The American Journal of Clinical Nutrition 119 (2024) 885 895

Original Research Article

Ben Nichols 1,y, Anny Briola 1,y, Michael Logan1, Jaroslav Havlik2, Anna Mascellani 2, Konstantinos Gkikas1, Simon Milling 3, Umer Zeeshan Ijaz4, Christopher Quince 5, Vaios Svolos1, Richard K Russell6, Richard Hansen7,8, Konstantinos Gerasimidis 1,*

1 Human Nutrition, School of Medicine, University of Glasgow, Glasgow Royal Infirmary, Glasgow, United Kingdom; 2 Department of Food Science, Czech University of Life Sciences Prague, Prague, Czech Republic; 3 School of Infection and Inflammation, University of Glasgow, Glasgow, United Kingdom; 4 Civil Engineering, School of Engineering, University of Glasgow, Glasgow, United Kingdom; 5 Organisms and Ecosystems, Earlham Institute, Norwich, United Kingdom; 6 Department of Paediatric Gastroenterology, Hepatology and Nutrition, Royal Hospital for Children and Young People, Edinburgh, United Kingdom; 7 Department of Paediatric Gastroenterology, Hepatology and Nutrition, Royal Hospital for Children, Glasgow, United Kingdom; 8 Department of Child Health, Division of Clinical and Molecular Medicine, School of Medicine, University of Dundee, Dundee, United Kingdom

A B S T R A C T

Background: Predicting response to exclusive enteral nutrition (EEN) in active Crohn’s disease (CD) could lead to therapy personalization and pretreatment optimization. Objectives: This study aimed to explore the ability of pretreatment parameters to predict fecal calprotectin (FCal) levels at EEN completion in a prospective study in children with CD.

Methods: In children with active CD, clinical parameters, dietary intake, cytokines, inflammation-related blood proteomics, and diet-related metabolites, metabolomics and microbiota in feces, were measured before initiation of 8 wk of EEN. Prediction of FCal levels at EEN completion was performed using machine learning. Data are presented with medians (IQR).

Results: Of 37 patients recruited, 15 responded (FCal < 250 μg/g) to EEN (responders) and 22 did not (nonresponders). Clinical and immunological parameters were not associated with response to EEN. Responders had lesser (μmol/g) butyrate [responders: 13.2 (8.63 18.4) compared with nonresponders: 22.3 (12.0 32.0); P ¼ 0.03], acetate [responders: 49.9 (46.4 68.4) compared with nonresponders: 70.4 (57.0 95.5); P ¼ 0.027], phenylacetate [responders: 0.175 (0.013 0.611) compared with nonresponders: 0.943 (0.438 1.35); P ¼ 0.021], and a higher microbiota richness [315 (269–347) compared with nonresponders: 243 (205–297); P ¼ 0.015] in feces than nonresponders. Responders consumed (portions/1000 kcal/d) more confectionery products [responders: 0.55 (0.38 0.72) compared with nonresponders: 0.19 (0.01 0.38); P ¼ 0.045]. A multicomponent model using fecal parameters, dietary data, and clinical and immunological parameters predicted response to EEN with 78% accuracy (sensitivity: 80%; specificity: 77%; positive predictive value: 71%; negative predictive value: 85%). Higher taxon abundance from Ruminococcaceae, Lachnospiraceae, and Bacteroides and phenylacetate, butyrate, and acetate were the most influential variables in predicting lack of response to EEN.

Conclusions: We identify microbial signals and diet-related metabolites in feces, which could comprise targets for pretreatment optimization and personalized nutritional therapy in pediatric CD.

Keywords: exclusive enteral nutrition, Crohn’s disease, microbiome, precision therapy, metabolome, short chain fatty acids, o'link, cytokines

Abbreviations: 1H NMR, proton nuclear magnetic resonance; CD, Crohn’s disease; CRP, C-reactive protein; EEN, exclusive enteral nutrition; FCal, fecal calprotectin; FFQ, food frequency questionnaire; IBD, inflammatory bowel disease; NPV, negative predictive validity; OTU, operational taxonomic unit; PCA, principal component analysis; PPV, positive predictive validity; RF, random forest; ROC, receiver operating characteristic; SCFA, short-chain fatty acid; wPCDAI, weighted pediatric Crohn’s disease activity index.

* Corresponding author.

E-mail address: [email protected] (K. Gerasimidis). y BN and AB contributed equally to this work.

https://doi.org/10.1016/j.ajcnut.2023.12.027 Received 6 September 2023; Received in revised form 6 December 2023; Accepted 22 December 2023; Available online 19 February 2024 0002-9165/© 2024 The Authors. Published by Elsevier Inc. on behalf of American Society for Nutrition. This is an open access article under the CC BY license (http:// creativecommons.org/licenses/by/4.0/).

