Fecal acute phase proteins in cats with chronic enteropathies

Abstract

Background

Chronic enteropathies (CE) are common in cats and reliable biomarkers that can distinguish different causes and predict or monitor response to treatment are currently lacking.

Hypothesis

To evaluate certain acute phase proteins in feces that could potentially be used as biomarkers in cats with CE.

Animals

Twenty‐eight cats with either inflammatory bowel disease (IBD; n = 13), food‐responsive enteropathy (FRE; n = 3) or small cell gastrointestinal lymphoma (SCGL; n = 12) and 29 healthy control cats were prospectively enrolled.

Methods

Fecal concentrations of haptoglobin, alpha‐1‐acid‐glycoprotein (AGP), pancreatitis‐associated protein‐1 (PAP‐1), ceruloplasmin, and C‐reactive protein (CRP) were measured using Spatial Proximity Analyte Reagent Capture Luminescence (SPARCL) immunoassays before and after initiation of treatment. Cats were treated with diet and/or prednisolone (IBD cats), plus chlorambucil (SCGL cats).

Results

Compared with controls, median fecal AGP concentrations were significantly lower (25.1 vs 1.8 μg/g; P = .003) and median fecal haptoglobin (0.17 vs 0.5 μg/g), PAP‐1 (0.04 vs 0.4 μg/g) and ceruloplasmin (0.15 vs 4.2 μg/g) concentrations were significantly higher (P < .001) in cats with CE. Median fecal AGP concentrations were significantly lower (P = .01) in cats with IBD and FRE (0.6 μg/g) compared with cats with SCGL (10.75 μg/g). A significant reduction was found in CE cats after treatment for median fecal ceruloplasmin concentrations (6.36 vs 1.16 μg/g; P = .04).

Conclusions

Fecal AGP concentration shows promise to differentiate cats with SCGL from cats with IBD and FRE. Fecal ceruloplasmin concentrations may be useful to objectively monitor response to treatment in cats with CE.

Abbreviations

AGP

alpha1‐acid glycoprotein

APPs

acute phase proteins

ARE

antibiotic‐responsive enteropathy

CE

chronic enteropathy

CRP

C‐reactive protein

FCEAI

feline chronic enteropathy activity index

fPLI

feline pancreatic lipase immunoreactivity

FRE

food‐responsive enteropathy

fTLI

feline trypsin‐like immunoreactivity

GI

gastrointestinal

IBD

inflammatory bowel disease

PAP‐1

pancreatitis‐associated protein

SCGL

small cell GI lymphoma

Spec fPL

specific feline pancreatic lipase

α1‐PI

α1‐proteinase inhibitor

1. INTRODUCTION

Chronic enteropathy (CE) is a collective term used to describe a diverse group of gastrointestinal (GI) diseases in cats with increasing incidence over the past decades. ¹ There is no uniformly applied classification scheme for CE, although CE has been categorized based on response to treatment to food‐responsive enteropathy (FRE), idiopathic inflammatory bowel disease (IBD, often used interchangeably with steroid‐responsive enteropathy) and small cell GI lymphoma (SCGL). ¹ , ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ Currently, there is no convincing evidence supporting the presence of true antibiotic‐responsive enteropathy (ARE) in cats. ⁹ SCGL is the most common GI neoplasia in cats and the prevalence has increased during the past 2 decades. ¹⁰ , ¹¹ Progression of IBD to SCGL over months to years has long been suspected based on the frequent coexistence of inflammatory and neoplastic lesions in cats with SCGL, but this progression has not been proven yet. ⁵ , ¹² Currently, establishment of a definitive diagnosis of IBD or SCGL is based on histopathology, often requiring immunophenotyping of lymphocytes ¹³ and/or lymphocyte clonality testing. ⁵ , ¹⁴ , ¹⁵

The ability to predict and monitor disease activity and treatment efficacy with noninvasive tools, such as biomarkers, is highly desirable in veterinary medicine. In 2001, the Biomarkers Definitions Working Group defined a biomarker as a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention (Group BDW 2001). In inflammatory or neoplastic conditions, proteins, such as acute phase proteins (APPs), and other molecules are commonly released by affected cells or in response to tissue dysfunction and some of them may serve as biomarkers. ¹⁶ In cats with CE, fecal biomarkers seem to have several advantages as sample collection is easy, serial evaluation can be carried out, and direct contact with the inflamed or neoplastic intestine may increase their diagnostic utility compared with serum biomarkers that can be affected by extraintestinal disorders.

