Received: 17 December 2021 Accepted: 18 August 2022 DOI: 10.1111/jvim.16530 STANDARD ARTICLE Association of fecal and serum microRNA profiles with gastrointestinal cancer and chronic inflammatory enteropathy in dogs Janne G. Lyngby1 | Marta Gòdia2,3,4 | Louise Brogaard4 | 1 4 Annemarie T. Kristensen | Merete Fredholm | Ellen Skancke5 | Joanna Morris6 | Nana Dupont1 | Silke Salavati Schmitz7 | David Argyle7 | Armand Sánchez2,3 | Charlotte R. Bjørnvad1 | Susanna Cirera4 | Lise N. Nielsen1 1 Department of Veterinary Clinical Sciences, University of Copenhagen, Copenhagen, Denmark 2 Department of Animal Medicine and Surgery, School of Veterinary Sciences, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, Catalonia, Spain 3 Centre for Research in Agricultural Genomics, The Spanish National Research Council (CSIC), Institute of Agrifood Research and Technology (IRTA), Autonomous University of Barcelona (UAB), and University of Barcelona (UB), Cerdanyola del Vallès, Catalonia, Spain 4 Department of Veterinary and Animal Sciences, University of Copenhagen, Copenhagen, Denmark 5 Department of Companion Animal Clinical Sciences, Norwegian University of the Life Sciences, Oslo, Norway 6 College of Medical, Veterinary and Life Sciences, School of Veterinary Medicine, University of Glasgow, Glasgow, United Kingdom 7 Hospital for Small Animals, Royal (Dick) School of Veterinary Studies, The Roslin Institute, College of Medicine and Veterinary Medicine, University of Edinburgh, Easter Bush, Midlothian, United Kingdom Correspondence Janne G. Lyngby, Department of Veterinary Abstract Clinical Sciences, University of Copenhagen, Background: Reliable biomarkers to differentiate gastrointestinal cancer (GIC) from Dyrlægevej 16, 1870 Frederiksberg C, Denmark. chronic inflammatory enteropathy (CIE) in dogs are needed. Fecal and serum micro- Email: janne.lyngby@sund.ku.dk RNAs (miRNAs) have been proposed as diagnostic and prognostic markers of GI Present address disease in humans and dogs. Marta Gòdia, Animal Breeding and Genomics, Hypothesis/Objectives: Dogs with GIC have fecal and serum miRNA profiles that dif- Wageningen University and Research, Wageningen, The Netherlands. fer from those of dogs with CIE. Aims: (a) identify miRNAs that differentiate GIC from CIE, (b) use high-throughput reverse transcription quantitative real-time PCR (RT-qPCR) Funding information Fondet for sygdomsbekæmpelse hos vore to establish fecal and serum miRNA panels to distinguish GIC from CIE in dogs. familiedyr (“Fund for Disease Control in our Animals: Twenty-four dogs with GIC, 10 dogs with CIE, and 10 healthy dogs, all Companion Animals”), Grant/Award Number: A5386; The Independent Research Fund client-owned. Denmark - Technology and Production, Abbreviations: %CV, coefficient of variance; ANOVA, analysis of variance; AUC, area under the curve; CCECAI, Canine Chronic Enteropathy Clinical Activity Index; Cfa, Canis familiaris; CI, confidence interval; CIE, chronic inflammatory enteropathy; CPM, counts per million; CRP, C-reactive protein; Cq, quantification cycle; GI, gastrointestinal; GIC, gastrointestinal cancer; GIST, gastrointestinal stromal tumor; IBD, inflammatory bowel disease; LR, likelihood ratio; miRNA, microRNA; mRNA, messenger RNA; noPAP, not containing poly(A) polymerase; NPV, negative predictive values; NSAID, nonsteroidal anti-inflammatory drug; small RNAseq, small RNA sequencing; TLI, trypsin-like immunoreactivity; ROC, receiver operating characteristic; RT-qPCR, high- throughput reverse transcription quantitative real-time PCR; PLE, protein losing enteropathy; PPV, positive predictive values; UCPH, University of Copenhagen; WSAVA, World Small Animal Veterinary Association. Janne G. Lyngby, Marta Gòdia, and Louise Brogaard contributed equally as first authors. Susanna Cirera and Lise N. Nielsen contributed equally as the last authors. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2022 The Authors. Journal of Veterinary Internal Medicine published by Wiley Periodicals LLC on behalf of American College of Veterinary Internal Medicine. J Vet Intern Med. 2022;36:1989–2001. wileyonlinelibrary.com/journal/jvim 1989 1990 LYNGBY ET AL. Grant/Award Number: DFF-6111-00124; The Danish Kennel Club Methods: An international multicenter observational prospective case-control study. Small RNA sequencing was used to identify fecal and serum miRNAs, and RT-qPCR was used to establish fecal and serum miRNA panels with the potential to distinguish GIC from CIE. Results: The best diagnostic performance for distinguishing GIC from CIE was fecal miR-451 (AUC: 0.955, sensitivity: 86.4%, specificity: 100%), miR-223 (AUC: 0.918, sensi- tivity: 90.9%, specificity: 80%), and miR-27a (AUC: 0.868, sensitivity: 81.8%, specificity: 90%) and serum miR-20b (AUC: 0.905, sensitivity: 90.5%, specificity: 90%), miR-148a-3p (AUC: 0.924, sensitivity: 85.7%, specificity: 90%), and miR-652 (AUC: 0.943, sensitivity: 90.5%, specificity: 90%). Slightly improved diagnostic performance was achieved when combining fecal miR-451 and miR-223 (AUC: 0.973, sensitivity: 95.5%, specificity: 90%). Conclusions and Clinical Importance: When used as part of a diagnostic RT-qPCR panel, the abovementioned miRNAs have the potential to function as noninvasive biomarkers for the differentiation of GIC and CIE in dogs. KEYWORDS biomarker, CIE, miRNA qPCR, neoplasia, RT-qPCR, small RNA sequencing 1 | I N T RO DU CT I O N length, which primarily suppress the expression of target genes by binding to messenger RNA (mRNA) and inhibiting translation.17 Gastrointestinal cancer (GIC) is a heterogeneous group of diseases MiRNAs have been detected in feces and serum from dogs and display accounting for 8% of cancers in dogs.1 Clinical signs of GIC are non- substantial extracellular stability.14,15 Aberrant serum miRNA profiles have specific, including vomiting, diarrhea, weight loss, and anorexia, which been detected in numerous cancers in dogs,18 including intestinal T-cell makes it challenging to differentiate from other gastrointestinal (GI) disor- lymphoma19 as well as in GI tissues of dogs with CIE.13,19 However, the ders such as chronic inflammatory enteropathies (CIE). Though commonly potential of miRNAs as noninvasive diagnostic fecal and serum bio- part of the diagnostic work-up for enteropathy, abdominal ultrasonogra- markers to differentiate GIC from CIE remains unknown. phy has proven beneficial or vital in only 15% of dogs with chronic diar- We hypothesized that dogs with GIC display fecal and serum rhea, but having an abdominal or rectal mass increased the diagnostic miRNA profiles that differ from those of healthy dogs and dogs with utility.2 In some cases with solitary tumors or easy accessible lesions, min- CIE, the aims were: (a) to identify relevant miRNAs by an initial global imally invasive cytological biopsies can be acquired. However, 1 study screening of the fecal and serum miRNAome in dogs using small RNA 3 found that only 66% of cytological samples are clinically useful, and sequencing (small RNAseq) combined with a thorough investigation of when clinically useful, cytology was in complete agreement with the his- human and veterinary literature to identify fecal and serum miRNAs topathological diagnosis in 64% to 67% of the cases.3,4 In diffuse neopla- with potential to differentiate between GIC, CIE, and healthy dogs. sia or even if a solitary tumor is visualized on ultrasound, endoscopic or (b) To use high-throughput reverse transcription quantitative real-time surgical biopsies for histopathology might still remain necessary to achieve PCR (RT-qPCR) to establish a fecal and serum miRNA panel with the a diagnosis. This is costly, potentially invasive, and the required anesthesia potential to distinguish GIC from CIE in dogs. carries an inherent risk. However, achieving a correct and timely diagnosis is in turn essential to instigate appropriate treatment, improve outcomes, and provide owners with a prognosis. Even with histology, some cancers 2 | M A T E R I A L S A N D M ET H O D S such as small cell lymphoma can be challenging to distinguish from CIE, requiring additional diagnostics, delaying the time to diagnosis, and increas- 2.1 | Study design and ethics approval ing the costs. Hence, there is a need for noninvasive biomarkers to reliably differentiate GIC from