Introduction

Treatment with exclusive enteral nutrition (EEN) induces clinical remission in 80% of children with active Crohn’s disease (CD), but fewer patients show normalization of gut inflammatory biomarkers, such as fecal calprotectin (FCal) at treatment completion [1,2]. There is currently strong interest in stratified or personalized medicine; particularly for conditions in which response to therapies is variable, such as EEN in CD. The evolution of machine learning and the progress of high-throughput sequencing have begun to answer important questions in the etiology and management of inflammatory bowel disease (IBD), particularly through the integration of multiple parameters such as disease phenotype, blood and immune function markers, and the intestinal microbiota and its metabolites [3–6]. Such technologies and system biology may help predict therapeutic outcomes and lead to a novel understanding of underpinning mechanisms of disease pathogenesis. Integrating this approach into routine clinical care could ultimately allow patient stratification to guide treatment decisions, pretreatment optimization and therefore a more efficacious and cost-effective approach to patient care.

The literature exploring predictive factors of EEN response is sparse and mostly focused on clinical parameters such as disease phenotype [2] and disease severity during the initial period of treatment [7]. Hence, the objective of this study was to analyze, for the first time, to our knowledge, an extensive set of pretreatment factors as predictors of response to EEN. We included disease phenotype and characteristics, anthropometry, dietary intake, routine disease markers, inflammatory cytokines, plasma inflammation-related proteomic markers and diet-related bacterial metabolites, and the metabolome and microbiota in feces; all before EEN initiation. Although widely applied in clinical practice, disease activity indices, such as the weighted pediatric Crohn’s disease activity index (wPCDAI) [8], only very broadly correlate with histological and endoscopic activity [9,10]. We therefore chose FCal, which is more sensitive to detecting endoscopic activity in IBD [9,11] and thus might serve as a more appropriate biomarker for assessing response to EEN in the absence of endoscopy and as a potential “treat-to-target” biomarker.

Methods

Subjects

Children with active CD receiving EEN (Modulen IBD) for 8 wk were recruited prospectively at the Royal Hospital for Children, Glasgow, between October 2014 and May 2017. The clinical outcomes of the entire patient cohort were published previously [1]. In the current study, we included only participants who completed EEN and provided paired fecal samples at treatment initiation and completion (Supplemental Table 1 and Supplemental Figure 1). Exclusion criteria included use of antibiotics and probiotics 1 mo prior to EEN initiation, and concomitant use of other induction therapy during EEN. Patients who were administered antibiotics and probiotics during their course of EEN were subsequently excluded too. Clinical and anthropometric parameters, dietary intake, cytokines, inflammation-related blood proteomics, fecal diet-related metabolites, and metabolome and microbiota parameters were explored as predictors of response to treatment with EEN.

Disease characteristics and clinical parameters

Data on routine blood inflammatory markers [for example, Creactive protein (CRP), FCal, demographics, anthropometry, wPCDAI,

Bristol stool chart score, and disease phenotype] were collected prospectively [12]. FCal was measured in house at the end of the study (Calpro). The primary outcome utilized for prediction of EEN response was normalization of FCal concentration (250 μg/g plus 10% to account for in-house measurement assay inter- and intravariation) at EEN completion.

Dietary intake

Prior to initiation of EEN, the intake of macronutrients, fiber, and energy was estimated using the Scottish collaborative food frequency questionnaire (FFQ) for children [13]. The 148 food items in the FFQ were grouped under 16 food groups with frequency of portion consumption per day. Macronutrient intake was expressed as a percentage of total energy intake, except for fiber which was expressed as g/1000 kcal/d. Food group intake was standardized as portion/1000 kcal/d.

Plasma inflammatory cytokines and inflammation-related proteomics

The absolute concentration of 19 cytokines and relative concentration of 92 inflammation-related proteomic markers were measured in plasma with the Meso Scale Diagnostics platform (Meso Scale Diagnostics) and Olink assays (Olink Proteomics), respectively.