Although development and validation of assays for the measurement of certain APPs in cats' feces have been reported, ¹⁷ , ¹⁸ , ¹⁹ to the authors' knowledge, only α1‐proteinase inhibitor (α1‐PI) has been evaluated in cats with CE. ²⁰ Our hypothesis was that certain APPs could be useful as a biomarker to diagnose, differentiate, and monitor cats with CE. Therefore, the aim of this study was to evaluate fecal concentrations of haptoglobin, alpha1‐acid glycoprotein (AGP), pancreatitis‐associated protein (PAP‐1), ceruloplasmin, and C‐reactive protein (CRP) in cats with CE and compare them with those of healthy control cats. A secondary aim of this study was to compare fecal concentrations of the APPs mentioned above between cats with SCGL and those with IBD or FRE. A third aim of this study was to evaluate changes in the fecal concentrations of these APPs before and during treatment in cats with CE.

2. MATERIALS AND METHODS

2.1. Ethics approval

The study protocol was reviewed and approved by the Animal Ethics Committee of the University of Thessaly, Greece (AUP number: 115/6.10.2020). The owners of each cat enrolled in the study signed an informed owner consent form.

2.2. Study population

Cats presenting from September 2020 to October 2022 to the Clinic of Medicine of the University of Thessaly and to the referral hospital Animal Medical Center in Athens, Greece, for persistent or intermittent GI signs of at least 3 weeks' duration were considered for inclusion into the study. Cats were included in the study if they (1) had clinical signs compatible with CE (ie, diarrhea or loose stools of >3 weeks' duration; vomiting ≥2 times per month for ≥2 months, weight loss, hyporexia, and/or anorexia); (2) had no other diseases that could explain the cat's clinical signs; and (3) had no history of administration of antibiotics, corticosteroids, or other immunosuppressive agents for at least 3 weeks before presentation. However, cats with a clinical presentation compatible with CE that had concurrent diseases that were well controlled with appropriate treatment were eligible for enrollment. Cats with concurrent pancreatitis, and/or inflammatory liver disease were also enrolled into the study. Adult cats (at least 1 year of age) without a prior or current history of chronic GI signs or antibiotic and corticosteroid administration for at least 1 year, with no abnormal findings on physical examination and with negative PCR for Trichomonas foetus were recruited as healthy controls.

2.3. Study design and fecal sample collection

All cats with CE underwent the same diagnostic investigations, consisting of a collection of a complete clinical history, physical examination, a complete blood count (CBC), serum biochemistry with electrolytes, serum concentrations of cobalamin, folate, feline pancreatic lipase immunoreactivity (measured as Spec fPL), ²¹ , ²² and feline trypsin‐like immunoreactivity (fTLI), ²³ urinalysis, fecal parasitology using satirized zinc sulfate floatation, abdominal ultrasound, PCR for Tritrichomonas foetus, upper and lower GI endoscopy, and biopsy collection for histopathology and immunochemistry examination. The feline chronic enteropathy activity index (FCEAI) ³ was also calculated for all cats with CE. Reexamination along with CBC, serum biochemistry with electrolytes, serum concentrations of cobalamin, folate, Spec fPL, and fTLI were repeated at least at day 30 and day 90 after endoscopy.

Endoscopy was performed by a veterinarian with experience in endoscopy using a flexible videogastroscope (Pentax EG2490Κ, PENTAX Medical EMEA) with an outer diameter of 8 mm and a 2.4 mm biopsy channel. Biopsies were obtained using 2.2 mm forceps. Before endoscopy, food was withheld for 12 hours, and 4 to 5 warm water enemas were performed to clean the colon while the cat was under general anesthesia. Feces were collected at baseline (T0; before endoscopy) and day 90 (T1) after endoscopy. A single fecal sample was also collected from each cat in the control group. All fecal samples were placed in Eppendorf tubes and stored at −80°C, until analyses.

2.4. Histopathology and immunohistochemistry

Formalin‐fixed, paraffin‐embedded endoscopic biopsy samples were stained with hematoxylin and eosin (H&E) for histopathologic examination. All the endoscopic biopsy samples were examined by 2 veterinary specialists in small animal GI histopathology. Findings were reported descriptively and numerically scored according to the WSAVA histopathologic scoring system. ²⁴ Both inflammatory (presence of lymphocytes, plasma cells, eosinophils, neutrophils, and macrophages in the lamina propria), as well as morphological features (eg, surface epithelial injury, crypt lesions such as dilatation, distortion, or hyperplasia, atrophy and fibrosis) were assessed histologically and assigned a score (normal = 0, mild = 1, moderate = 2, and marked = 3). Immunohistochemistry was conducted using a stepwise approach. Staining for T‐, B‐, and natural killer (NK) cell markers (CD3, CD79a, granzyme B, respectively) was performed if deemed necessary based on the pathologist's discretion depending on the results of H&E staining (ie, number, size, and distribution of mucosal lymphocytes). Based on pathologist discretion clonality testing (PARR) was also performed.