CIE in dogs before they undergo more invasive This international multicenter observational prospective case-control diagnostics. study was carried out from 2018 to 2020 and initiated by the Univer- A number of biomarkers for GIC have been investigated in human sity of Copenhagen (UCPH). Participating centers included: Norwe- and veterinary medicine.5-10 MicroRNAs (miRNAs) have recently been gian University of Life Sciences, University of Glasgow, University of proposed as promising diagnostic and prognostic markers of GI dis- Edinburgh. Each institution obtained their own local ethical approval ease in both humans and dogs.11-16 MicroRNAs are small, non– (#2017-9 + 2017-12; 14/04723-72; Ref 05a/18; VERC#41.18). In protein-coding RNA molecules of approximately 20 to 23 nucleotide addition, the Animal Experiments Inspectorate under the Ministry of LYNGBY ET AL. 1991 Food, Agriculture, and Fisheries of Denmark, Danish Veterinary and characterization the healthy dogs enrolled in the small RNAseq screening Food Administration (case #2017-15-0201-01353) approved the study. underwent an upper and lower GI endoscopy under general anesthesia Owners provided informed written consent for study participation and with mucosal pinch biopsies from the stomach, duodenum, ileum, and could withdraw this at any time. colon. Dogs were recruited at UCPH, the Norwegian University of Life Sciences, University of Glasgow, the University of Edinburgh, or at Evi- densia Karlslunde Animal Hospital, or Evidensia Faxe Animal Hospital. 2.3 | Sample collection All healthy dogs were recruited at UCPH. Blood samples were collected at presentation and processed within 30 minutes of sampling. Serum for miRNA analysis was aliquoted and 2.2 | Dog recruitment stored at 20 C. Fecal samples were collected and aliquoted (approxi- mately 1 g per aliquot) into cryotubes within 1 hour of defecation and Client-owned dogs were prospectively recruited for the study and stored at 20 C for miRNA quantification. This was done at the pre- allocated into 1 of 3 groups: GIC, CIE, and healthy dogs (Table S1). sentation or at home by the client after receiving thorough verbal and writ- Dogs presenting for evaluation of chronic (>2-3 weeks duration) signs ten instructions and the necessary supplies as previously described.15,28 of GI disease, suspicion of GIC, or both, and undergoing endoscopic Mucosal endoscopic pinch biopsies or surgical full-thickness biopsies were or surgical biopsies were enrolled. obtained at the clinician's discretion. Histopathologic tissue evaluation was Dogs in all groups had to be ≥1 year of age and have a body- performed by a reference laboratory, and inflammatory changes were eval- weight ≥2.5 kg. Dogs were not included if they were being fed a raw uated using the WSAVA Standardization Grading System.26,27 All frozen food diet, had received antimicrobial therapy, corticosteroids, or other samples were stored at 80 C for batch analysis at a later stage. immunomodulatory drugs within the last 6 weeks, or nonsteroidal anti-inflammatory drugs within the 2 weeks before enrolment. All 3 groups of dogs underwent a standardized minimum diagnostic work- 2.4 | RNA isolation up including a history, physical exam, complete blood count, serum bio- chemistry, serum C-reactive protein (CRP) concentration, serum folate, For fecal samples, total RNA was isolated from 100 mg fecal material using and cobalamin concentrations, urinalysis, fecal flotation (Fecalyzer, the miRNeasy Mini Kit (Qiagen, Hilden, Germany) as described previously.28 Vetoquinol, Lure, France), and Giardia (Immunocard STAT! Crypto/Giar- For serum samples, 200 μL were used for isolation of total RNA dia, Meridian Bioscience, San Diego, California, USA) testing. The using the miRNeasy Serum/Plasma kit (Qiagen, Hilden, Germany) Canine Chronic Enteropathy Clinical Activity Index (CCECAI)20 was cal- without including the accompanying Caenorhabditis elegans miR-39 culated and a fecal score was determined using the Purina 7-point fecal spike-in extraction control. 21,22 scoring system (Nestlé S.A., Vevey, Switzerland). Concentration and purity of each RNA sample were determined In dogs suspected of CIE, additional diagnostic procedures were by assessing the 260/280 and 260/230 absorbance ratios on a Nano- performed including but not limited to serum basal cortisol concentra- Drop ND-1000 Spectrophotometer (Thermo Scientific, Hvidovre, Denmark), tion, ACTH stimulation test if basal cortisol ≤55 nmol/L,23 and serum and stored at 80 C until further use. trypsin-like immunoreactivity (TLI) concentration to rule out other GI and non-GI diseases.24,25 All CIE dogs underwent a dietary trial with a hydrolyzed or novel protein diet for a minimum of 14 days, and were 2.5 | Small RNAseq and data analysis excluded if they improved.12 Dogs with antibiotic-responsive enterop- athy were likewise excluded; however, because of a focus on prudent Small RNAseq libraries were prepared using the NEBNext Small RNA antibiotic use and strict Danish antimicrobial guidelines for dogs Library Prep Set for Illumina (New England Biolabs, Ipswich, Massa- recruited in Denmark, an antibiotic trial was not routinely undertaken. chusetts) using 6 μL total RNA from each fecal or serum sample. This Further exclusion criteria for the initial small RNAseq screening were was carried out by the sequencing service provider (Genomics Unit, the presence of concurrent diseases, serum albumin <2.0 g/dL (<20 g/L), Center for Genomic Regulation, Barcelona Biomedical Research Park, or histopathological evidence of lymphangiectasia. For RT-qPCR analysis, Barcelona, Spain) using the Illumina HiSeq 2500 system to produce comorbidities were permissible in dogs with GIC, and treatment with anti- 50 bp single-end reads. A fully detailed description of small RNAseq biotics, glucocorticoids, or other medication was allowed. Abdominal data analysis can be found in File S1.29-37 imaging (ultrasound or computed tomography scan) was performed in all GIC or CIE dogs and thoracic imaging was encouraged. Histopathology including World Small Animal Veterinary Association (WSAVA) scoring of 2.6 | Validation of candidate miRNAs using intestinal biopsies was available for all cases of CIE.26,27 RT-qPCR Healthy client-owned dogs were enrolled based on history and physi- cal exam and no episodes of signs of GI disease within the last 2 months. Identification of GIC biomarker candidates was performed using As part of a meticulous screening to ensure a clear phenotypical RT-qPCR. The high-throughput qPCR platform Biomark HD (Fluidigm, 1992 LYNGBY ET AL. San Francisco, California) was applied for this purpose in combination determining sensitivity and specificity as well as positive and negative with the appertaining 96.96 Dynamic Array integrated fluidic circuit likelihood ratios (LR+, LR). To this end, the optimal cutoff level for chips (Fluidigm, San Francisco, California). A fully detailed description each miRNA was calculated using the maximum value of Youden's of RT-qPCR can be found in File S1. index. Multiple logistic regression was applied to investigate the com- bined effect of multiple miRNAs for discrimination between GIC and CIE dogs using GraphPad Prism. 2.7 | Data analyses 2.7.1 | Clinical data 3 | RE SU LT S Data with continuous variables were assessed for normality using the 3.1 | Dog characteristics Shapiro-Wilk test. Summary statistics for continuous and ordinal vari- ables were reported as median and range and categorical data were The GIC group consisted of 24 dogs (median age: 101 months, [range, presented as absolute and relative frequencies (number, n, and per- 27-192 months]; 9 females/15 males). The most common breeds centage, %). The CCECAI score was considered an ordinal variable. were Eurasians (n = 4), French bulldogs (n = 2), and mix breed dogs Continuous variables were compared using an unpaired Student's (n = 3). A complete list of breeds can be found in Table S1. Eleven t test or Mann-Whitney test when comparing 2 groups, and a 1-way dogs had hematemesis, melena or hematochezia. Fourteen dogs had analysis of variance (ANOVA) or Kruskal-Wallis test when comparing an endoscopy, 7 dogs had a laparotomy, and additionally, 3 dogs ini- 3 groups. Categorical data were compared using Fisher's exact test or tially had an endoscopy but because of concern of not having the Chi-squared test. Bonferroni's multiple comparisons test or false dis- correct diagnosis, they underwent a laparotomy with full thickness covery rate using the Benjamini, Krieger, and Yekutieli method were biopsies; these 3 dogs all had small or intermediate cell lymphoma. used to correct for multiple comparisons. For all statistics, a P value < .05 Final diagnoses included: lymphoma (n = 9/24; small cell lymphoma was considered significant. Statistical software packages (GraphPad Prism n = 3, intermediate cell lymphoma n = 1, large cell lymphoma n = 5; 9.2.0) were used for all statistical analyses and preparation of graphs. gastric n = 2, small intestinal n = 7), carcinoma (n = 10/24; gastric n = 8, small intestinal n = 1, rectal n = 1), and sarcomas (n = 4/24; gastrointestinal stromal tumor [GIST] n = 3, nonclassified n = 1). 