Diet-related bacterial metabolites and fecal metabolomics

The entire bowel motion was collected fresh in an empty, disposable container. Immediately following defecation an anaerobic sachet (Anaerocult A, Merck) was placed above the recipient container to reduce oxygen concentration and the samples were placed in a cool bag along with icepacks. Within 4 h of defecation, samples were transported to the laboratory, homogenized using mechanical kneading with a blender and aliquots were stored for downstream analyses. Aliquots for the measurement of short-chain fatty acids (SCFAs, stabilized with NaOH 1M) and total (stabilized with zinc acetate 0.11M) and free sulfide (stabilized with NaOH 1.25M) were stored in 20 C, whereas aliquots intended for LC-MS, Proton nuclear magnetic resonance (1H NMR) metabolomics and microbiota analysis were stored in 80 C. In feces, the concentration of short (SCFA), branched, and medium chain fatty acids was measured using gas chromatography coupled with a flame ionization detector (GC) (Supplemental Methods), and ammonia was measured on the day of sample collection with an automated analyzer (HI 96715, Hanna Instruments), and free/total sulfide in stored samples with colorimetric assays [14]. Fecal pH and fecal water content (%) were also measured [15]. 1H NMR metabolomics of fecal samples was performed using a 500 MHz spectrometer, using a One dimension Nuclear Overhauser Effect Spectroscopy pulse sequence with water suppression (Supplemental Methods). Comparisons were carried out in annotated and quantified metabolites [16].

Fecal microbiota

Genomic DNA was extracted from feces within 2 mo of sample collection, as described previously and in Supplemental Methods [17]. Total bacterial load in feces was measured by qPCR (forward primer: CGG TGA ATA CGT TCC CGG and reverse primer: TAC GGC TAC CTT GTT ACG ACT T) using Taqman chemistry [14]. The V4 region of the 16S rRNA gene was sequenced (MiSeq) in fecal samples using 2

250-bp paired-end reads. The V4 region was amplified (forward primer: GTGCCAGCMGCCGCGGTAA and reverse primer: GGACTACHVGGGTWTCTAAT) using fusion Golay adapters barcoded on

the reverse strand. Barcoded amplicons were purified using the Zymoclean Gel DNA Recovery Kit (D4001, Zymo Research).

Bioinformatics

Operational taxonomic units (OTUs) were constructed from the raw 16S rRNA sequencing data at a similarity of 97% using the VSEARCH pipeline (https://github.com/torognes/vsearch/wiki/VSEARCHpipeline) [18]. Paired reads were merged and quality filtering was performed with a maximum expected error value of 0.5. Sequences longer than 275 bp and shorter than 225 bp were discarded, reads were dereplicated across all samples and singleton sequences were filtered out. Chimeras were identified and eliminated using the VSEARCH implementation of the UCHIME de novo algorithm after preclustering at 98%. The UCHIME reference-based chimera detection method was then applied using the “Gold” ChimeraSlayer database [19]. OTUs were generated by clustering the remaining sequences at 97%. Taxonomic classification to genus level was performed using the Ribosomal Database Project Naïve Bayes Classifier algorithm in conjunction with the SILVA (version 123) database [20,21].

In plasma proteomic and fecal metabolome analysis, data below the limit of detection were replaced by half of the minimum detected value. For multivariate analysis, the fecal metabolite concentrations from 1H NMR analysis initially expressed as μg/g of wet fecal sample, were further normalized by total sum, log-transformed, and scaled using unit variance scaling, whereby the data are mean-centered and divided by the standard deviation for each variable.

Statistical analysis

Statistical analyses were carried out with Excel (version 2310, Microsoft Corporation) and R statistical software (version 4.2.3, R Foundation) with RStudio (version 2023.03.0, Posit Software). For nonparametric data, Mann–Whitney U tests were used for all comparisons between groups, and Spearman’s rank correlation test was performed for correlations. Continuous variables were summarized with medians and IQR, and categorical parameters were presented by counts and frequencies.

Fecal microbial community structure was analyzed using the vegan R package, and responders and nonresponders were compared in terms of α diversity, using the Chao1 richness estimate and Shannon diversity index, and β diversity, using Bray–Curtis dissimilarity, nonmetric multidimensional scaling and β dispersion [22]. For taxon abundance analysis, we normalized the dataset using total sum scaling normalization combined with centered log-ratio transformation. We removed low abundant features by keeping taxa that accounted for >0.01% of all reads. All microbial diversity and taxon abundance analyses were carried out at the OTU level.

The plasma inflammation-related proteomic profile was evaluated using the performance of principal components analysis (PCA). Separation between groups was assessed with the use of permutation analysis of variance (ANOVA) tests on Euclidean distance matrices. Discriminant proteins were identified with the use of Mann–Whitney U tests. Results of differential analysis for all datasets were corrected for multiple testing with the Benjamini Hochberg method. P values below

1. 0.05 were considered statistically significant. For multivariate analysis of the 1H NMR metabolome data, PCA and Orthogonal Projections to Latent Structures Discriminant Analysis ordination were applied to identify differences between groups using MetaboAnalystR. Levels of individual metabolites were compared using Mann–Whitney tests between groups.

