = eosinophil esophagitis histology scoring system. PPI = proton-pump inhibitor. *Simplified ESS used.

Of 185 subjects in the discovery cohort, the age range for all subjects was 3.5 to 69.6 years, with 88 children and 97 adults. The pediatric and adult groups both exhibited a male predominance. Although there were different proportions of pediatric-onset versus adult-onset EoE, the length of time from the initial EoE diagnosis to the biopsy sample collection were similar in pediatric and adult individuals. Peak eosinophil counts ranged from 0 to 174 eosinophils/HPF. Approximately half of the subjects (46.5%, n=86) had active EoE (≥15 eosinophils/HPF) and 20.5% of subjects (n=38) had biopsy specimens without eosinophils. There were no significant differences in peak eosinophil counts between pediatric and adult individuals with active EoE; however, some clinical and endoscopic findings differed by age. Among subjects with active EoE, a larger proportion of pediatric subjects with EoE had an inflammatory phenotype at endoscopy than did adults with EoE (p=0.0008), and significantly more adults had a fibrostenotic phenotype than did pediatric individuals (p<0.0001). There was no significant difference in the EDP score nor HSS score between pediatric and adult individuals with active EoE, whereas the EREFS total score was significantly higher in adults than children (p<0.0001). The distribution of the different types of therapy were similar between the pediatric and adult populations. Among the subjects with a history of swallowed steroid treatment (n=91), there were fewer subjects in the category of steroid-refractory (n=25/91, 27.5%) than steroid-sensitive (n=66/91, 72.5%), and this distribution was similar in pediatric and adult individuals (p=0.82).

The EoE Diagnostic Panel (EDP) showed consistency across sites (Spearman ρ=−0.73−−0.80, p<0.0001) and had similar values across pediatric and adult subjects with EoE (p=0.11) (appendix p 20). To define the relationship among various clinical, endoscopic, and histologic features in relation to the accepted gold standard of assessing disease activity, the esophageal eosinophil level, we evaluated the associations between peak eosinophil counts and disease parameters (EDP, HSS, and EREFS). Using total scores, which represent the overall values of each platform, we found significant correlations between peak eosinophil counts and each platform (FIG. 1A EoE EDP score: Spearman ρ=−0.74, p=1.9E-22; FIG. 1B HSS score: Spearman ρ=0.81, p=1.8E-30; and FIG. 1C EREFS score: Spearman ρ=0.51 p=5.4E-13, respectively).

We subsequently focused on individual components of HSS and EREFS and their relationship with overall EDP. There were relatively strong associations between the EDP's 95 genes and several HSS domains (absolute median Spearman ρ=0.30 [IQR, 0.20-0.40]) (FIG. 2A). In particular, the basal zone hyperplasia (BZH) domain from the distal esophagus exhibited the highest magnitude of correlation with the overall EDP (absolute median Spearman ρ=0.47 [IQR, 0.36-0.61]). There were moderate associations between several EREFS domains and the EDP's 95 genes (absolute median Spearman ρ=0.25; [IQR, 0.11-0.38]) (FIG. 2B). In particular, distal furrows as a single endoscopic feature exhibited the highest magnitude of correlation with overall EDP (absolute median Spearman ρ=0.43 [IQR: 0.32-0.50]). A clustering tree based on the Spearman correlations showed their hierarchic relationships, supporting that the HSS and EREFS features aligned with the biological features associated with EoE.

To test the hypothesis that active EoE demonstrates heterogeneous molecular profiling, we focused on analyzing subjects with active EoE (n=86). Consensus clustering based solely on the EDP was examined to assess stability for a number of potential cluster numbers varying from 2 to 12. This established 3 stable groups (i.e., endotypes, referred to as EoEe1-3) after resampling, as defined by a flat middle part of the consensus CDF (FIG. 3A) and well-defined squares within the consensus matrix (FIG. 3B), in addition to the CLC values and silhouette widths. Although a few clinical sites enrolled a higher number of subjects, there was no significant difference in the distribution of the endotypes at any given site. On the heat map (FIG. 3C) and 3-dimensional plot by principal component analysis (PCA, FIG. 3D), these endotypes were well separated from each other. As a control, multiple biopsies obtained from the same endoscopy maintained each separate endotype (n=4, each endotype) (appendix p 24). Differentially expressed genes between each endotype were identified using a Benjamini-Hochberg false discovery rate of less than 0.05 and greater than 2-fold change. Using this threshold, there were a total of 15 differentially expressed genes in EoEe1, 20 in EoEe2, and 9 in EoEe3 (FIG. 3E). The differentially expressed genes between each endotype are detailed in Table 2 below.