2.7.2 | RT-qPCR data Sarcomas were located in the stomach (n = 1) and small intestine (n = 3). One dog was diagnosed with a GI mast cell tumor (heavy Inspection of amplification curves, melting curves, and standard curves as tumor burden in the stomach and duodenum). Twelve dogs had well as negative controls (noPAP and nontemplate controls) was per- comorbidities, including cardiac disease or hypothyroidism, or had formed using the Fluidigm Real-Time PCR Analysis software (v. 4.7.1, Flui- received glucocorticoids, antibiotics, NSAIDs, or a combination thereof, digm). GenEx Pro software (v. 7.1.1.118, MultiD Analyses AB) was used before enrolment. However, when comparing dogs with comorbidities or for interplate calibration, qPCR efficiency correction of Cq values, evalua- therapeutic intervention to dogs without, there was no statistical signifi- tion of endogenous miRNA normalizers using the algorithms NormFin- cance in age (P = .09), body weight (P = .11), sex (P = .67), duration of der38 and geNorm, 39 normalization with the most suitable normalizers, disease (P = .64), clinical signs (P ≥ .99), CCECAI score (P = .64), serum calculation of relative quantities, and statistical analysis of miRNA amounts cobalamin concentration (P = .7), serum folate concentration (P = .76), or between groups. When comparing 3 groups, statistical significance of the serum CRP concentration (P = .87). These 24 dogs are described as difference in miRNA amounts was assessed with 1-way ANOVA and 1 GIC group throughout the article. Tukey-Kramer's test after pairwise comparisons. When comparing miRNA Ten dogs with CIE were enrolled. All dogs had clinical signs of amounts in only 2 groups, Student's t test was applied. Significance levels small intestinal disease alone or together with gastritis or colitis. Dogs were set to ɑ = .0016 (feces) and ɑ = .00070 (serum) to account for per- with CIE had a longer duration of disease (median: 284 days, range, forming multiple comparisons (Šidàk corrected P values, as employed by 45-730 days) compared to GIC (median: 120 days, range, 3-730 days; the GenEx Pro software). Analyses yielding P values between .05 and the P = .02). The CCECAI score was significantly different with a median corrected P values were interpreted with caution. MiRNA fold changes of 6 (range, 3-12) in CIE and 11 (1-17) in GIC (P = .04), but the distri- were computed as the ratio between the average relative miRNA amounts bution of clinical signs did not differ between CIE and GIC groups in the 2 groups being compared, and with a 95% confidence interval. (Table 1). All dogs had serum resting cortisol concentration measured, Correlation of miRNA quantification as determined by small RNA- 5/10 had an ACTH stimulation test, 9/10 had serum TLI concentra- seq and RT-qPCR was assessed using Pearson's correlation coefficient tion measured, and 5/10 had serum pancreatic lipase immunoreactiv- (Pearson's r) using GraphPad Prism. ity concentration measured. Four dogs had hematemesis, melena, The ability of selected miRNAs to discriminate between GIC and hematochezia, or a combination hereof. All dogs had an upper and CIE dogs was assessed by receiver operating characteristic (ROC) lower endoscopy performed and the ileum was accessed in 8 of the curve analysis using GraphPad Prism. Diagnostic performance of indi- 10 dogs. The dominant histopathological diagnoses were vidual miRNAs or combinations of miRNAs was evaluated by lymphocytic-plasmacytic enteritis (n = 7), gastritis and colitis (n = 1), LYNGBY ET AL. 1993 TABLE 1 Dog characteristics Group characteristic Healthy dogs CIE GIC P valuea Total number, n 10 10 24 Dog characteristics Age in monthsb, median (range) 61 (15-130) 53 (18-132) 101 (27-192) .02 Sex, n, female/male 5/5 3/7 9/15 .87 Body weight in kgc, median (range) 14 (6.5-33) 10 (2.6-22) 15 (2.4-74) .31 Clinical variables Disease duration in daysc, median (range) NA 284 (45-730) 120 (3-730) .02 Clinical signs, n (%) Vomiting NA 5/10 (50%) 19/23 (83%) .09 Diarrhea NA 7/10 (70%) 14/23 (61%) .71 Weight loss NA 3/9 (33%) 14/23 (61%) .24 Hyporexia/ anorexia NA 8/10 (80%) 21/23 (91%) .57 GI bleeding (hematemesis, hematochezia, or melena) NA 4/10 (40%) 11/24 (46%) >.99 CCECAI scored, median (range) 1 (0-4) 6 (3-12) 11 (1-17) .04 Clinical disease severity, n (%) 10/10 (100%) 9/10 (90%) 18/24 (75%) .33 Mild (CCECAI score ≤5) 10 (100%) 3 (33%) 2 (11%) Moderate (CCECAI score 6-8) 0 3 (33%) 4 (22%) Severe (CCECAI score 9-11) 0 1 (11%) 4 (22%) Very severe (CCECAI score ≥ 12) 0 2 (22%) 8 (45%) Clinicopathological variables Serum cobalamine (pmol/L), median (range) 341 (242-1066) 311 (145-573) 204 (105-675) .06 Serum folatef (nmol/L), median (range) 27.9 (20.5-54.4) 24.3 (6.2-93.1) 16.7 (2.5-50.8) .18 f CRP (g/L), median (range) 5.7 (0-7.4) 6.8 (0.7-10.2) 14.1 (0-104.7) .15 Note: All 3 study groups are compared in the statistical analyses except of “Clinical variables” as this was not relevant for the healthy dogs. Abbreviations: CCECAI, canine chronic enteropathy clinical activity index; CIE, chronic inflammatory enteropathy; CRP, C-reactive protein; GIC, gastrointestinal cancer; GI, gastrointestinal. a Significant difference between groups (P < .05) is indicated with boldface values. b Significant difference was observed when comparing all 3 groups in an ANOVA; the difference remains significant only between GIC and CIE when using Bonferroni's multiple comparisons test (P = .04). c One dog was censored because of missing data. d Significant difference between GIC and CIE dogs using Mann–Whitney test. Statistical comparisons to the healthy group were not made for the CCECAI score. Seven dogs were censored because of missing values in CCECAI. e No significant difference was observed when comparing all 3 groups in a Kruskal–Wallis test; however, there was a significant difference between GIC and healthy dogs when the false discovery rate using the Benjamini, Krieger, and Yekutieli method was applied post hoc (P = .02). Not documented in 6 dogs. f Not documented in 5 dogs. gastritis and enteritis (n = 1), or gastritis, enteritis, and colitis (n = 1). serum samples ranged from 7.5 to 49.6 ng/μL (average 16.5 ng/μL) with The 1 dog with gastritis combined with colitis as the dominant histo- mean 260/280 and 260/230 ratios of 1.45 and 0.38, respectively. pathological diagnosis did also have mild enteritis as well as clinical signs of small intestinal disease. 