2.5. Assays for determination of fecal APPs concentrations

Feline ceruloplasmin and CRP were measured using 96‐well SPARCL assays from Life Diagnostics, Inc (West Chester, PA). Feline AGP and haptoglobin were measured using single tube SPARCL assays from Veterinary Biomarkers, Inc (West Chester, PA). A canine PAP‐1 assay from Veterinary Biomarkers was used to measure feline PAP‐1. All biomarkers were measured in fecal extracts. The assays used in the study are commercially available and listed on the Life Diagnostics website. Internal laboratory validation has shown that they measure biomarker levels reproducibly, and that parallelism (dilutional linearity) is found for native samples (serum, plasma, or fecal extracts). The approach that has been used to validate the SPARCL assays is described by Josipa Kules. ²⁵ Use of the cat AGP, ceruloplasmin, and haptoglobin assays is described by Liu. ²⁶ The dog PAP‐1 assay, which cross‐reacts with cat, is described by O'Reilly. ²⁷

Fecal extracts were prepared by mixing ~100 mg of feces with 9 weight‐volumes of 150 mM NaCl, 10 mM Tris, 1 mM EDTA, pH 7.4, in 1.5 mL microcentrifuge tubes. After vertexing several times over a period of 30 minutes, the suspensions were centrifuged for 5 minutes at 15 000 rpm. The supernatants were saved and stored as aliquots at −20°C until analyses.

2.6. Treatments in cats with CE

All cats received the same preventative antiparasitic treatment (Broadline, Boehringer Ingelheim) for the duration of the study. After endoscopy and biopsy collection, owners were advised to feed their cats the same hydrolyzed protein diet (Anallergenic, Royal Canin). In addition, all cats received cobalamin injections (250 μg/cat SQ every 2 weeks) for the duration of the study regardless of their serum cobalamin concentration. Cats with a diagnosis of IBD were treated with prednisolone (starting dose 2 mg/kg, once daily, with gradual tapering), while cats with a diagnosis of SCGL were treated with prednisolone (same dose protocol as for cats with IBD) plus chlorambucil (2 mg/cat every 2nd or 3rd day). Immunomodulatory treatment was started 14 to 30 days after endoscopy in all cats depending on the severity of the disease and the time needed for histopathology results.

2.7. Statistical analysis

Fisher's exact test was performed to determine the significance of each clinical sign in terms of predicting a diagnosis of IBD or FRE vs SCGL. Continuous data were tested for normality using the Shapiro‐Wilk test. Normally distributed data were compared using t‐tests. Mann‐Whitney tests were used to compare non‐normally distributed data. Fecal APPs were compared between groups at T0 using a Mann‐Whitney test. Fecal APPs before and during treatment were compared using a Wilcoxon test. Spearman correlation (r) was used to assess the relationship between fecal APPs and FCEAI. Significance was set at P < .05. All data were analyzed using a commercially available statistical software package (GraphPad Prism 8).

3. RESULTS

3.1. Signalment

A total of 28 cats with CE were included in the study. Of these 21 were DSH, 3 DLH, 2 Birman, 1 Persian, and 1 Bengal. Twenty‐nine healthy cats were included as a control group and all of these were DSH. Mean body weight, median age, and sex of these cats are displayed in Table 1.

TABLE 1.

Characteristics of healthy control (HC) cats, CE cats, and subgroups IBD + FRE and SCGL.

Variable	HC (n = 29)	CE (n = 28)	P‐values	IBD + FRE (n = 13 + 3)	SCGL (n = 12)	P‐values
Male	12	16		11	5
Female	17	12		5	7
Age (years)	3 (1‐12)	11 (3‐15)	<.0001	10.38 (3‐15)	9.83 (3‐15)	.68
Body weight (kg)	4.25 (2.5‐8)	4.9 (2.8‐8.5)	.71	5.15 (2.8‐8.5)	4.22 (2.58‐6)	.14
Clinical score (FCEAI)	—	9.1(5‐8)		8.81 (4‐13)	9 (4‐15)	.86
Decreased activity	—	16		9	7
Decreased appetite	—	19		11	8
Vomiting	—	25		14	11
Diarrhea	—	20		10	10
Weight loss	—	22		11	11
Concurrent diseases
Pancreatitis	—	5		3	2
Hyperthyroidism	—	2		1	1
CKD II	—	2		2
Stomatitis	—	2		2
Idiopathic hypertension	—	1			1

3.2. Final diagnoses

Based on histopathology and immunochemistry results, a total of 13 of the 28 diseased cats were diagnosed with IBD, 12 were diagnosed with SCGL. The remaining 3 cats were diagnosed with FRE based on the response to a hydrolyzed‐protein diet.