TABLE 2 Differentially expressed genes between each endotype. Fold change Adjusted p value Comparison 1 Comparison 2 EoEe1 EoEe1 vs EoEe2 EoEe1 vs EoEe3 . . . ALOX12 3.3 6.3 6.0 × 10−6 ALOX15 −263.1 −176.4 3.3 × 10−11 APOBEC3A −11.0 −8.3 6.8 × 10−8 CDA 10.4 10.1 7.0 × 10−9 COL8A2 −6.3 −4.3 2.8 × 10−4 CRISP3 307.6 540.6 6.0 × 10−11 ENDOU 6.1 5.6 1.0 × 10−8 EPB41L3 −3.3 −2.0 1.5 × 10−2 GCNT3 −3.6 −2.7 8.0 × 10−7 HILPDA 5.3 7.5 3.9 × 10−9 IGFL1 2.9 2.7 1.4 × 10−3 PLAUR −7.3 −4.4 9.0 × 10−7 SPINK7 4.9 5.3 1.9 × 10−5 UPK1A 5.4 9.1 4.3 × 10−6 ZNF365 2.8 5.5 3.5 × 10−6 EoEe2 EoEe2 vs EoEe1 EoEe2 vs EoEe3 . . . ACTG2 28.9 44.2 2.0 × 10−7 CCR3 24.0 12.9 8.7 × 10−11 CLEC16A 3.3 3.0 1.5 × 10−7 FFAR3 18.0 14.5 1.6 × 10−6 FKBP5 3.2 5.1 1.0 × 10−5 HPGDS 9.1 6.1 1.8 × 10−7 IL4 9.9 10.8 1.0 × 10−8 IL5RA 7.3 7.8 1.1 × 10−6 KRT23 5.4 3.1 1.2 × 10−5 LRRC32 7.1 9.2 1.4 × 10−7 MUC4 7.1 4.2 2.3 × 10−8 NTRK1 8.4 10.6 2.3 × 10−8 PTGFRN 2.5 2.4 1.2 × 10−7 RGS9BP 14.8 21.1 3.4 × 10−6 RUNX2 7.9 6.1 2.7 × 10−8 SAMSN1 13.4 7.2 1.0 × 10−8 TGFB1 2.5 2.2 2.3 × 10−5 TSLP 20.8 14.6 1.6 × 10−6 UPK1B 4.6 3.3 3.1 × 10−4 WDR36 3.2 4.1 8.1 × 10−9 EoEe3 EoEe3 vs EoEe1 EoEe3 vs EoEe2 . . . ACPP −2.8 −3.1 9.2 × 10−9 CITED2 −2.9 −5.7 7.6 × 10−8 CTNNAL1 −5.9 −8.0 2.9 × 10−10 EML1 −10.7 −5.6 1.6 × 10−6 FLG −6.4 −5.5 1.3 × 10−7 GRPEL2 −3.4 −4.0 2.7 × 10−9 MT1M −4.9 −4.5 1.2 × 10−5 PNLIPRP3 −21.8 −12.8 2.4 × 10−9 TSPAN12 −10.1 −5.3 2.0 × 10−7