3.1.2 | Small RNAseq screening of feces and serum from dogs 3.1.1 | RNA isolation A total of 21 dogs (6 healthy dogs, 9 dogs with CIE, and 6 dogs with Of the 44 enrolled dogs, feces were obtained from 42 dogs and serum GIC) were included in the initial small RNAseq screening (Figure 1, was obtained from 41 dogs, to be used for RNA isolation (Figure 1, Table S1). Small RNAseq libraries yielded an average of 11.2 M and Table S1). Yield of RNA from fecal samples ranged from 27.9 to 10.9 M reads for fecal and serum samples, respectively. Over 98% of 864.1 ng/μL (average 282.6 ng/μL) with average 260/280 and reads passed quality control filters (Table S3). On average, 24% and 260/230 ratios of 1.98 and 1.37, respectively. Yield of RNA from 84% of the filtered reads mapped to the Canis familiaris (cfa) genome 1994 LYNGBY ET AL. 3.1.3 | RT-qPCR identification of GIC biomarkers Dog recruitment Of the 96 miRNAs included in the panel, 36, and 75 assays were suc- Healthy : 6 cessfully quantified using RT-qPCR in feces and serum, respectively GIC: 6 (Tables S240-57 and S6). Successful quantification was determined CIE: 9 based on the absence of or negligible fluorescence in noPAP controls, mono-peaked melting curves indicating specific amplification of the Addional: desired product, acceptable qPCR efficiency (ranged from 93% to Healthy : 4 109% for feces, 91% to 114% for serum, Table S2),40-57 and good Screening GIC: 18 Small RNA sequencing CIE: 1 cDNA reproducibility (less than 1 Cq difference between cDNA repli- cates). miR-16, miR-29a, and miR-148b were used for the normaliza- tion of fecal data, and miR-148b and miR-320 were used for the Healthy : 6 normalization of serum data. GIC: 6 Significantly higher relative quantity of miR-451 were detected in CIE: 9 the feces of dogs with GIC compared to CIE, confirming results obtained in the small RNAseq screening (Table 2, Figure 2A). Addition- ally, RT-qPCR results demonstrated significantly higher miR-27a and Biomarker idenficaon miR-223 amounts in the feces of dogs with GIC compared to CIE RT-qPCR (Table 2, Figure 2A). Increased amounts of miR-143-3p and miR-145 in serum of GIC dogs compared to CIE as demonstrated in the small Healthy: 10 dogs (feces n = 10; serum n = 10) RNAseq screening were confirmed by RT-qPCR, although there was GIC: 24 dogs (feces n = 22; serum n = 21) CIE: 10 dogs (feces n = 10; serum n = 10) no statistical significance after correcting for multiple testing (Table 2, Figure 2B). RT-qPCR also demonstrated significantly increased serum amounts of miR-20b, miR-148a-3p, and miR-652 F I G U R E 1 Study design and recruitment. Six healthy dogs, 9 dogs with chronic inflammatory enteropathy (CIE), and 6 dogs with in GIC compared to CIE (Table 2, Figure 2B). gastrointestinal cancer (GIC) were included in the initial small RNAseq No significant differences in any fecal or serum miRNA amounts screening for miRNA GIC biomarkers in feces and serum. An were demonstrated when comparing CIE to healthy dogs (Table 2, additional 4 healthy dogs, 1 dog with CIE, and 18 dogs with GIC were Table S6). Within the GIC group, having comorbidities or having subsequently recruited for RT-qPCR identification of biomarker received glucocorticoids, antibiotics, or NSAIDs did not affect miRNA candidates. As such, the total number of dogs included in qPCR analysis was 10 healthy, 10 CIE, and 24 GIC dogs. Of the GIC dogs, amounts (Table S6). Also, across different types or locations of GIC, no feces was obtained from 22, and serum was obtained from 21 significant differences in miRNA amounts were observed (Table S6). in fecal and serum samples, respectively (Table S3). Of these mapped 3.1.4 | Diagnostic performance of fecal and serum reads, 0.02% in feces and 4.3% in serum mapped to annotated mature miRNAs miRNAs in the C familiaris genome (Table S3; Figure S1). Of the 453 mature C familiaris miRNAs annotated in miRBase (v.22.1) a total When assessing miRNAs for their ability to discriminate between GIC of 210 and 283 miRNAs were identified in at least 1 fecal or serum and CIE, the best performances were seen for fecal miR-451, miR- sample, respectively (Table S4). On average, the most abundant miRNAs 223, and miR-27a, and serum miR-20b, miR-148a-3p, and miR-652, in feces were cfa-miR-21 (177 930 counts per million [CPM]), cfa-miR- with an area under the curve (AUC) of 0.955, 0.918, and 0.868, and 192 (104 096 CPM), and cfa-miR-215 (99 857 CPM), whereas the most 0.905, 0.924, and 0.943, respectively (Figure 3, Table 3). Combining abundant miRNAs in serum were cfa-miR-486-3p (286 956 CPM), cfa- miRNAs resulted in slight improvements in diagnostic performance, miR-486 (263 105 CPM), and cfa-miR-92a (63 269 CPM; Table S4). with the greatest improvement observed for the combination of fecal By setting a threshold of a q value <.05 and a jlog2FCj ≥1.5, 1 miR-451 and miR-223. This combination of markers yielded an AUC fecal miRNA (cfa-miR-451) and 5 serum miRNAs (cfa-miR-133c, cfa- of 0.973, a sensitivity of 95.5%, a specificity of 90%, and LR+ and miR-133a, cfa-miR-145, cfa-miR-143, cfa-miR-1) were found to be LR of 9.55 and 0.05, respectively (Table 3). present at significantly different amounts in GIC and CIE dogs, and all 6 miRNAs were present at higher amounts in GIC compared to CIE (Figure S2, Table S5). Comparison of GIC and healthy dogs 4 | DI SCU SSION revealed 2 additional fecal miRNAs (cfa-miR-194, lower in GIC, and cfa-miR-320, higher in GIC) and 1 serum miRNA (cfa-miR-122, This study demonstrates an association between the miRNAome in higher in GIC) to be different (Figure S2, Table S5). When comparing feces and serum in clinically well-characterized dogs with either GIC CIE and healthy dogs, no differences were found. or CIE. Fecal miR-451, miR-223, and miR-27a, as well as serum miR- LYNGBY ET AL. 1995 TABLE 2 Potential miRNA GIC biomarkers in feces and serum from dogs identified with qPCR Significant correlation FC (GIC vs CIE) FC (GIC vs healthy) FC (CIE vs healthy) between small RNAseq P value (1-way ANOVA) (±95% CI) (±95% CI) (±95% CI) and qPCR? (Pearson's r) Fecal miRNAs miR-451 <.001 59 (±38) 59 (±38) 0.99 (±0.63) Yes, P < .001, r = .78 miR-223 <.001 25 (±13) 19 (±10) 0.76 (±0.58) Nob, P = .09, r = .53 miR-27a <.001 2.3 (±0.40) 2.1 (±0.37) 0.93 (±0.31) Nob, P = .51, r = .18 Serum miRNAs miR-20b <.001 2.3 (±0.53) 2.6 (±0.61) 1.2 (±0.21) Yes, P = .003, r = .62 miR-148a-3p <.001 2.2 (±0.39) 2.2 (±0.39) 1.0 (±0.16) Yes, P < .001, r = .75 miR-652 <.001 5.0 (±2.2) 3.7 (±1.7) 0.74 (±0.21) Yes, P < .001, r = .77 miR-145 .0034a 5.0 (±3.1) 4.9 (±3.1) 0.98 (±0.23) Yes, P < .001, r = .89 miR-143-3p .0075 a 5.3 (±3.1) 3.1 (±1.8) 0.59 (±0.12) Yes, P < .001, r = .86 Abbreviations: ANOVA, analysis of variance; CIE, chronic inflammatory enteropathy; CI, confidence interval; FC, fold change; GIC, gastrointestinal cancer. a P value below .05, but does not survive correction for multiple comparisons in 1-way ANOVA. b Not quantifiable in several samples with small RNAseq (48% missing values for miR-223; 28% missing values for miR-27a). Fold change value in bold: P value for this pairwise comparison (Tukey–Kramer's test) survives correction for multiple comparisons. Fold change value underlined: P value below .05 for this pairwise comparison (Tukey–Kramer's test), but does not survive correction for multiple comparisons. All other comparisons: nonsignificant. 95% CI for fold changes are indicated in brackets. 20b, miR-148a-3p, and miR-652, were identified as noninvasive associated with resistance toward chemotherapy.65 The promotion of potential biomarkers for the differentiation of GIC and CIE by quan- cancer cell proliferation and metastasis has likewise been described in tifying a comprehensive panel of miRNAs in feces and serum using relation to gastric cancer.66,67 In contrast to miR-451 and miR-223, RT-qPCR. Upon validation in a larger and clinically more diverse the observed increase of miR-27a in feces from dogs with GIC is more group of dogs with chronic GI disease, future application of some or likely to be a direct effect of up-regulation in cancerous tissue, as all of these miRNAs in a diagnostic qPCR assay might facilitate early occurs in humans with colorectal cancer11,68 and gastric cancer,66,69 and timely differentiation between GIC and CIE in dogs before more and might therefore be a more reliable GIC biomarker if no GI bleed- invasive diagnostics are performed. ing is present, as the increase of fecal miR-27a amounts in GIC dogs Results demonstrate miR-451 and miR-223 to be robust fecal would not rely on GI bleeding. However, the role of miR-27a in GIC is biomarkers for GIC in dogs, which mirrors findings regarding colo- not currently clear, and further investigations are warranted in order rectal cancer in people.58,59 miR-451 is highly expressed in erythro- to confirm its oncomiR function in dogs. The varying fold change cytes and miR-223 is highly abundant in, for example, neutrophils.60,61 amounts that we observe between GIC and CIE dogs also support the The increase in miR-451 and miR-223 amounts observed in feces of GIC different mechanisms suggested for the observed increased fecal dogs could therefore reflect increased presence of blood in the amounts of miR-451, miR-223, and miR-27a in GIC dogs. Here, we feces of dogs with GIC compared to CIE. For this same reason, these observed a 2.3-fold increase in fecal miR-27a amounts in GIC dogs com- 2 miRNAs are suggested as specific fecal biomarkers of necrotizing pared to CIE, whereas the increase for miR-451 and miR-223 was 62 enterocolitis in preterm infants. Consequently, whether the pres- 59- and 25-fold, respectively. This discrepancy suggests a differ- ence of miR-451 and miR-223 in feces from dogs is unambiguously ence in the underlying mechanisms resulting in the presence of these miR- indicative of GIC should be addressed in future studies, as other GI NAs in the feces of GIC dogs, supporting that the slight increase in miR- diseases might likewise cause blood to be present in the feces. 