Random forest modeling

For the prediction of EEN response from the various datasets, we generated random forest (RF) models, which use a machine learning algorithm widely applied for classification and prediction purposes on multiomics data [23]. RF analysis was performed using the R package, randomForest [24], separately for microbiota, 1H NMR metabolome, cytokines, inflammation-related proteomics, SCFA, dietary intake, routine clinical datasets, and inflammatory biomarkers (for example, CRP and FCal) at EEN initiation. Variable optimization was applied using the FeatureTerminatoR R package which minimizes the number of variables in the model without reducing model performance. For all models, 50,000 decision trees were grown and candidate variables at each split were set to default. To account for class imbalances, the data were stratified by response type. The importance of each feature in the model was assessed as mean decrease in Gini impurity index, which shows the change in classification accuracy between a model with and without the variable of interest, with the Gini impurity representing the probability that a specific sample will be classified incorrectly when labeled randomly. Model significance was determined after running a permutation test 1000 times using the rf.significance function in the rfUtilities R package [25]. Finally, a receiver operating characteristic (ROC) curve was plotted, and the AUC was calculated with the R package pROC. In multicomponent analysis, all datasets were combined in a single RF model. In multicomponent analysis, missing data were replaced by the corresponding variable medians.

Ethics

The study was approved by the West of Scotland Research Ethics Committee (14/WS/1004) and registered at clinicaltrials.gov (NCT02341248). All patients/carers provided informed consent. All authors had access to the study data, and reviewed and approved the final manuscript.

Results

Participant characteristics and clinical parameters

Of the 66 children with active CD, 54 completed EEN (Supplemental Figure 1). For 37 of 54 (69%) patients, FCal was measured at both baseline and EEN completion and these children were included in the current study. At EEN initiation, all patients had FCal levels of >250 μg/g. Of these, 15 of 37 (41%) displayed FCal levels of 250 μg/ g after EEN and were classified as responders. The rest of the patients with FCal levels of >250 μg/g were classified as nonresponders (n ¼ 22/37, 59%). There were no differences in disease characteristics between patients included in this study and those in the complete cohort (Supplemental Table 1). Pretreatment wPCDAI, FCal, anthropometry, disease phenotype, use of immunosuppressants, and routine inflammatory biomarkers in blood were not different between the 15 responders and the 22 nonresponders (Table 1). An RF using participants’ characteristics and clinical parameters failed to differentiate between responders and nonresponders (permutation test P ¼ 0.158).

Pretreatment dietary intake in responders and nonresponders

Treatment responders reported a higher baseline intake of confectionery and ice cream products compared with nonresponders [median (IQR), responders: 0.55 portions/1000 kcal/d (Q1: 0.38, Q3: 0.72) compared with nonresponders: 0.19 portions/1000 kcal/d (Q1: 0.01,

Pretreatment patient characteristics in responders and nonresponders to treatment with exclusive enteral nutrition

Variables1 Responders (N ¼ 15) Nonresponders (N ¼ 22) P value

Age (y) 12.1 (10.6, 14.8) 12.4 (10.3, 14.8) 0.99 Sex, male, n (%) 11 (73) 14 (64) 0.36 BMI z-score 0.74 ( 1.73, 0.06) 0.23 ( 1.19, 0.32) 0.15 Height z-score 0.43 ( 0.96, 0.27) 0.06 ( 0.53, 0.62) 0.31 Weight z-score 0.85 ( 1.52, 0.32) 0.30 ( 0.88, 0.08) 0.18 wPCDAI 27.5 (16.3, 46.3) 41.3 (27.5, 57.5) 0.15 wPCDAI classification, n (%) 0.84

Severe (wPCDAI > 57.5) 3 (20) 5 (23) Moderate (wPCDAI > 40) 3 (20) 6 (27) Mild (wPCDAI > 12.5) 9 (60) 10 (45) Remission (wPCDAI 12.5) 0 (0) 1 (5)

FCal (μg/g) (n ¼ 34) 1352 (863, 1857) 1699 (1386, 1835) 0.17 CRP (mg/L) (n ¼ 34) 5.00 (2.25, 20.5) 14.0 (3.00, 25.5) 0.49 Albumin (g/L) (n ¼ 36) 34.5 (26.8, 38.0) 32.5 (27.0, 37.0) 0.99 ESR (mm/h) (n ¼ 30) 12 (8.5, 32.5) 19 (13, 35.5) 0.49 Hemoglobin (g/L) (n ¼ 34) 126 (124, 128) 119 (111, 130) 0.39 Use of immunosuppressants, n (%) 7 (46.7) 10 (45.5) 0.94