TABLE 2A Names of differentially expressed genes. Gene abbreviation Name ALOX12 Arachidonate 12-lipoxygenase ALOX15 Arachidonate 15-lipoxygenase APOBEC3A Apolipoprotein B mRNA editing enzyme catalytic subunit 3A CDA Cytidine deaminase COL8A2 Collagen type VIII alpha-2 chain CRISP3 Cysteine-rich secretory protein 3 ENDOU Endonuclease, poly(U)-specific EPB41L3 Erythrocyte membrane protein band 4.1 like 3 GCNT3 Glucosaminy1 (N-acetyl) transferase 3, mucin-type HILPDA Hypoxia-Inducible Lipid Droplet Associated protein IGFL1 IGF-like family member 1 protein PLAUR Plasminogen Activator, Urokinase Receptor SPINK7 Serine-peptidase inhibitor, Kazel Type 7 (putative) UPK1A Uroplakin 1A ZNF365 Zinc finger protein 365 ACTG2 Actin, Gamma 2, Smooth Muscle CCR3 C-C motif chemokine receptor 3 CLEC16A C-type lectin domain containing 16A FFAR3 Free fatty acid receptor 3 FKBP5 FKBP prolyl isomerase 5 HPGDS Hematopoietic prostaglandin D synthase IL4 Interleukin 4 IL5RA Interleukin 5 receptor subunit alpha KRT23 Keratin 23 LRRC32 Leucine-rich repeat containing 32 MUC4 Mucin 4, cell surface associated NTRK1 Neurotrophic receptor tyrosine kinase 1 PTGFRN Prostaglandin F2 receptor inhibitor RGS9BP Regulator of G-protein signaling 9 binding protein RUNX2 Runt-related transcription factor 2 SAMSN1 SAM domain, SH3 domain and nuclear localization signals 1 TGFB1 Transforming growth factor beta 1 TSLP Thymic stromal lymphopoietin UPK1B Uroplakin 1B WDR36 WD repeat domain 36 ACPP Acid phosphatate, prostate CITED2 Cbp/P300 Interacting Transactivator With Glu/Asp Rich Carboxy-Terminal Domain 2 CTNNAL1 Catenin alpha-like 1 EML1 Echinoderm Microtubule-Associated Protein Like 1 FLG Filaggrin GRPEL2 GrpE Like 2, Mitochondrial MT1M Metallothionein 1M PNLIPRP3 Pancreatic lipase-related protein 3 TSPAN12 Tetraspanin 12

The clinical and demographic characteristics (which were not included in the consensus Clustering) for each EoE endotype are described below and in FIG. 4. The 3 endotypes did not differ significantly by their peak eosinophil level, age at time of biopsy collection, gender, race, nor length of time since diagnosis of EoE to biopsy collection. To address whether the identified endotypes have histologic and endoscopic distinctions, we evaluated the association of each endotype with several disease parameters (peak eosinophil counts, EDP, HSS, and EREFS).

Notably, even though there was no difference in peak eosinophil counts among the endotypes, EDP, HSS, and EREFS parameters were associated with endotype classification (FIG. 4A). Among the HSS domains, the BZH and surface epithelial alteration (SEA) domains showed the most significant association with endotype (FIG. 4B). BZH was significantly higher in EoEe2 compared to EoEe1 (p=0.0004), whereas SEA was significantly higher in EoEe3 compared to EoEe1 (p=0.0178). Among the EREFS features, edema, exudates, and furrows showed significant association with endotypes (FIG. 4B).

Endoscopic edema was significantly higher in EoEe2 and EoEe3 compared to EoEe1 (p=0.0096 and p=0.0017, respectively). Occurrence of exudates was significantly higher in EoEe2 compared to EoEe1 (p=0.0317), whereas occurrence of furrows was significantly higher in EoEe3 compared to EoEe1 (p=0.0163).

To determine which clinical characteristics were specific to each endotype, logistic regression analysis was performed. From logistic regression modeling adjusted by age at time of biopsy, we observed strong associations between EoEe1 and normal-appearing esophagus (aOR=4.96 [95% CI 1.12-22.02]; p=0.035), EoEe2 and steroid refractory (aOR=4.91 [95% CI 1.19-20.30]; p=0.028), and EoEe3 and the presence of narrow-caliber esophagus (aOR=10.08 [95% CI 1.96-51.83]; p=0.006) (FIG. 4C). MCA was performed to demonstrate the pattern of the endotypes with regards to clinical features (FIG. 4D). EoEe1 was situated near atopic, pediatric-onset, steroid-sensitivity, and normal endoscopic appearance. EoEe2 was situated close to inflammatory and steroid-resistance, whereas EoEe3 was located near non-atopic, adult-onset, and fibrostenotic phenotypes.

To facilitate potential translation to clinical practice, a clinically reproducible method to identify endotypes was developed. Briefly, stepwise discriminant analysis was performed using the stepwise forward selection method. The discovery cohort was split into training (75%) and testing sets (25%) by random selection, estimation of the variables in the training set and classification of the testing set. Variables were selected from the EDP genes based on a p-value of 0.05 to determine which variables (if any) to include in the discriminating function and whether some of the entered variables should be excluded from the model. The procedure continued until none of the excluded variables had a p-value below the threshold and none of the entered variables had a p-value above the threshold (the stopping rule was applied).