27a amounts could be because of its up-regulation locally in tumor tissue Investigating fecal miRNA amounts in relation to intestinal bleeding and after release into feces, whereas the substantial increase of miR-451 was not an aim of the present study; however, we subsequently and miR-223 is likely because of increased amount of blood in feces. identified the occurrence of intestinal bleeding (melena, hematoche- Should fecal miR-27a amounts indeed be a reflection of local zia, hematemesis, or a combination thereof) to be comparable in the tumor up-regulation of expression, then it could potentially not be GIC and CIE groups. Neutrophil influx during intestinal inflammation equally well suited as a biomarker for all types of GIC in dogs. should also be considered as a source of miR-223, in case these cells Although we did not observe any significant differences in miR-27a are shed or leaked into feces even in the absence of intestinal amounts when cancer dogs were grouped according to cancer type bleeding.63 (carcinoma, lymphoma, sarcoma), we did however find that miR-27a miR-27a is generally described as an oncomiR (miRNA onco- was inferior to miR-451 and miR-223 in discriminating between GIC gene).64 High amounts of miR-27a in human colorectal cancer have and CIE dogs; in this regard, the likely blood-derived biomarkers miR- been shown to support cell proliferation and tumor growth and to be 451 and miR-223 performed more robustly. 1996 LYNGBY ET AL. (A) (A) miR-451 (feces) (B) miR-223 (feces) P < .001 Healthy 100 100 1 000 000 P < .001 CIE P < .001 GIC 80 80 100 000 p < .001 60 60 Relative miRNA level 10 000 P = .011* 40 40 1000 20 20 P < .001 AUC = 0.955 AUC = 0.918 100 0 0 0 20 40 60 80 100 0 20 40 60 80 100 10 1 (C) miR-27a (feces) (D) miR-20b (serum) 100 100 0.1 80 80 miR-451 miR-223 miR-27a 60 60 (B) 40 40 P < .001 1000 P = .014* 20 20 P = .012* AUC = 0.868 AUC = 0.905 P < .001 0 0 P = .017* Sensitivity% (TPR) 0 20 40 60 80 100 0 20 40 60 80 100 100 P < .001 Relative miRNA level P < .001 P < .001 (E) miR-148a-3p (serum) (F) miR-652 (serum) P < .001 10 100 100 80 80 60 60 1 40 40 20 20 0.1 AUC = 0.924 AUC = 0.943 0 0 0b p 52 45 3p 0 20 40 60 80 100 0 20 40 60 80 100 -3 -6 -1 -2 a- 43 iR iR iR 48 -1 m m m -1 iR iR m (G) (H) m miR-145 (serum) miR-143-3p (serum) F I G U R E 2 Differentially abundant miRNAs in A (feces) and B 100 100 (serumB). Boxes extend from 25th to 75th percentile; whiskers 80 80 extend to the minimum and maximum values; horizontal lines indicate 60 60 median value. P values indicate the significance level of pairwise comparisons (Tukey–Kramer's test). Healthy dogs: n = 10 (feces and 40 40 serum); chronic inflammatory enteropathy (CIE): n = 10 (feces and 20 20 serum); GIC: n = 22 (feces) and n = 21 (serum). *Significance level of AUC = 0.795 AUC = 0.810 pairwise comparison does not survive correction for multiple testing. 0 0 Individual miRNA amounts in all 3 dog groups can be found in 0 20 40 60 80 100 0 20 40 60 80 100 Table S6 100% - Specificity% (FPR) F I G U R E 3 Receiver operating characteristic curves for fecal Fecal samples proved challenging to collect in several dogs, because miRNAs miR-451 (A), miR-223 (B), and miR-27a (C) and serum dogs with prolonged anorexia or hyporexia can have decreased fecal vol- miRNAs miR-20b (D), miR-148a-3p (E), miR-652 (F), miR-145 (G), and ume. In such dogs, serum biomarkers would prove more suitable, but in miR-143-3p (H). AUC, area under the curve; FPR, false positive rate; TPR, true positive rate. x- and y-axes show 100% - specificity (%) and the majority of dogs, feces is easily accessible and the ideal noninvasive sensitivity (%) with regards to the diagnosis of GIC vs CIE, matrix for GI disease. From a technical perspective, obtaining successful respectively quantification of miRNAs with RT-qPCR proved challenging using feces. The same 96 assays were applied to both fecal and serum samples, and whereas 75 assays were successful in serum, only 36 achieved acceptable and of other RNA molecules before extraction, yielding short RNA degra- results in feces. The primary issue appeared to be nonspecific amplifica- dation products that might cross-react with some miRNA primers. tion in the majority of the assays, which could be a consequence of the Due to accessibility, ease of testing, and extracellular stability of hostile environment offered by feces, resulting in degradation of miRNAs miRNAs in serum, this sample material is an obvious target for LYNGBY ET AL. 1997 TABLE 3 Diagnostic performance of miRNAs AUC 95% CI (AUC) Sn (%) 95% CI (Sn) Sp (%) 95% CI (Sp) LR+ LR Fecal miRNAs miR-451 0.955 0.888-1.0 86.4 66.7-95.3 100 72.3-100 Infinity 0.14 miR-223 0.918 0.810-1.0 90.9 72.2-98.4 80 49.0-96.5 4.55 0.11 miR-27a 0.868 0.715-1.0 81.8 61.5-92.7 90 59.6-99.5 8.18 0.20 miR-451 + miR-223 0.973 0.925-1.0 95.5 78.2-99.8 90 59.6-99.5 9.55 0.05 Serum miRNAs miR-20b 0.905 0.778-1.0 90.5 71.1-98.3 90 59.6-99.5 9.05 0.11 miR-148a-3p 0.924 0.830-1.0 85.7 65.4-95.0 90 59.6-99.5 8.57 0.16 miR-652 0.943 0.863-1.0 90.5 71.1-98.3 90 59.6-99.5 9.05 0.11 miR-145 0.795 0.636-0.955 52.4 32.4-71.7 100 72.3-100 Infinity 0.48 miR-143-3p 0.810 0.659-0.960 57.1 36.6-75.5 100 72.3-100 Infinity 0.43 Note: Diagnostic performance of miRNAs with regards to distinguishing between GIC and CIE. For any analyses including only fecal miRNA: n = 32 (22 GIC, 10 CIE); for any analyses including only serum miRNA: n = 31 (21 GIC, 10 CIE). Abbreviations: AUC, area under the curve; CI, confidence interval; LR, negative likelihood ratio; LR+, positive likelihood ratio; Sn, sensitivity; Sp, specificity. biomarker identification.70 The serum miRNA profile could represent All dogs in this study were systematically characterized through any pathological change occurring in an individual dog, and the pro- specified inclusion and exclusion criteria, clinical and paraclinical diag- cess of validating serum biomarkers should include ascertaining that nostic workup, scoring, and phenotyping. This is the strength of the changes to serum amounts of a given miRNA can be ascribed to the current study, and make the results highly applicable when evaluating disease in question. miR-652 could have potential as a serum bio- a dog fulfilling the inclusion criteria. However, it also introduces selec- marker for the differentiation between GIC and CIE in dogs, as this tion bias to our cohort as many dogs with chronic enteropathy, includ- was the best performing serum miRNA with regards to the ability to ing food-responsive and antibiotic-responsive enteropathy, as well as discriminate between GIC and CIE. The relevance of miR-652 is mir- all CIE dogs with comorbidities or dogs receiving medication were rored in the literature, where it has been identified as a plasma bio- excluded. We did exclude CIE dogs with primary protein losing enter- marker of gastric cancer as well as a prognostic serum biomarker of opathy (PLE) or albumin below 2.0 g/dL. It is unknown if dogs with 71,72 relapse in colorectal cancer in humans. miR-148a-3p amounts were PLE would have an altered fecal miRNA profile, because of leakage of significantly higher in GIC serum relative to both healthy and CIE dogs, miRNAs into the intestinal lumen. which was substantiated by a strong correlation with small RNAseq As the main objective was to identify dogs with GIC, a subgroup results. This miRNA has previously been described as a down-regulated of cancer dogs with comorbidities or that had received medication serum biomarker of human colorectal cancer relative to healthy con- (or both) were included in the RT-qPCR biomarker identification, but trols.73 The discrepancy between these studies