Azathioprine, n (%) 7 (46.7) 8 (36.4) Mercaptopurine, n (%) 0 (0) 1 (4.55) Methotrexate, n (%) 0 (0) 1 (4.55)

Disease location, n (%) 0.33 L1 1 (7) 2 (9) L2 2 (13) 3 (14) L2, L4a 1 (7) 4 (18) L2, L4a, L4b 1 (7) 1 (5) L3 3 (20) 3 (14)

1. L3, L4a 0 (0) 7 (32)
2. L3, L4b 1 (7) 0 (0) L3, L4a, L4b 6 (40) 2 (9)

Perianal involvement, n (%) 2 (13) 2 (9) 1.00 FCal post-EEN (μg/g) 133 (37, 244) 1047 (735, 1705) <0.001 wPCDAI post-EEN 0 (0, 7.5) 7.5 (0, 19.4) 0.07

Abbreviations: CRP, C-reactive protein; EEN, exclusive enteral nutrition; ESR, erythrocyte sedimentation rate; FCal, fecal calprotectin; L1, distal 1/3 ileal limited caecal disease; L2, colonic; L3, ileocolonic; L4a, upper disease proximal to Ligament of Treitz; L4b, upper disease distal to Ligament of Treitz and proximal to distal 1/3 ileum; wPCDAI, weighted pediatric Crohn’s disease activity index.

1 Data are presented as median (IQR) or numbers. Chi-square tests were used for categorical variables, and Mann Whitney U tests for numerical variables.

Q3: 0.38); P ¼ 0.045] (Table 2). No other significant differences were observed in macronutrient or food group intake between the 2 groups, except for a marginally significant (P ¼ 0.09) higher fiber intake in nonresponders (Table 2). An RF model, using the intakes of food groups (portions/1000 kcal) and macronutrients (% of energy intake) as input variables, yielded an accuracy of 67% with sensitivity of 55%, specificity of 75%, positive predictive validity (PPV) of 60%, and negative predictive validity (NPV) of 71%, (permutation test P ¼

1. 0.006) to predict response (Figure 1). The foods that contributed the most to the model were confectionery, which were higher in responders, and fruit, starch, and fiber, which were all higher in nonresponders.

Pretreatment diet-related bacterial metabolites in responders and nonresponders

Pretreatment concentration (μmol/g) of fecal butyrate, measured with GC, in responders was about half that of nonresponders, when data expressed per dry [median (IQR), responders: 64.9 (Q1: 48.0, Q3: 86) compared with nonresponders: 112 (Q1: 65.7, Q3: 162); P ¼ 0.04] or per wet fecal matter [median (IQR), responders: 13.0 (Q1: 7.79, Q3: 16.9) compared with nonresponders: 20.6 (Q1: 7.99, Q3: 30.5); P ¼ 0.06] (Supplemental Table 2). This signal was not an artifact of differences in microbial load (P ¼ 0.12), or fecal water content between the 2 groups, as neither fecal water content (P ¼ 0.87) nor Bristol stool

scale (P ¼ 0.83) differed between the 2 groups (Supplemental Table 2). A similar trend (P ¼ 0.09) was observed for acetate (Supplemental Table 2). Fiber intake, from dietary analysis, correlated positively with acetate (dry matter: rho ¼ 0.50; P ¼ 0.017 and wet matter: rho ¼ 0.52; P ¼ 0.013) but not with propionate or butyrate. No significant differences were observed in feces between the 2 groups in pH, ammonia, and free or total sulfide (Supplemental Table 2). An RF using dietrelated metabolites in feces failed to predict response to EEN (permutation test P ¼ 0.144).