Stepwise linear discriminant analysis using the same 96 EDP genes for active EoE (including EoEe1-3) identified the 8 strongest discriminatory genes as PNLIPRP3, CRYM, WDR36, DSG1, TSLP, TNFAIP6, ANO1, and UPK1A (FIG. 5A). Using these 8 genes, 84 (98%) patients in the discovery cohort were assigned to the appropriate endotype. The three endotypes were discriminated well from each other with good diagnostic accuracy (FIG. 5B).

FIG. 5B shows a canonical biplot based on a linear discriminant analysis to predict endotype classification based on the observed continuous variables, or covariates, of gene expression. Gene expression for each endotype is shown in FIG. 5C-J.

In FIG. 5B, the biplot axes are the first two canonical variables which define the two dimensions providing maximum separation between endotypes. Each canonical variable is a linear combination of the gene expression covariates. The biplot shows how each subject is represented in terms of canonical variables and how each covariate contributes to the canonical variables. The subjects are represented as points on the biplot and are expressed in terms of the first two canonical variables. A 95% confidence level ellipse is plotted for each mean. If two groups differ significantly, the confidence ellipses tend not to intersect. An ellipse denoting a 50% contour is plotted for each group. This depicts a region in the space of the first two canonical variables that contains approximately 50% of the observations, assuming normality. The set of rays that appears in the plot represents the gene expression covariates. For each canonical variable, the coefficients of the gene expression covariates in the linear combination can be interpreted as weights. To facilitate comparisons among the weights, the gene expression covariates are standardized so that each has a mean of 0 and standard deviation of 1. The coefficients for the standardized covariates are referred to as the canonical weights. The larger a covariate's canonical weight, the greater its association with the canonical variable. The length of each ray is a multiple of the canonical weight. The rays emanate from the point (0,0), which represents the grand mean of the data in terms of the canonical variables. The length and direction of each ray in the biplot indicates the degree of association of the corresponding covariate with the first two canonical variables. Thus, the length and direction of each ray indicates the degree of association of gene expression with the canonical variables. Comparison of continuous variables between diagnostic groups was performed by Wilcoxon/Kruskal-Wallis tests (rank sums), using nonparametric comparison for all pairs using Dunn method for joint ranking, as shown in FIG. 5C-5J.

To further validate the eosinophilic esophagitis endotype findings, the same analysis was done in the validation cohort, which comprised 100 patients with active eosinophilic esophagitis. Two separate strategies (consensus clustering based solely on the EDP, and endotype prediction based on the highly discriminatory genes) were used for assignment of endotype, then the validation cohort was compared with the discovery cohort. For validation of endotypes by clustering, 60 patients were segregated into the three endotypes with optimum quality and stability. The remaining 40 patients were assigned to one of the three endotypes based on results of the endotype-prediction algorithm developed with the discovery cohort. The endotypes generated from the validation cohort were like those generated from discovery cohort, in that the gene expression relations among the endotypes were maintained. Furthermore, differences in eosinophilic esophagitis scores and several gene expression levels among the endotypes were similar between validation and discovery cohorts. In the validation cohort, peak eosinophil counts among the three endotypes did not differ. Moreover, consistent with the discovery cohort, the three endotypes showed similar differential trends in clinical and endoscopic findings.

DISCUSSION

Herein, we have dissected EoE disease molecular heterogeneity via the EoE EDP, across a multi-site cohort of subjects associated with CEGIR, and assessed its relevance using a combination of standardized histologic, endoscopic, and clinical platforms. First, we demonstrated that the EDP showed consistency across sites, had similar findings between pediatric and adult EoE subjects, and correlated with esophageal eosinophil levels. Second, we report the existence of three disease endotypes and present evidence for their clinical, histologic and endoscopic significance. Notably, these disease endotypes remained stable using distinct statistical methodology including unsupervised clustering, 3-dimensional principal component analysis (PCA), and cumulative distribution functionality. Third, disease endotypes occurred independent of peak eosinophil counts, underscoring that these findings surpass information provided by eosinophil counts, consistent with prior findings that disease severity and clinical symptoms do not simply reflect eosinophil levels. Fourth, focusing on the unique features of the disease endotypes, we report that EoEe1 has the mildest phenotype, most closely resembling findings seen in healthy tissue of normal biopsies; EoEe2 is characterized by substantial inflammatory changes, type-2 immune responses, and evidence of refractoriness to steroids; and EoEe3 shows a strong association with the presence of a narrow-caliber esophagus, the highest degree of endoscopic and histologic severity, and the lowest expression of epithelial differentiation genes. Fifth, we demonstrate that machine learning can be used to reproducibly separate disease endotypes. Six, we have uncovered a strong association between eosinophilic inflammation, BZH, and endoscopic furrowing. Lastly, we have identified genes that are modulated within each of the endotypes, establishing insight into distinct disease mechanisms