3.3. Concurrent diseases

Of the 28 cats with CE included into the study, 5 cats (3 with IBD and 2 with SCGL) had concurrent pancreatitis (based on an increased serum Spec fPL concentration ≥5.4 μg/L), 2 (1 with IBD and 1 with SCGL) had hyperthyroidism under controlled therapy (based on an increased serum T4 concentration ≥3.82 μg/day), 2 (both with IBD) had IRIS stage II chronic kidney disease (based on serum creatinine concentration between 1.6 and 2.8), 2 (both with IBD) had stomatitis (based on physical examination findings), 1 with SCGL had idiopathic hypertension and was being treated with amlodipine, and 1 cat with IBD had also asthma and was being treated with inhaled fluticasone (Table 1).

3.4. Clinical signs

The most common presenting clinical sign in cats with CE was vomiting (25/28; 89.3%), followed by weight loss (22/28; 78.6%), diarrhea (20/28; 71.4%), and hyporexia/anorexia (19/28; 67.8%%) (Table 1). In cats with IBD or FRE, the most common clinical signs were vomiting (14/16, 87.5%), hyporexia/anorexia (11/16, 68.8%), weight loss (11/16, 68.8%), and diarrhea (10/16, 62.5%). In cats with SCGL, the most common clinical signs were vomiting (11/12, 91.7%), weight loss (11/12, 91.6%), diarrhea (10/12, 83.3%), hyporexia/anorexia (8/12, 66.7%) (Table 1). Based on Fisher's exact test there was no association between any of these clinical signs and the group of cats. No significant difference was found in FCEAI between IBD or FRE and SCGL groups at T0 (Table 1).

3.5. Treatment and follow‐up

All cats were switched to the same hydrolyzed protein diet after endoscopy (Anallergenic Royal Canin). However, only 11 out of 28 cats actually consumed the diet exclusively, while the remaining 17 continued to consume their previous diet because of their refusal to consume the hydrolyzed diet.

Of the 28 cats with CE included in the study, follow‐up information was available for 17 cats at T1 (Day 90), 9 with IBD, 3 with FRE, and 5 with SCGL. Of these 9 IBD cats, by the time when the feces were collected, 5 were treated with prednisolone for 30 to 45 days plus hydrolyzed diet for 90 days and the remaining 4 were treated only with prednisolone for 30 to 45 days. Of the 5 cats with SCGL, by the time when feces were collected, 3 received treatment with prednisolone plus chlorambucil for 30 to 45 days plus hydrolyzed diet for 90 days and 1 received treatment with prednisolone plus chlorambucil for 30 days. In the remaining 1 cat, the owner refused to start treatment because the cat experiencing remission of clinical signs with a hydrolyzed diet only during the study, but clinical signs relapsed after 5 months of the initial diagnosis and owner decide to start prednisolone with chlorambucil then. Mean FCEAI was 9.1 (SD 2.5) at T0 and 3.5 (SD 1.8) at T1. Median FCEAI was significantly decreased in cats at T1 compared with T0 (P < .001). No significant difference in FCEAI was found between IBD and SCGL groups at T1.

3.6. Fecal APP concentrations in cats with CE and healthy cats at baseline

Fecal AGP concentrations were significantly lower (median 1.8 μg/g, range 0.2‐77.07 vs median 25.1 μg/g, range 0.07‐133.7; P = .003) and fecal haptoglobin (median 0.5 μg/g, range 0.07‐55.99, vs median 0.17 μg/g, range 0.03‐1.78), PAP‐1 (median 0.4 μg/g, range 0.04‐7.64 vs median 0.04 μg/g, range 0‐4.23) and ceruloplasmin (median 4.2 μg/g, range 0‐29 vs median 0.15 μg/g, range 0‐6.68) concentrations were significantly higher (P < .001 for all) in cats with CE compared with controls (Figure 1). After repeating the analysis after excluding cats with concurrent pancreatitis, fecal AGP concentrations remained significantly lower (median 4.16 μg/g, range 0.2‐77.07 vs median 25.1, range 0.07‐133.7 μg/g; P = .01) and fecal haptoglobin (median 0.49 μg/g, range 0.07‐55.99 vs median 0.17 μg/g, range 0.03‐1.75), PAP‐1 (median 0.29 μg/g, range 0.04‐7.64 vs median 0.04 μg/g, range 0‐4.23) and ceruloplasmin (4.05 μg/g, range 0‐29 vs 0.15 μg/g, range 0‐6.68) concentrations remained significantly higher (P < .001 for all) in cats with CE compared with controls. No significant differences found in fecal concentrations of CRP between cats with CE and controls (Figure 1).

FIGURE 1.