Pretreatment fecal 1H NMR metabolome in responders and nonresponders

Twenty-nine metabolites from 1H NMR analysis were annotated (Supplemental Table 3). No difference in global metabolome structure was found (permutation ANOVA P ¼ 0.737; Supplemental Figure 2), but significant differences were discovered for individual metabolites and, in accordance with findings observed for the same metabolites quantified with GC (Supplemental Table 2). Thus, the pretreatment concentration of fecal butyrate (μmol/g) was significantly lower in responders than nonresponders [median (IQR), responders: 13.2 (Q1: 8.63, Q3: 18.4) compared with nonresponders: 22.3 (Q1: 12.0, Q3: 32.0); P ¼ 0.03], and so was the case for the concentration of acetate [median (IQR), responders: 49.9 (Q1: 46.4, Q3: 68.4) compared with nonresponders: 70.4 (Q1: 57.0, Q3: 95.5); P ¼ 0.027], phenylacetate

Pretreatment dietary intake in responders and nonresponders to treatment with exclusive enteral nutrition

Variables1 Responders (N ¼ 11) Nonresponders (N ¼ 16) P value Energy and nutrient intake

Energy intake (kcal) 1322 (1213, 1658) 1525 (1365, 1712) 0.42 %EAR 66.0 (50.3, 84.6) 64.7 (52.9, 81.2) 0.86 Protein (g) 49.7 (42.7, 57.7) 53.5 (44.7, 65.9) 0.61 % energy from protein 13.8 (12.6, 16.2) 13.6 (12.8, 14.7) 0.67 Fat (g) 52.4 (46.4, 63.1) 55.3 (44.1, 58.0) 0.98 % energy from fat 34.6 (32.0, 35.7) 33.3 (30.0, 35.5) 0.36 Saturated fatty acids (g) 21.5 (18.7, 27.1) 22.0 (15.4, 25.9) 0.90 % energy from saturated fatty acids 14.5 (13.2, 16.0) 13.6 (12.4, 15.1) 0.15 Carbohydrates (g) 180 (156, 222) 215 (188, 258) 0.37 % energy from carbohydrates 54 (51.7, 55.9) 55.3 (53.2, 61.0) 0.23 Sugars (g) 77.1 (58.4, 96.2) 85.9 (66.1, 107) 0.68 % energy from sugars 24.9 (20.7, 26.7) 22.0 (20.6, 26.4) 0.61 Fiber (g) 8.7 (6.94, 10.1) 10.7 (8.76, 12.4) 0.13 Fiber/1000 kcal 6.37 (5.8, 6.78) 7.03 (6.18, 7.92) 0.09 Starch (g) 96 (85.5, 114.5) 113 (97.8, 142) 0.19 % energy from starch 27.8 (26.0, 29.3) 32.4 (26.1, 36.2) 0.16

Food groups (portions per day per 1000 kcal)

Breakfast cereals 0.823 (0.683, 1.189) 1.13 (0.356, 1.775) 0.35 Bread 0.994 (0.404, 1.381) 1.073 (0.517, 1.669) 0.68 Milk 0.85 (0.537, 1.341) 1.242 (0.637, 1.56) 0.51 Yogurt, cheese, and eggs 0.458 (0.278, 0.833) 0.342 (0.201, 0.579) 0.37

Meat 1.033 (0.883, 1.352) 0.822 (0.558, 1.369) 0.25 Fish 0.138 (0.081, 0.278) 0.215 (0.112, 0.353) 0.43 Potatoes, rice, and pasta 0.807 (0.631, 1.012) 0.885 (0.752, 1.129) 0.61 Savory dishes 1.05 (0.535, 1.147) 0.738 (0.497, 1.076) 0.65 Vegetables 0.486 (0.283, 0.955) 1.052 (0.312, 1.445) 0.27 Fruit 0.425 (0.334, 0.732) 0.768 (0.557, 1.07) 0.15 Juices and drinks 3.649 (2.407, 3.942) 2.446 (1.704, 4.218) 0.65 Jam, sugar, and spreads 0.137 (0, 0.428) 0.496 (0.046, 1.287) 0.16 Crisps, nuts, and savory snacks 0.496 (0.396, 0.696) 0.447 (0.281, 0.672) 0.61 Biscuits and cakes 1.216 (0.652, 1.366) 0.497 (0.3, 1.02) 0.11 Desserts 0.027 (0, 0.201) 0.028 (0, 0.109) 0.86 Confectionery and ice cream 0.55 (0.38, 0.72) 0.19 (0.01, 0.38) 0.045

Abbreviation: EAR, estimated average requirements. 1 Data are presented as median (IQR). Mann Whitney U tests were used for statistical analysis.