could be because of the only after statistical evaluation confirming no differences in clinical or heterogeneity of both cancer patients and tumor types in our study miRNA profiles between dogs with comorbidities or that had received group. In humans, serum miR-20b is suggested as a biomarker for both medication and dogs that had not. This approach more strongly 74,75 gastric cancer and inflammatory bowel disease. Whereas our results reflects the dogs encountered in a clinical setting. demonstrated significantly increased amounts of miR-20b in GIC dogs rel- There are 3 different tumor types investigated in this study. This ative to both CIE and healthy dogs, we did not find any significant differ- serves both as a strength and as a limitation. The number of dogs with ences between CIE and healthy dogs for this miRNA. In contrast to each tumor type varies, and reflects the prevalence of diseases, how- human findings, this might suggest that miR-20b could function as a spe- ever evaluating our proposed biomarkers on a larger group of dogs cific GIC serum biomarker in dogs. However, evaluation of miR-20b with each cancer type could elucidate other associations. Only 7/24 including a more heterogeneous group of CIE dogs with higher CCECAI GIC dogs had thoracic imaging to evaluate pulmonary pathology, scores could serve to clarify the ability of miR-20b in discriminating fully hence primary tumors or metastases could have been missed. between GIC and CIE dogs. Another limitation of this study was that all CIE and the majority Our results from the CIE dogs do not support the previous findings of GIC dogs had endoscopic pinch biopsies rather than full-thickness of increased amounts of miR-16, miR-122, miR-146a, and miR-147 and biopsies. Though endoscopy is noninvasive and might facilitate locali- decreased amounts of miR-185, miR-192, and miR-223 in serum in dogs zation of lesions, the endoscope is rarely able to access the jejunum, 13 with large intestinal inflammatory bowel disease. This discrepancy might and in some dogs, it can be challenging to gain access to and obtain be because of differences in the investigated dog populations in age, loca- biopsies from the ileum. Thus, the quality and extent of pinch biopsies tion of disease, as well as differences in CCECAI scores.13 are inferior to that of full-thickness biopsies, potentially leading to 1998 LYNGBY ET AL. incorrect diagnoses.76-78 Dogs in all groups had biopsies evaluated by INSTITU TIONAL ANIMAL C AR E AND USE COMMITTEE a pathologist at a reference laboratory, however, we did not have a (IACUC) OR OTHER APPROVAL DECLARATION single pathologist review all samples. Even though the interobserver The study was approved by the local ethics committees (University of variation between pathologists can be significant,79 histopathologic Copenhagen: #2017-9 + 2017-12; Norwegian University of Life Sci- scoring was not an objective of this study, and was not pursued ences: 14/04723-72, University of Glasgow: Ref 05a/18, University further. of Edinburgh: VERC#41.18) at all involved institutions as well as by The CIE dogs in our study had a median CCECAI score of 6 (mod- the Animal Experiments Inspectorate under the Ministry of Food, erate disease). This CCECAI score was assessed at the time of enroll- Agriculture, and Fisheries of Denmark, Danish Veterinary and Food ment and sample collection, and most dogs did have episodes of Administration (case #2017-15-0201-01353). higher CCECAI scores before or after initial sample collection. Exclud- ing dogs with PLE could also account for the moderate CCECAI score HUMAN E THICS APPROVAL DECLARATION in these population. In future investigations, it will be necessary to Authors declare human ethics approval was not needed for this study. address the size and composition of the investigated CIE group, including a larger and more heterogeneous group in validation of the bio- DATA AVAILABILITY STAT EMEN T marker candidates. This should include other types of chronic enter- All small RNA sequencing data is available from the Sequencing Read opathies including food-responsive and antibiotic-responsive Archive (https://www.ncbi.nlm.nih.gov/sra; BioProject ID: PRJNA789413). enteropathy, PLE, dogs with comorbidities or medical intervention, as well as dogs with GI bleeding. Food-derived miRNAs (xenomiRs) OR CID are detectable in feces and blood of humans,80,81 and it is, therefore, Janne G. Lyngby https://orcid.org/0000-0003-0160-743X pertinent to evaluate whether meat-rich or raw food diets affect Louise Brogaard https://orcid.org/0000-0003-4571-6822 fecal miRNA amounts in dogs. Annemarie T. Kristensen https://orcid.org/0000-0003-3391-4928 For the potential future application of the identified GIC miRNA Merete Fredholm https://orcid.org/0000-0002-3563-7648 biomarkers in a diagnostic RT-qPCR assay, it is important to consider Joanna Morris https://orcid.org/0000-0001-5975-2475 the performance of the endogenous miRNA normalizers applied in Silke Salavati Schmitz https://orcid.org/0000-0003-1084-7013 qPCR data analysis.82 Four different miRNAs were employed as nor- Charlotte R. Bjørnvad https://orcid.org/0000-0002-2876-1667 malizers in the present study because of their stability as demon- Susanna Cirera https://orcid.org/0000-0001-8105-1579 strated by the NormFinder38 and geNorm39 algorithms. Three were Lise N. Nielsen https://orcid.org/0000-0002-3988-0979 included in these qPCR panels based on either previous reports of suitability as normalizer in feces or serum (miR-16, miR-320),16,83-86 RE FE RE NCE S or based on apparently stable amounts across dog groups in small 1. Dobson JM, Samuel S, Milstein H, Rogers K, Wood JLN. Canine neo- RNAseq (miR-148b). Additionally, miR-29a was identified as a suitable plasia in the UK: estimates of incidence rates from a population of endogenous normalizer although it was originally included in the panel insured dogs. J Small Anim Pract. 2002;43(6):240-246. doi:10.1111/j. because of being increased in serum of ulcerative colitis patients.74 1748-5827.2002.tb00066.x 2. Leib MS, Larson MM, Grant DC, et al. Diagnostic utility of abdominal This highlights the necessity for the identification of suitable endoge- ultrasonography in dogs with chronic diarrhea. J Vet Intern Med. nous miRNA qPCR normalizers specifically in the investigated sample 2012;26(6):1288-1294. doi:10.1111/j.1939-1676.2012.01009.x material, as good normalizers might not always be found only in 3. Turner RBS, Liffman R, Woodward AP, Beck C, Courtman N, literature. Dandrieux JRS. Assessment of the clinical usefulness of ultrasound-guided cytological specimens obtained from gastrointestinal lesions in dogs and In conclusion, we identified miR-451, miR-223, and miR-27a in feces cats. J Small Anim Pract. 2021;62(2):114-122. doi:10.1111/jsap.13260 and miR-20b, miR-148a-3p, miR-652 in serum, as noninvasive biomarker 4. Bonfanti U, Bertazzolo W, Bottero E, et al. Diagnostic value of cyto- candidates for the optimization of a diagnostic RT-qPCR panel with the logic examination of gastrointestinal tract tumors in dogs and cats: potential to differentiate between GIC and CIE in dogs. 83 cases (2001-2004). J Am Vet Med Assoc. 2006;229(7):1130-1133. doi:10.2460/javma.229.7.1130 5. Carethers JM, Fecal DNA. Testing for colorectal cancer screening. ACKNOWLEDGMENT Annu Rev Med. 2020;71:59-69. doi:10.1146/annurev-med-103018- This study was funded by the Independent Research Fund Denmark - 123125 Technology and Production, The Danish Kennel Club, and “Fondet for 6. Herrera-Pariente C, Montori S, Llach J, Bofill A, Albeniz E, Moreira L. sygdomsbekæmpelse hos vore familiedyr” (“Fund for Disease Control Biomarkers for gastric cancer screening and early diagnosis. Biomedi- cine. 2021;9(10):1448. doi:10.3390/biomedicines9101448 in our Companion Animals”). 7. Seim-Wikse T, Skancke E, Nødtvedt A, et al. Comparison of body con- dition score and other minimally invasive biomarkers between dogs CONF LICT OF IN TE RE ST DEC LARAT ION with gastric carcinoma and dogs with chronic gastritis. J Am Vet Med The authors declare no conflict of interest. Assoc. 2019;254(2):226-235. doi:10.2460/javma.254.2.226 8. Seim-Wikse T, Kolbjørnsen T, Jörundsson E, et al. Tumour gastrin expression and serum gastrin concentrations in dogs with gastric car- OFF- LABE L ANT IMICR OBIAL DE CLARAT ION cinoma are poor diagnostic indicators. J Comp Pathol. 