Collectively, the new endotypes described here stratify patients with EoE into subgroups with clinical and therapeutic significance, thereby providing a framework for a precision medicine approach to EoE therapy. For example, because fibrostenotic eosinophilic esophagitis is typically steroid resistant, EoEe2 and EoEe3 likely represent more complex or severe phenotypes and could benefit from treatments in addition to, or even instead of inflammatory control. By uncovering three distinct disease endotypes, each associated with different clinical features, and importantly, molecular pathways and hence mechanisms, our findings provide a framework for distinct prognostic medicine and therapeutic intervention strategies targeted to specific EoE patient subpopulations.

In the present study, we highlighted the strong association between the EDP, the HSS and the EREFS, which are important assessments of disease severity. The overall EDP exhibited the strongest association with the basal zone hyperplasia (BZH) domain. This is consistent with recent work by others suggesting a substantial role of the basal epithelium, particularly related to loss of cellular differentiation. We also found that the endoscopic finding of furrowing stands out as a unique feature, related to transcript changes, particularly those involved in inflammatory responses. Of note, this association was consistent across age groups, even though it is well recognized that clinical and endoscopic features differ between children and adults.

To our knowledge, the present study is the first to characterize endotypes in EoE. EoEe1, representing 35% of patients, had relatively small changes in epithelial differentiation genes, a pauci-inflammatory state, and a greater proportion of normal-appearing esophagus by endoscopy. EoEe2, representing 29% of patients, had particularly high type-2 immune response mechanisms and a steroid-refractory phenotype. EoEe3, representing 36% of patients, had particularly low expression of epithelial differentiation genes and a greater frequency of narrow-caliber esophagus. Endotypes were associated with distinct clinical features, including pediatric-onset versus adult-onset EoE (EoEe1 and EoEe2 vs. EoEe3), atopic versus non-atopic (EoEe1 and EoEe2 vs. EoEe3), normal versus inflammatory versus fibrostenotic endoscopic appearance (EoEe1 vs. EoEe2 vs. EoEe3), and steroid-sensitive versus steroid-refractory (EoEe1 vs EoEe2 and EoEe3).

Our study demonstrates that EoE populations are molecularly and biological distinct and thus has implications for targeting therapy to specific subgroups. EoEe1 is characterized by markedly low expression of ALOX15, suggesting that this gene may be associated with a more severe phenotype and that its suppression and/or suppression of the metabolic products of 15-lipoxygenase may be therapeutic, particularly in subjects with EoEe2 or EoEe3. EoEe2 is characterized by a pronounced inflammatory response, observed by endoscopy and molecular transcript profiling. EoEe2 transcript profiles are notable for relatively high expression of a variety of pro-inflammatory cytokines, especially those characterized by type-2 immune responses (e.g., IL-4 and TSLP). The highest relative expression is seen in the ACTG2 gene, encoding for the actin gamma smooth muscle 2 protein. This protein has been shown to be involved in epithelial cell responses including mesenchymal transition, which is observed in EoE. EoEe3 is enriched for epithelial genes that lose expression, particularly ACPP, CITED2, CTNNALI, EML1, FLG, GRPEL2, MT1M, PNLIPPR3 and TSPAN12. This is the first molecular analysis of the fibrostenotic disease group and provides pathogenic insight and potential points of therapeutic intervention for this difficult-to-treat EoE subgroup. For example, TSPAN12 is tetraspan protein involved in epithelial cell contact, proliferation, and migration and therefor represents a promising therapeutic target.