[median (IQR), responders: 0.175 (Q1: 0.013, Q3: 0.611) compared with nonresponders: 0.943 (Q1: 0.438, Q3: 1.35); P ¼ 0.021] and 3-(3hydroxyphenyl)propionic acid [median (IQR), responders: 0.013 (Q1: 0.013, Q3: 0.013) compared with nonresponders: 0.061 (Q1: 0.013, Q3: 0.308); P ¼ 0.011] (Figure 2A–D). An RF model, using the levels of all 1H NMR metabolites as input, produced an accuracy of 74% with sensitivity of 73%, specificity of 74%, PPV of 69% and NPV of 78% (permutation test P ¼ 0.02) to predict response to EEN. The most influential metabolites in this model were butyrate followed by phenylacetate and acetate with concentrations much higher in nonresponders (Figure 2E, F). Acetate correlated strongly with butyrate (r ¼ 0.72; P < 0.001) and both SCFAwith 3-(3-hydroxyphenyl)propionic acid (acetate [r ¼ 0.46; P ¼ 0.007]; butyrate [r ¼ 0.45; P ¼ 0.008]). No correlations were observed between these 3 metabolites with phenylacetate, suggesting a dependent relationship between the 3 former metabolites, but none between them and phenylacetate.

Pretreatment inflammation-related plasma proteome and cytokines in responders and nonresponders

The absolute concentration of 19 inflammatory cytokines measured with the multiplex assay did not differ between responders and nonresponders (Supplemental Table 4). There was no clustering of inflammation-related proteome structure according to the EEN response (permutation ANOVA P ¼ 0.737) (Supplemental Figure 3) and an RF

using those 19 inflammatory cytokines did not predict EEN (permutation test P ¼ 0.293). From the 92 inflammation-related proteomic markers measured with the Olink technology, the concentration of NT-3 was lower in responders [median (IQR), responders: 0.69 (Q1: 0.69, Q3: 0.69) compared with nonresponders: 1.49 (Q1: 0.69, Q3: 1.79); P ¼ 0.03], whereas the concentration of CXCL6 was higher in responders [median (IQR), responders: 11.1 (Q1: 10.5, Q3: 11.2) compared with nonresponders: 10.5 (Q1: 10.3, Q3: 10.9); P ¼ 0.046] (Supplemental Table 5). However, an RF generated with the 92 proteomic markers was not predictive of response to EEN (permutation test P ¼ 0.473).

Pretreatment fecal microbiota in responders and nonresponders

16S sequencing yielded a mean of 49,238 reads per sample (minimum: 10,039; maximum: 106,518; SD: 29,300). Microbiota Chao1 richness (α diversity) was higher in responders [median (IQR), responders: 315 (Q1: 269, Q3: 347) compared with nonresponders: 243 (Q1: 205, Q3: 297); P ¼ 0.015] (Figure 3A) but community composition (β diversity) did not differ significantly between the 2 groups (Figure 3B, C). Of the 238 OTUs, 13 OTUs were differentially abundant between groups (Figure 3D). RF analysis produced a model with an accuracy of 79%, sensitivity of 73%, specificity of 84%, PPVof 79%, and NPV of 80% (permutation test P ¼ 0.001) to differentiate between responders and nonresponders (Figure 3E, F). The most

Variable

1. FIGURE 1. Random forest classification between responders and nonresponders using the intakes of macronutrients, fiber, and food groups from food frequency questionnaires. Starch was reported as percentage energy intake and the other variables per 1000 kcal/d. The top 15 variables with the highest mean decrease in Gini impurity and ROC curve are shown. Non-RS, fecal calprotectin nonresponders; ROC, receiver operating characteristic; RS, fecal calprotectin responder.

influential OTUs, which all had a higher abundance in nonresponders than responders, were assigned to Bacteroides, Lachnospiraceae, Ruminococcaceae, and Anaerococcus (Figure 3E).

Multicomponent prediction of responses to EEN

Last, a multicomponent RF model was generated by including the entire study’s clinical parameters and omics datasets in the model, except for SCFA measured with GC to avoid replication of the same data measured with 1H NMR. The final model yielded an accuracy of 78%, sensitivity of 80%, specificity of 77%, PPVof 71%, and NPVof 85% (permutation test P ¼ 0.001) (Figure 4).

The most influential variables in this model were OTUs from Lachnospiraceae, Ruminococcaceae, and Bacteroides, phenylacetate, butyrate, and acetate, all of which were higher in nonresponders. Conversely, OTUs from Acidaminococcus and Collinsella, which were

higher in responders, were the most influential variables of response to EEN. The concentration of the 3 differential metabolites (that is, butyrate, acetate, and phenylacetate) correlated significantly with the relative abundances of several differential OTUs from the same model and, most importantly, retained the direction of their effect in predicting responses to EEN (Supplemental Figure 4).