2014;151(2–3): The authors declare no off-label use of antimicrobials. 207-211. doi:10.1016/j.jcpa.2014.05.002 LYNGBY ET AL. 1999 9. Heilmann RM, Berghoff N, Mansell J, et al. Association of fecal calpro- 28. Lyngby JG, Kristensen AT, Fredholm M, Nielsen LN, Cirera S. Evalua- tectin concentrations with disease severity, response to treatment, tion of fecal microRNA stability in healthy cats. Vet Clin Pathol. 2019; and other biomarkers in dogs with chronic inflammatory enteropa- 48(3):455-460. doi:10.1111/vcp.12757 thies. J Vet Intern Med. 2018;32(2):679-692. doi:10.1111/jvim.15065 29. Martin M. Cutadapt removes adapter sequences from high- 10. Gerou-Ferriani M, Allen R, Noble PJM, German AJ, Caldin M, throughput sequencing reads. EMBnetj. 2011;17(1):10-12. doi:10. Batchelor DJ. Determining optimal therapy of dogs with chronic 14806/ej.17.1.200 enteropathy by measurement of serum citrulline. J Vet Intern Med. 30. Langmead B, Trapnell C, Pop M, Salzberg SL. Ultrafast and memory- 2018;32(3):993-998. doi:10.1111/jvim.15124 efficient alignment of short DNA sequences to the human genome. 11. Duran-Sanchon S, Moreno L, Augé JM, et al. Identification and valida- Genome Biol. 2009;10(3):R25. doi:10.1186/gb-2009-10-3-r25 tion of MicroRNA profiles in Fecal samples for detection of colorectal 31. Yang L, Smyth Gordon K, Wei S. FeatureCounts: an efficient general cancer. Gastroenterology. 2020;158(4):947-957.e4. doi:10.1053/j. purpose program for assigning sequence reads to genomic features. gastro.2019.10.005 Bioinformatics. 2014;30(7):923-930. 12. Heilmann RM, Steiner JM. Clinical utility of currently available bio- 32. Kozomara A, Birgaoanu M, Griffiths-Jones S. MiRBase: from micro- markers in inflammatory enteropathies of dogs. J Vet Intern Med. RNA sequences to function. Nucleic Acids Res. 2019;47(D1):D155- 2018;32(5):1495-1508. doi:10.1111/jvim.15247 D162. doi:10.1093/nar/gky1141 13. Konstantinidis AO, Pardali D, Adamama-Moraitou KK, et al. Colonic 33. McCarthy DJ, Chen Y, Smyth GK. Differential expression analy- mucosal and serum expression of microRNAs in canine large intestinal sis of multifactor RNA-Seq experiments with respect to biologi- inflammatory bowel disease. BMC Vet Res. 2020;16(1):69. doi:10. cal variation. Nucleic Acids Res. 2012;40(10):4288-4297. doi:10. 1186/s12917-020-02287-6 1093/nar/gks042 14. Enelund L, Nielsen LN, Cirera S. Evaluation of microRNA stability in 34. Love MI, Huber W, Anders S. Moderated estimation of fold change plasma and serum from healthy dogs. MicroRNA. 2017;6(1):42-52. and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014; doi:10.2174/2211536606666170113124114 15(12):550. doi:10.1186/s13059-014-0550-8 15. Cirera S, Willumsen LM, Johansen TT, Nielsen LN. Evaluation of 35. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a prac- microRNA stability in feces from healthy dogs. Vet Clin Pathol. 2018; tical and powerful approach to multiple testing. J R Stat Soc B. 1995; 47(1):115-121. doi:10.1111/vcp.12566 57(1):289-300. 16. Wohnhaas CT, Schmid R, Rolser M, et al. Fecal MicroRNAs show 36. Balcells I, Cirera S, Busk PK. Specific and sensitive quantitative RT- promise as noninvasive Crohn's disease biomarkers. Crohn's Colitis PCR of miRNAs with DNA primers. BMC Biotechnol. 2011;11:70. doi: 360. 2020;2(1):otaa003. doi:10.1093/crocol/otaa003 10.1186/1472-6750-11-70 17. Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. 37. Busk PK. A tool for design of primers for microRNA-specific quantita- Cell. 2004;116(2):281-297. doi:10.1016/S0092-8674(04)00045-5 tive RT-qPCR. BMC Bioinform. 2014;15:29. doi:10.1186/1471-2105- 18. Sahabi K, Selvarajah GT, Abdullah R, Cheah YK, Tan GC. Compar- 15-29 ative aspects of microRNA expression in canine and human can- 38. Andersen CL, Jensen JL, Ørntoft TF. Normalization of real-time quan- cers. J Vet Sci. 2018;19(2):162-171. doi:10.4142/jvs.2018.19. titative reverse transcription-PCR data: a model-based variance esti- 2.162 mation approach to identify genes suited for normalization, applied to 19. Joos D, Leipig-Rudolph M, Weber K. Tumour-specific microRNA bladder and colon cancer data sets. Cancer Res. 2004;64(15):5245- expression pattern in canine intestinal T-cell-lymphomas. Vet Comp 5250. doi:10.1158/0008-5472.CAN-04-0496 Oncol. 2020;18(4):502-508. doi:10.1111/vco.12570 39. Vandesompele J, De Preter K, Pattyn F, et al. Accurate normalization 20. Allenspach K, Wieland B, Gröne A, Gaschen F. Chronic enteropathies of real-time quantitative RT-PCR data by geometric averaging of mul- in dogs: evaluation of risk factors for negative outcome. J Vet Intern tiple internal control genes. Genome Biol. 2002;3(7):research0034. Med. 2007;21(4):700-708. doi:10.1111/j.1939-1676.2007.tb03011.x doi:10.1186/gb-2002-3-7-research0034 21. Purina Institute. Purina fecal scoring chart. https://www. 40. Zahm AM, Thayu M, Hand NJ, Horner A, Leonard MB, Friedman JR. purinainstitute.com/sites/g/files/auxxlc381/files/2021-04/fecal- Circulating microRNA is a biomarker of pediatric crohn disease. chart.pdf. Accessed February 7, 2022 J Pediatr Gastroenterol Nutr. 2011;53(1):26-33. doi:10.1097/MPG. 22. Bybee SN, Scorza AV, Lappin MR. Effect of the probiotic enterococ- 0b013e31822200cc cus faecium SF68 on presence of diarrhea in cats and dogs housed in 41. Wang J, Sun Z, Yan S, Gao F. Effect of miR-145 on gastric cancer an animal shelter. J Vet Intern Med. 2011;25(4):856-860. doi:10. cells. Mol Med Rep. 2019;19(5):3403-3410. doi:10.3892/mmr.2019. 1111/j.1939-1676.2011.0738.x 10015 23. Bovens C, Tennant K, Reeve J, Murphy KF. Basal serum cortisol con- 42. Paraskevi A, Theodoropoulos G, Papaconstantinou I, Mantzaris G, centration as a screening test for Hypoadrenocorticism in dogs. J Vet Nikiteas N, Gazouli M. Circulating MicroRNA in inflammatory bowel Intern Med. 2014;28(5):1541-1545. doi:10.1111/jvim.12415 disease. J Crohns Colitis. 2012;6(9):900-904. doi:10.1016/j.crohns. 24. Jergens AE. Inflammatory bowel disease in veterinary medicine. Front 2012.02.006 Biosci. 2012;4:1404-1419. doi:10.2741/470 43. Schönauen K, Le N, Von Arnim U, Schulz C, Malfertheiner P, Link A. 25. Dandrieux JRS. Inflammatory bowel disease versus chronic enteropa- Circulating and fecal microRNAs as biomarkers for inflammatory thy in dogs: are they one and the same? J Small Anim Pract. 2016; bowel diseases. Inflamm Bowel Dis. 2018;24(7):1547-1557. doi:10. 57(11):589-599. doi:10.1111/jsap.12588 1093/ibd/izy046 26. Day MJ, Bilzer T, Mansell J, et al. Histopathological standards for 44. Ahmed FE, Jeffries CD, Vos PW, et al. Diagnostic microRNA markers the diagnosis of gastrointestinal inflammation in endoscopic for screening sporadic human colon cancer and active ulcerative colitis in biopsy samples from the dog and cat: a report from the world stool and tissue. Cancer Genomics Proteomics. 2009;6(5):281-295. small animal veterinary association gastrointestinal standardiza- 45. Yau TO, Wu CW, Dong Y, et al. microRNA-221 and microRNA-18a tion group. J Comp Pathol. 2008;138(Suppl 1):S1-S43. doi:10. identification in stool as potential biomarkers for the non-invasive 1016/j.jcpa.2008.01.001 diagnosis of colorectal carcinoma. Br J Cancer. 2014;111(9):1765- 27. Washabau RJ, Day MJ, Willard MD, et al. Endoscopic, biopsy, and his- 1771. doi:10.1038/bjc.2014.484 topathologic guidelines for the evaluation of gastrointestinal inflam- 46. Sierzega M, Kaczor M, Kolodziejczyk P, Kulig J, Sanak M, mation in companion animals. J Vet Intern Med. 2010;24(1):10-26. Richter P. Evaluation of serum microRNA biomarkers for gastric doi:10.1111/j.1939-1676.2009.0443.x cancer based on blood and tissue pools profiling: the importance of 2000 LYNGBY ET AL. miR-21 and miR-331. Br J Cancer. 2017;117(2):266-273. doi:10.1038/ 64. Zhang J, Cao Z, Yang G, You L, Zhang T, Zhao Y. MicroRNA-27a (miR- bjc.2017.190 27a) in solid tumors: a review based on mechanisms and clinical observa- 47. Zhu M, Huang Z, Zhu D, et al. A panel of microRNA signature in tions. Front Oncol. 2019;9:893. doi:10.3389/fonc.2019.00893 serum for colorectal cancer diagnosis. Oncotarget. 2017;8(10):17081- 65. Barisciano G, Colangelo T, Rosato V, et al. miR-27a is a master regula- 17091. doi:10.18632/oncotarget.15059 tor of metabolic reprogramming and chemoresistance in colorectal 48. Wang N, Wang L, Yang Y, Gong L, Xiao B, Liu X. A serum exosomal cancer. Br J Cancer. 