Our findings also suggest distinct therapeutic strategies for patients falling within each of the endotypes described here. For example, our results indicate that EoEe2 is relatively more responsive to specific anti-type-2 immune therapies, such as anti-IL-4Ra, rather than anti-IL-13 therapy, which shows less differentiation between the three endotypes. Taken together with recent work by others suggesting the presence of a subgroup having a T-helper-2-type inflammatory profile with high expression of TSLP (Lexmond W S, et al. Clin Exp Allergy 2013; 43(8): 902-13), the present findings also indicate that EoEe2 patients represent a subgroup that will be responsive to anti-TSLP therapy. Our findings also indicate that pediatric and adult EoE have comparable pathogenesis and therefor are likely to be similarly responsive to therapy.

A growing body of evidence supports the reassessment of clinical trial designs to include biomarkers reflecting the status of the host response. The molecular endotypes described here show that EoE presents heterogeneously with distinct pathophysiologic profiles that are not distinguishable by esophageal eosinophil counts alone. This has potential value for future clinical trials that could stratify participants prospectively or retrospectively to identify subgroups with distinct responsiveness to therapy. In addition, by deriving machine learning prediction for each endotype, we provided evidence that molecular subtyping of subjects with EoE is feasible in clinical practice and that the technology to produce such tests with automated generation of results exists.

In conclusion, we have established that at least three EoE endotypes exist, each having distinct features of gene expression, histology, and endoscopy, that correlate with clinical features and can be used to stratify patients for personalized therapy.

Study Design and Participants

This study was done within the wider context of CEGIR, which is part of Rare Diseases Clinical Research Network (RDCRN) of the National Institutes of Health. For the discovery cohort, between 2015-2017, children and adults with EoE (≥3 years of age) were enrolled in a multicenter prospective observational study associated with CEGIR. Data were entered and managed by the Data Management and Monitoring Center (DMCC) associated with the RDCRN. Subjects with EoE were defined as having symptomatic esophageal dysfunction and a peak count of 15 or more esophageal eosinophils/HPF. For the validation cohort, children and adults with EoE (≥3 years of age) presenting for standard of care were enrolled in an independent, local Cincinnati cohort. Subjects in the validation cohort were not in the discovery cohort. This study was approved by the institutional review boards of the participating institutions via a central institutional review board at Cincinnati Children's Hospital Medical Center.

Procedures

Transcriptomic signatures in distal esophageal biopsy samples were obtained using an EDP as previously reported. Histologic and endoscopic features were assessed by peak eosinophil counts, the EoE HSS, and EREFS. Clinical features of subjects were captured across sites by the CEGIR questionnaires, which include self-reported demographic, race/ethnicity, exposure assessments, and clinical variables. Clinical phenotypes were defined using metrics previously reported. In addition, steroid-sensitivity/-resistance was determined using a positive/negative response, respectively, to whether swallowed topical steroids have been effective on the basis of pathology (see the REGID [Registry for Eosinophilic Gastrointestinal Disease] questionnaire in the appendix p 29-30).

We evaluated the associations between peak eosinophil counts and disease parameters (EDP, HSS, and EREFS). Furthermore, the EDP (either as a whole or individual genes) was examined for association with the HSS and EREFS features. Spearman correlation analysis was performed between the gene expression levels on the EDP and the HSS and EREFS features.

EDP data from subjects with active EoE were further examined by an unbiased/unsupervised clustering. Consensus clustering was performed by the partition-around-medoids (PAM) algorithm with the Euclid distances metric. Bootstrapping was performed by randomly removing 10% of the data and repeating the clustering a total of 1,000 times. To assess the optimal number of clusters, stability was assessed by the cumulative distribution function (CDF), cluster-consensus (CLC) values, and silhouette width analysis. To identify specific clinical associations for each cluster, logistic regression modeling adjusted by age at biopsy, which could act as potential confounder, was performed. Multiple correspondence analysis (MCA) was also performed to present the pattern of relationships among each cluster and several phenotypes. To develop the algorithm for identifying EoE endotypes, stepwise linear discriminant analysis was performed with a stepwise selection method. The diagnostic performance of the algorithm was evaluated by receiver operator characteristic (ROC) analyses to calculate the area under the curve (AUC). Results were validated with a validation cohort.

The primary objectives of this study were to establish the relationships between various endoscopic, histologic, and molecular features and to determine whether EoE endotypes exist and their significance in terms of histologic, endoscopic, and clinical features.