(A) Fecal AGP concentrations in cats with CE and healthy control (HC) cats. Fecal AGP concentrations were significantly lower in cats with CE (median 1.8 μg/g, range 0.2‐77.07) compared with HC cats (median 25.1 μg/g, range 0.07‐133.7; P = .003), but there was a large degree of overlap between groups. (B) Fecal haptoglobin concentrations in cats with CE and healthy control (HC) cats. Fecal haptoglobin concentrations were significantly higher in cats with CE (median 0.5 μg/g, range 0.07‐55.99) compared with HC cats (median 0.17 μg/g, range 0.03‐1.78; P < .001), but there was a large degree of overlap between groups. (C) Fecal PAP‐1 concentrations in cats with CE and healthy control (HC) cats. Fecal PAP‐1 concentrations were significantly higher in cats with CE (median 0.4 μg/g, range 0.04‐7.64) compared with HC cats (0.04 μg/g, range 0‐4.23; P < .001), but there was a large degree of overlap between groups. (D) Fecal ceruloplasmin concentrations in cats with CE and healthy control (HC) cats. Fecal ceruloplasmin concentrations were significantly higher in cats with CE (median 4.2 μg/g, range 0‐29) compared with HC cats (median 0.15 μg/g, range 0‐6.68; P < .001), but there was a large degree of overlap between groups. (E) Fecal CRP concentrations in cats with CE and healthy control (HC) cats. Fecal CRP concentrations were not significantly different in cats with CE (median 2.24 μg/g, range 0.03‐207.8) compared with HC cats (median 1.71 μg/g, range 0.81‐36.09; P = .2).

3.7. Fecal APP concentrations in cats with IBD compared with cats with SCGL at baseline

Fecal AGP concentrations were significantly lower (P = .01) in cats with IBD or FRE (median 0.6 μg/g, range 0.02‐77.07) compared with cats with SCGL (median 10.75 μg/g, range 0.23‐59.63) (Figure 2). When the 3 cats with FRE were excluded, median fecal AGP concentrations were also significantly lower (P = .02) in cats with IBD (median 0.7 μg/g, range 0.2‐77.07) compared with cats with SCGL (median 10.75 μg/g, range 0.23‐59.63). No significant differences were found in fecal concentrations of haptoglobin, PAP‐1, ceruloplasmin, and CRP between cats with IBD or FRE and cats with SCGL (Figure 2). When the 5 cats with pancreatitis were excluded, fecal AGP concentrations were remained significantly lower (P = .02) in cats with IBD or FRE (median 0.64 μg/g, range 0.2‐77.07) compared with cats with SCGL (median 11.27 μg/g, range 0.23‐59.63). No significant differences were found in fecal concentrations of haptoglobin, PAP‐1, ceruloplasmin, and CRP between cats with IBD or FRE and cats with SCGL after the exclusion of cats with pancreatitis.

FIGURE 2.

(A) Fecal AGP concentrations in cats with IBD or FRE and cats with SCGL. Fecal AGP concentrations were significantly lower in cats with IBD or FRE (median 0.6 μg/g, range 0.02‐77.07) compared with cats with SCGL (median 10.75 μg/g, range 0.23‐59.63; P = .01). (B) Fecal haptoglobin concentrations in cats with IBD or FRE and cats with SCGL. Fecal haptoglobin concentrations were not significantly different in cats with IBD or FRE (median 0.6 μg/g, range 0.18‐55.99) compared with cats with SCGL (median 0.5 μg/g, range 0.07‐8.8; P = .95). (C) Fecal PAP‐1 concentrations in cats with IBD or FRE and cats with SCGL. Fecal PAP‐1 concentrations were not significantly different in cats with IBD or FRE (median 0.18 μg/g, range 0.04‐7.1) compared with cats with SCGL (median 1.45 μg/g, range 0.04‐7.64; P = .27). (D) Fecal ceruloplasmin concentrations in cats with IBD or FRE and cats with SCGL. Fecal ceruloplasmin concentrations were not significantly different in cats with IBD or FRE (median 5.98 μg/g, range 0.55‐29) compared with cats with SCGL (median 3.63 μg/g, range 0‐23.84; P = .59). (E) Fecal CRP concentrations in cats with IBD and cats with SCGL. Fecal CRP concentrations were not significantly different in cats with IBD or FRE (median 1.57 μg/g, range 0.03‐9.09) compared with cats with SCGL (median 2.66 μg/g, range 0.09‐207.8; P = .22).

3.8. Fecal APP concentrations in cats with CE before and during treatment

A significant reduction from T0 to T1 was found for fecal ceruloplasmin concentrations in all 17 cats with CE (median 6.36 μg/g, range 0.23‐29 vs median 1.16 μg/g, range 0‐25.78; P = .04), as well as in cats with CE after exclusion of the 3 FRE cats (median 7.56 μg/g, range 0.23‐29 vs median 0.96 μg/g, range 0‐5.66; P = .01). No correlations were found between the fecal APP concentrations and clinical scoring. No significant differences were found in fecal concentrations of haptoglobin, PAP‐1, ceruloplasmin and CRP between 12 cats with IBD or FRE and 5 cats with SCGL. A significant reduction was found for fecal ceruloplasmin concentrations in 9 cats with IBD (mean 10.84 μg/g, SD 9.19 vs mean 1.99 μg/g, SD 1.99; P = .01; Figure 3).