2020;122(9):1354-1366. doi:10.1038/s41416- microRNA panel as a potential biomarker test for gastric cancer. Bio- 020-0773-2 chem Biophys Res Commun. 2017;493(3):1322-1328. doi:10.1016/j. 66. Liu T, Tang H, Lang Y, Liu M, Li X. MicroRNA-27a functions as an bbrc.2017.10.003 oncogene in gastric adenocarcinoma by targeting prohibitin. Cancer 49. Wu CW, Ng SSM, Dong YJ, et al. Detection of miR-92a and Lett. 2009;273(2):233-242. doi:10.1016/j.canlet.2008.08.003 miR-21 in stool samples as potential screening biomarkers for colorectal 67. Wang J, Guan X, Zhang Y, et al. Exosomal miR-27a derived from gas- cancer and polyps. Gut. 2012;61(5):739-745. doi:10.1136/gut.2011. tric cancer cells regulates the transformation of fibroblasts into 239236 cancer-associated fibroblasts. Cell Physiol Biochem. 2018;49(3):869- 50. Wang H, Lv Y, Wang C, et al. Systematic analysis of intestinal micro- 883. doi:10.1159/000493218 RNAs expression in HCC: identification of suitable reference genes in 68. Colangelo T, Polcaro G, Ziccardi P, et al. Proteomic screening iden- fecal samples. Front Genet. 2019;10:687. doi:10.3389/fgene.2019. tifies calreticulin as a MIR-27a direct target repressing MHC class I 00687 cell surface exposure in colorectal cancer. Cell Death Dis. 2016;7(2): 51. Koga Y, Yamazaki N, Yamamoto Y, et al. Fecal miR-106a is a useful e2120. doi:10.1038/cddis.2016.28 marker for colorectal cancer patients with false-negative results in 69. Liang Y, Zhao Y, Li L, et al. MicroRNA profiles in five pairs of early immunochemical fecal occult blood test. Cancer Epidemiol Bio- gastric cancer tissues and adjacent non-cancerous tissues. Oncol Lett. markers Prev. 2013;22(10):1844-1852. doi:10.1158/1055-9965. 2021;22(2):595. doi:10.3892/ol.2021.12856 EPI-13-0512 70. Mori MA, Ludwig RG, Garcia-Martin R, Brandão BB, Kahn CR. 52. Niu Y, Wu Y, Huang J, et al. Identification of reference genes for cir- Extracellular miRNAs: from biomarkers to mediators of physiology and culating microRNA analysis in colorectal cancer. Sci Rep. 2016;6: disease. Cell Metab. 2019;30(4):656-673. doi:10.1016/j.cmet.2019. 35611. doi:10.1038/srep35611 07.011 53. Kim SY, Jeon TY, Choi CI, et al. Validation of circulating miRNA bio- 71. Shin VY, Ng EKO, Chan VW, Kwong A, Chu KM. A three-miRNA sig- markers for predicting lymph node metastasis in gastric cancer. nature as promising non-invasive diagnostic marker for gastric cancer. J Mol Diagnostics. 2013;15(5):661-669. doi:10.1016/j.jmoldx.2013. Mol Cancer. 2015;14:202. doi:10.1186/s12943-015-0473-3 04.004 72. Ji D, Qiao M, Yao Y, et al. Serum-based microRNA signature predicts 54. Wu CW, Ng SC, Dong Y, et al. Identification of microrna-135b in relapse and therapeutic outcome of adjuvant chemotherapy in colo- stool as a potential noninvasive biomarker for colorectal cancer and rectal cancer patients. EBioMedicine. 2018;35:189-197. doi:10.1016/ adenoma. Clin Cancer Res. 2014;20(11):2994-3002. doi:10.1158/ j.ebiom.2018.08.042 1078-0432.CCR-13-1750 73. Peng X, Wang J, Zhang C, et al. A three-miRNA panel in serum as a 55. Li X, Zhang G, Luo F, et al. Identification of aberrantly expressed miR- noninvasive biomarker for colorectal cancer detection. Int J Biol NAs in rectal cancer. Oncol Rep. 2012;28(1):77-84. doi:10.3892/or. Markers. 2020;35(3):74-82. doi:10.1177/1724600820950740 2012.1769 74. Iborra M, Bernuzzi F, Correale C, et al. Identification of serum and tis- 56. Zheng G, Wang H, Zhang X, et al. Identification and validation of ref- sue micro-RNA expression profiles in different stages of inflammatory erence genes for qPCR detection of serum microRNAs in colorectal bowel disease. Clin Exp Immunol. 2013;173(2):250-258. doi:10.1111/ adenocarcinoma patients. PLoS One. 2013;8(12):e83025. doi:10. cei.12104 1371/journal.pone.0083025 75. Yang R, Fu Y, Zeng Y, et al. Serum miR-20a is a promising biomarker 57. Wang H, Zhang S, Yu Q, et al. Circulating MicroRNA223 is a new for gastric cancer. Biomed Rep. 2017;6(4):429-434. doi:10.3892/br. biomarker for inflammatory bowel disease. Med (United States). 2017.862 2016;95(5):e2703. doi:10.1097/MD.0000000000002703 76. Kleinschmidt S, Meneses F, Nolte I, Hewicker-Trautwein M. Retro- 58. Phua LC, Chue XP, Koh PK, Cheah PY, Chan ECY, Ho HK. Global fecal spective study on the diagnostic value of full-thickness biopsies from microRNA profiling in the identification of biomarkers for colorectal the stomach and intestines of dogs with chronic gastrointestinal dis- cancer screening among Asians. Oncol Rep. 2014;32(1):97-104. doi: ease symptoms. Vet Pathol. 2006;43(6):1000-1003. doi:10.1354/vp. 10.3892/or.2014.3193 43-6-1000 59. Chang PY, Chen CC, Chang YS, et al. MicroRNA-223 and microRNA- 77. Procoli F, Mõtsküla PF, Keyte SV, Priestnall S, Allenspach K. 92a in stool and plasma samples act as complementary biomarkers to Comparison of histopathologic findings in duodenal and ileal increase colorectal cancer detection. Oncotarget. 2016;7(9):10663- endoscopic biopsies in dogs with chronic small intestinal enter- 10675. doi:10.18632/oncotarget.7119 opathies. J Vet Intern Med. 2013;27(2):268-274. doi:10.1111/ 60. Pritchard CC, Kroh E, Wood B, et al. Blood cell origin of circulating jvim.12041 microRNAs: a cautionary note for cancer biomarker studies. Can- 78. Caulfield S, Priestnall SL, Kathrani A. Concordance of the histo- cer Prev Res. 2012;5(3):492-497. doi:10.1158/1940-6207.CAPR- pathologic diagnosis of concurrent duodenal and ileal biopsy 11-0370 specimens in dogs. Animals. 2021;11(10):2938. doi:10.3390/ 61. Juzenas S, Venkatesh G, Hübenthal M, et al. A comprehensive, cell ani11102938 specific microRNA catalogue of human peripheral blood. Nucleic Acids 79. Willard MD, Jergens AE, Duncan RB, et al. Interobserver variation Res. 2017;45(16):9290-9301. doi:10.1093/nar/gkx706 among histopathologic evaluations of intestinal tissues from dogs and 62. Ng PC, Chan KYY, Lam HS, et al. A prospective cohort study of Fecal cats. J Am Vet Med Assoc. 2002;220(8):1177-1182. doi:10.2460/ miR-223 and miR-451a as noninvasive and specific biomarkers for javma.2002.220.1177 diagnosis of necrotizing Enterocolitis in preterm infants. Neonatology. 80. del Pozo-Acebo L, Lo  pez de las Hazas MC, Margollés A, Dávalos A, 2021;117(5):555-561. doi:10.1159/000511655 García-Ruiz A. Eating microRNAs: pharmacological opportunities for 63. Verdier J, Breunig IR, Ohse MC, et al. Faecal micro-RNAs in inflamma- cross-kingdom regulation and implications in host gene and gut tory bowel diseases. J Crohns Colitis. 2020;14(1):110-117. doi:10. microbiota modulation. Br J Pharmacol. 2021;178(11):2218-2245. 1093/ecco-jcc/jjz120 doi:10.1111/bph.15421 LYNGBY ET AL. 2001 81. Link J, Thon C, Schanze D, et al. Food-derived Xeno-microRNAs: 86. Rotelli MT, Di Lena M, Cavallini A, et al. Fecal microRNA profile in influence of diet and detectability in gastrointestinal tract—proof-of- patients with colorectal carcinoma before and after curative surgery. principle study. Mol Nutr Food Res. 2019;63(2):e1800076. doi:10. Int J Colorectal Dis. 2015;30(7):891-898. doi:10.1007/s00384-015- 1002/mnfr.201800076 2248-0 82. Marabita F, De Candia P, Torri A, Tegnér J, Abrignani S, Rossi RL. Normalization of circulating microRNA expression data obtained by quantitative real-time RT-PCR. Brief Bioinform. 2016;17(2):204-212. SUPPORTING INF ORMATION doi:10.1093/bib/bbv056 Additional supporting information can be found online in the Support- 83. Link A, Balaguer F, Shen Y, et al. Fecal microRNAs as novel bio- ing Information section at the end of this article. markers for colon cancer screening. Cancer Epidemiol Biomarkers Prev. 2010;19(7):1766-1774. doi:10.1158/1055-9965.EPI-10-0027 84. Huang S, Wang J, Li J, et al. Serum microRNA expression profile as a diagnostic panel for gastric cancer. Jpn J Clin Oncol. 2016;46(9):811- How to cite this article: Lyngby JG, Gòdia M, Brogaard L, et al. 818. doi:10.1093/jjco/hyw085 Association of fecal and serum microRNA profiles with 85. Song J, Bai Z, Han W, et al. Identification of suitable reference gastrointestinal cancer and chronic inflammatory enteropathy genes for qPCR analysis of serum microRNA in gastric cancer in dogs. J Vet Intern Med. 2022;36(6):1989‐2001. doi:10. patients. Dig Dis Sci. 2012;57(4):897-904. doi:10.1007/s10620- 011-1981-7 1111/jvim.16530