Statistical Analysis

Data are n (%) or median (interquartile range [IQR]) unless otherwise stated. Statistical analyses were done using the JMP v13.0 (SAS Institute, Cary, NC), GeneSpring GX 12.6 (Agilent Technologies, Santa Clara, CA), GraphPad Prism 7 (GraphPad Software, Inc., San Diego, CA), and the R statistical computing environment (version 3.1.2). Correlation analysis was done using Spearman's rank correlation coefficient followed by Bonferroni adjustment. Adjusted odds ratio (aOR) and 95% confidence intervals (CI) were calculated for each endotype with reference to all other endotypes. To compare differences between endotypes, the Kruskal-Wallis with the Dunn's multiple comparison test were used for nonparametric, continuous variables and the chi-square tests for nonparametric, categorical variables. A significant p value was defined as less than 0.05.

EQUIVALENTS

Those skilled in the art will recognize or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be encompassed by the following claims.

All references cited herein are incorporated herein by reference in their entirety and for all purposes to the same extent as if each individual publication or patent or patent application was specifically and individually indicated to be incorporated by reference in its entirety for all purposes.

The present invention is not to be limited in scope by the specific embodiments described herein. Indeed, various modifications of the invention in addition to those described herein will become apparent to those skilled in the art from the foregoing description and accompanying figures. Such modifications are intended to fall within the scope of the appended claims.

Claims

1. A method for treating eosinophilic esophagitis (EoE) in a subject determined to have a severe form of EoE, the method comprising

subjecting a biological sample from the subject to a method for gene expression analysis,
determining the expression of a panel of genes in the biological sample, the panel comprising ANO1, CRYM, DSG1, PNLIPRP3, TNFAIP6, TSLP, UPK1A, and WDR36,
determining the subject's EoE endotype as mild, intermediate, or severe based on the expression of the panel of genes, and
treating the patient with an EoE therapy comprising one or more of esophageal dilation, anti-cytokine therapy, and glucocorticoid therapy where the EoE endotype is determined to be severe, or
treating the patient with one or more of an anti-cytokine therapy, an anti-TSLP therapy, and an anti-ALOX15 therapy where the EoE endotype is determined to be intermediate, or
treating the patient with one or both of proton pump inhibitor (PPI) therapy and dietary therapy where the EoE endotype is determined to be mild;

2. The method of claim 1, wherein the panel of genes further comprises one or more of the genes set forth in Table 2.

3. The method of claim 1, wherein the panel of genes further comprises one or more of ALOX15, APOBEC3A, CDA, CRISP3, ACTG2, CCR3, FFAR3, IL4, RGS9BP, TSLP, CTNNAL1, EML1, FLG, PNLIPRP3 and TSPAN12.

4. The method of claim 1, wherein the biological sample is an esophageal biopsy sample.

5. The method of claim 1, wherein the determining the expression of a panel of genes in the biological sample is performed using a PCR-based method.

6. The method of claim 1, wherein the determining the subject's EoE endotype based on the expression of the panel of genes is performed by a method comprising linear discriminant analysis.

7. The method of claim 6, wherein the linear discriminant analysis comprises determining a probability distance.

8. The method of claim 7, wherein the distance is the Mahalanobis distance.

9. The method of claim 1, wherein the endotype is further characterized by one or more histologic, endoscopic, or clinical features.

10. The method of claim 9, wherein the one or more histologic features is selected from basal zone hyperplasia (BZH) and surface epithelial alteration (SEA).

11. The method of claim 9, wherein the one or more endoscopic features is selected from the occurrence of edema, exudates, and furrows.

12. The method of claim 9, wherein the one or more clinical features is selected from pediatric onset, adult onset, atopic, non-atopic, steroid sensitivity, steroid refractory, and fibrostenotic.

Patent History
Publication number: 20230374594
Type: Application
Filed: Aug 2, 2023
Publication Date: Nov 23, 2023
Applicant: Children's Hospital Medical Center (Cincinnati, OH)
Inventors: Marc E. Rothenberg (Cincinnati, OH), Tetsuo Shoda (Cincinnati, OH), Ting Wen (Cincinnati, OH)
Application Number: 18/229,323
Classifications
International Classification: C12Q 1/6883 (20060101);