FIGURE 3.

Fecal ceruloplasmin in 17 cats with CE before (T0) and after treatment (T1). Fecal ceruloplasmin concentrations were significantly decreased from T0 (median 6.36 μg/g, range 0.23‐29) to T1 (1.16 μg/g, range 0‐25.78; P = .04).

4. DISCUSSION

CE is commonly seen in cats and can be associated with a high morbidity and mortality. ³ , ⁴ In our prospective study we investigated whether certain fecal APPs (ie, haptoglobin, AGP, PAP‐1, ceruloplasmin, and CRP) could serve as biomarkers to diagnose, differentiate, and monitor cats with CE. To the authors' knowledge, this is the first study evaluating concentrations of these APPs in feces of cats with CE. We found that cats with CE had increased fecal haptoglobin, PAP‐1, and ceruloplasmin concentrations and decreased AGP concentrations compared with healthy cats. However, while there was a statistically significant difference in these markers between groups, the overlap of the actual values was broad, which would limit their potential usefulness as diagnostic markers. CRP concentrations were not different between cats with CE and healthy cats; this was not unexpected, however, because CRP has previously been reported not to be involved in the acute phase response in cats. ²⁸ , ²⁹

Interestingly, fecal AGP concentrations were significantly lower in cats with IBD or FRE compared with cats with SCGL, and between cats with IBD and SCGL with a limited degree of overlap. This finding is of particular interest as the differentiation between cats with IBD and SCGL poses a significant clinical problem that has yet to be reliably solved. This is complicated further by the possibility of IBD progressing to SCGL over time. Thus, a biomarker that could distinguish IBD from SCGL is highly desirable because many cats could be diagnosed with IBD via histopathology and overtime may develop SCGL. Ideally, in these cases, endoscopy and histopathology should be repeated but this happens rarely in clinical practice.

AGP is a glycoprotein synthesized mostly by hepatocytes and lymphocytes released into the bloodstream, ²⁹ , ³⁰ but its biological function has not been completely defined. However, an immunomodulatory and anti‐inflammatory role has been suggested as it can inhibit lymphocyte proliferation and platelet aggregation, down‐regulate neutrophil responsiveness, stimulate IL‐1R antagonist secretion by macrophages, and modulate the production of anti‐inflammatory cytokines by peripheral blood leucocytes. ³¹ In cats, based on previous data, AGP is a major positive APP. ²⁸ In cats with neoplastic conditions, serum AGP is increased inconsistently among studies. ³² In 1 study involving cats with various types of lymphoma, AGP concentration in serum was not correlated with either remission duration or survival time. ³³ However, cats with lymphoma had significantly higher serum AGP concentrations than healthy cats. ³³ In another study in cats, AGP concentrations in serum were higher in cats with lymphoma than in healthy cats and concentrations decreased significantly after antineoplastic treatment, when the disease was in remission. ³⁴ However other studies on feline serum AGP concentration have focused on infectious diseases, especially FIP. ³⁵

In humans, serum AGP concentrations have been correlated with IBD activity; however, the relatively long half‐life (5 days) has limited its utility in clinical practice. ³⁶ Fecal AGP seems more promising, and in a study in humans fecal AGP had good diagnostic accuracy for the differentiation of patients with IBD from healthy controls. ³⁷ In the same study, the high degree of correlation of fecal AGP with endoscopic findings suggests that AGP may be a promising biomarker for the differentiation of active from inactive IBD. ³⁷

In contrast to previous studies evaluating serum AGP in cats with various types of neoplasia ³³ , ³⁵ and fecal AGP in humans with IBD, ³⁷ in our study fecal AGP was significantly lower in cats with CE compared with controls. Moreover, fecal AGP was significantly higher in cats diagnosed with SCGL compared with those diagnosed with IBD. Our results showed that in IBD, AGP in feces is decreased, although based on previous data it is a positive acute phase protein in serum. The reason for this finding is unknown. In 1 study in humans with sepsis, serum AGP concentrations were found to be decreased in nonsurvivors compared with survivors but the reason for this finding remains unknown. ³⁸ In pigs, serum concentrations of AGP have also been investigated and have been reported to be associated with both positive and negative acute phase reactions in these species. ³⁹ , ⁴⁰ Therefore, it might be possible that AGP in cats might have negative acute phase reaction in some types of inflammation in cats. Moreover, in pigs, serum AGP concentrations are increased during birth and gradually decrease during growth until the adult values. ⁴¹ One possibility is that fecal AGP concentrations were decreased in cats with CE compared with healthy cats because of the age difference between these 2 groups. However, in our study, cats with IBD (which had the lowest serum AGP concentrations) were significantly younger than cats with SCGL. Further studies are needed to investigate fecal AGP concentrations in cats with IBD and SCGL.

Another important finding of our study was that fecal ceruloplasmin was significantly decreased in cats with CE during treatment at the same time when clinical scoring was significantly decreased. This might suggest that fecal ceruloplasmin could serve as a potential biomarker for assessing response to treatment in cats with CE, but more studies are needed to confirm these findings. Ceruloplasmin is an a2‐glycoprotein and a minor positive APP in dogs and cats. ²⁸ , ²⁹ It is produced by hepatocytes and activated macrophages, has ferroxidase with bactericidal activities, and binds most of the free serum copper. In the inflammatory process, it acts as a protectant against damage by free iron, which promotes free radical oxidation. ⁴² In a study in humans with CE, serum ceruloplasmin concentrations were significantly higher in patients with Crohn's disease and patients with celiac disease compared with the healthy controls, while data evaluating fecal ceruloplasmin in humans with IBD are lacking. ⁴³ Although it has been widely evaluated in several inflammatory diseases in dogs and cats, data evaluating ceruloplasmin in serum or feces of dogs or cats with CE are lacking.

Haptoglobin is a protein produced by the liver and attached to a certain type of hemoglobin. In our study, fecal haptoglobin concentrations were higher in cats with CE compared with healthy controls, but there was a significant degree of overlap between groups. In addition, no difference was found in fecal haptoglobin concentrations between cats with IBD or FRE and cats with SCGL or between baseline and at 3‐months after the initiation of treatment. In human medicine, 1 recent study showed that the fecal concentrations of haptoglobin were significantly increased in children with IBD compared with healthy controls and that children with ulcerative colitis had higher fecal haptoglobin than those with Crohn's disease. ⁴⁴ In 2 additional studies, fecal haptoglobin (in the form of hemoglobin‐haptoglobin complexes) was found to be increased in patients with colorectal cancer compared with controls, and in another study, fecal haptoglobin concentration was higher in patients with colorectal cancer compared with controls. ⁴⁵ , ⁴⁶ , ⁴⁷ In dogs with inflammatory CE, fecal haptoglobin concentrations were significantly higher, compared with controls and dogs with FRE. ²⁷ In our study, fecal haptoglobin concentrations failed to separate cats with IBD or FRE and cats with SCGL or cats with IBD and SCGL. Because of the small number of cats (n = 3) with FRE, statistical comparisons of FRE cats with IBD and SCGL were not possible.

In addition, we found a significant difference in fecal PAP‐1 concentrations between cats with CE and healthy cats, but there was a significant degree of overlap between groups. Also, no significant differences were found between cats with IBD or FRE and those with SCGL or between baseline and at 3‐months after the initiation of treatment. No correlation between PAP‐1 and the disease activity (the clinical scoring) was found. PAP is part of the group of proteins encoded by the regenerating islet‐derived (REG) gene family, and many of them are associated with epithelial inflammation. Human PAP is expressed in the GI tract and is mainly produced by Paneth cells of the jejunum and ileum and by the goblet cells and enterocytes in the colon and is upregulated in patients with IBD. PAP has a variety of activities, which include anti‐apoptotic, anti‐inflammatory, antibacterial and proliferative effects, maintaining host‐bacterial homeostasis in the mammalian gut. ⁴⁸ , ⁴⁹ In addition to increased intestinal PAP mRNA in human IBD patients, ⁵⁰ , ⁵¹ increased serum levels of PAP have also been reported in people with active IBD, with levels correlating with clinical and endoscopic disease severity, all of which make PAP a promising biomarker. ⁴⁹ In a recent study, serum PAP levels were higher at disease relapse than in inactive Crohn's disease in human patients and those with active disease also had increased levels of PAP compared with people in remission at the visit before disease flare. In the same study, no differences in fecal PAP concentrations in IBD human patients regardless of disease activity were found. ⁵² In a study in dogs, fecal PAP concentrations were significantly increased in dogs with CE compared with healthy controls, and were significantly higher in dogs with inflammatory CE compared with FRE dogs. ⁴⁸ Because of the small number of cats cases (n = 3) cats with FRE, statistical comparisons of FRE cats were not compared separately with cats with IBD and SCGL were not possible.

Our study had some limitations. One limitation is the relatively small number of cats with IBD, FRE, and SCGL and the even smaller number of cats that completed the 3‐month follow‐up period. This could have led potentially to type 2 statistical error. However, this was a prospective study, with relatively strict inclusion criteria and a 3‐month follow‐up period, and therefore it is challenging to enroll larger numbers of animals. Another limitation could be potentially the enrollment of cats with concurrent diseases (mainly pancreatitis), and this may have affected the fecal concentrations of at least some of the APPs. It is currently unknown whether and how these diseases affect the concentrations of these APPs in feces, although after exclusion of cats with pancreatitis and the cats with stomatitis the results did not change. In the authors' institutions, many (if not the majority) of cats present with more than 1 concurrent diseases. Therefore, we feel that the feline population enrolled in our study reflects what we commonly see in practice. Another important point to consider is that in our study APPs were measured in feces and not in serum samples. Fecal APPs are believed to be mainly produced by the inflamed intestinal mucosa and therefore reflect intestinal inflammation, while inflammation of other organs such as the pancreas or in the cases of stomatitis, might not lead to a significant increase of in fecal APPs. Also, cats included in the control group were not age‐matched with the cats in the CE group. This might have affected the results, but it is currently unknown whether age affects fecal concentrations of these biomarkers. Finally, cats in the control group were considered clinically healthy based on the 1‐year history and clinical examination but unfortunately no laboratory testing was available.

Another limitation of the study is that food trial in these cats was held with only 1 diet and some of these cats refused to eat it. It is possible that some of these cats diagnosed as having IBD (and respond to steroids) while they might have been FRE if more dietary trials were performed. Based on the study design we were not able to take time to do more food trials. However, this is a common problem that we often face in clinical practice. Many of these cats are presented with severe clinical signs (severe weight loss and low body condition score, severe diarrhea, vomiting, etc) and trying several diets sequentially (often ideally including home‐made diets) is not necessarily the best option as it might take months to see a response and their condition might have severely deteriorated by that time. In addition, some of the cats included in our study (3/13 cats with IBD) had been on several hypoallergenic diets by their referring veterinarian, with no or only temporary improvement in their clinical signs. Moreover, we would like to keep a homogeneity in our cats and different clinical diets would have led to mix conclusions etc if there is an impact of each diet in fecal APPs. The purpose of this study was to find biomarkers that could help us distinguish cats with SCAL from cats with inflammatory enteropathies, based on that misclassification of some FRE cats will not influence these results. Another purpose of the study was to evaluate fecal APPs as biomarkers for treatment efficacy, all of these cats improved with therapy and ceruloplasmin was also decreased so misclassification of some cats could not influence the results.

Another limitation of the study is that a small number of cats (2 with IBD and 1 with SCAL) had received prednisolone 3 weeks before enrollment, and it is possible that this has affected the fecal concentrations of at least some of the APPs evaluated in our study. It is known that APP expression is controlled by a combination of hormones, in particular glucocorticoids, and by proinflammatory cytokines, such as interleukin 1 (IL‐1), TNF‐a, and IL‐6. ⁵³ , ⁵⁴ In dogs, a significant increase in serum haptoglobin concentration has been demonstrated after administration of different doses of glucocorticoids, although concentrations of CRP and ceruloplasmin were not affected. ⁵⁵ Some of the cats in our study had received prednisolone up to 2 mg/kg/day up to 3 weeks before enrollment. Moreover, most of these cats received prednisolone as part of their treatment for CE and this could explain, at least in part, the fact that AGP, haptoglobin, and PAP‐1 were not significantly affected during treatment despite the fact that FCEAI was significantly decreased. Interestingly, fecal ceruloplasmin was significantly decreased in our cats, despite the fact that most of them were treated with prednisolone, potentially indicating that ceruloplasmin is not affected by glucocorticoids.

In conclusion, this is the first study evaluating fecal concentrations of AGP, haptoglobin, ceruloplasmin, PAP‐1, and CRP in cats with CE. Fecal AGP concentrations might show promise in the differentiation between cats with IBD and cats with SCGL. Fecal ceruloplasmin concentrations may be useful to objectively monitor response to treatment in cats with CE.

CONFLICT OF INTEREST DECLARATION

The feline CRP and ceruloplasmin 96‐well assays are listed for sale on the Life Diagnostics website. The feline AGP and haptoglobin, and canine PAP1 assays are listed on the Veterinary Biomarkers website. Dr. Chadwick has full ownership of Life Diagnostics and partial ownership of Veterinary Biomarkers. The study design, analysis of the results and discussion were performed by the other authors and no conflict of interest exists.

OFF‐LABEL ANTIMICROBIAL DECLARATION

Authors declare no off‐label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION

Approved by the Animal Ethics Committee of the University of Thessaly, Greece (AUP number: 115/6.10.2020).

HUMAN ETHICS APPROVAL DECLARATION

Authors declare human ethics approval was not needed for this study.

Karra DA, Chadwick CC, Stavroulaki EM, et al. Fecal acute phase proteins in cats with chronic enteropathies. J Vet Intern Med. 2023;37(5):1750‐1759. doi: 10.1111/jvim.16802