Discussion

The present study found that pretreatment dietary, microbiota and metabolomic gut signatures can predict with high accuracy 8 of the 10 patients who will show an FCal response to EEN; thus, closing the gap routine clinical parameters were unable to fill. Pretreatment butyrate, acetate, and phenylacetate concentrations were at higher levels in nonresponders; almost double those of responders. These findings are

mol/g

mol/g

mol/g

mol/g

1. FIGURE 2. (A–D) Pretreatment discriminant 1H NMR fecal metabolites between responders and nonresponders with mean decrease in (E) Gini impurity and (F) ROC curve for random forest classification. 1H NMR, proton nuclear magnetic resonance; non-RS, fecal calprotectin nonresponder; ROC, receiver operating characteristic; RS, fecal calprotectin responder.

counterintuitive, because butyrate is associated with anti-inflammatory pathways in intestinal mucosa [26], and the levels of several butyrate-producing species have been consistently reported to be reduced in CD [27]. It is unlikely the increased pretreatment levels of fecal acetate and butyrate in nonresponders stem from diminished absorption because of more extensive inflammation in colonocytes, as the same strength of association was not observed across all SCFA measured and neither FCal levels nor wPCDAI differed between the 2 study groups. Previous research found a decrease in fecal butyrate during EEN [14,15], but it remains unclear whether this is simply an epiphenomenon of the lack of fiber in EEN feed composition [28], or if

it is causally involved in its mechanisms of action [29]. In a recent study, fecal butyrate levels associated positively with higher FCal levels during early food reintroduction post-EEN [30] and another research group showed that unfermented β fructans exacerbated inflammation in certain patients with active IBD [31]. This unexpected positive relationship between fiber intake, SCFA production, at treatment initiation, and colonic inflammation at completion of EEN requires further investigation.

Phenylacetate was also a strong predictor of FCal responses to EEN, both in single and multicomponent models. Phenylacetate is a catabolic product of phenylalanine and other aromatic compounds which is

Chao1 Richness Estimate

Beta Dispersion

NMDS2

1. 0
2. 1
3. 2

OTU Abundance Difference

RS non−RS

−1

OTU_253_Genus_Weissella

OTU_246_Family_Lachnospiraceae

OTU_89_Family_Lachnospiraceae

OTU_1190_Genus_Bacteroides

OTU_116_Genus_Alistipes

OTU_100_Genus_Anaerococcus

OTU_322_Genus_Collinsella

OTU_75_Genus_Acidaminococcus

OTU_203_Genus_Staphylococcus

OTU_132_Genus_Ruminococcaceae_UCG.005

OTU_170_Genus_Ruminococcaceae_UCG.013

OTU_21_Genus_.Eubacterium._coprostanoligenes_group

OTU_623_Phylum_Firmicutes

OTU

1. FIGURE 3. Pretreatment fecal microbiome characteristics in responders and nonresponders. (A) Chao1 richness estimate, (B) β dispersion, (C) Bray–Curtis NMDS ordination, and (D) significantly discriminant OTUs. The top 15 OTUs with the highest mean decrease in (E) Gini impurity and (F) ROC curve are shown for random forest classification. NMDS, nonmetric multidimensional scaling; non-RS, fecal calprotectin nonresponder; OUT, operational taxonomic unit; ROC, receiver operating characteristic; RS, fecal calprotectin responder.

further metabolized by selected species including Escherichia coli. Its exact role in CD has not been described but it might comprise a biomarker of metabolism of certain pathobionts, such as adherent and invasive E. coli, which have been implicated extensively in the pathogenesis of CD [32]. The fact that only diet-related bacterial metabolites and organisms which are major fiber fermenters in the gut differentiated between responders and nonresponders, in the current study, further underlines the importance of dietary factors in the management of active CD and possibly its underlying etiology.

Patients with higher microbiota diversity in feces benefited the most from EEN. Reduced bacterial richness is a consistent finding in fecal and mucosal samples of patients with CD, compared with healthy controls [27] and has been correlated with the extent of intestinal inflammation, assessed using FCal, in pediatric CD [33]. In contrast, no differences in α diversity metrics were observed between patients who had a >50% decrease in FCal levels and others who did not, in another study [34] suggesting that specific organisms may be more important than crude metrics of community structure. Indeed, in multicomponent

Variable

1. FIGURE 4. Random forest classification between responders and nonresponders using a multicomponent model with fecal OTUs, 1H NMR metabolites, Chao1 richness estimate of the fecal microbiome, cytokine, proteomic, diet-related metabolites, dietary intake, disease, and routine clinical datasets. The top 15 variables with the highest mean decrease in Gini impurity and ROC curve are shown. 1H NMR, proton nuclear magnetic resonance; non-RS, fecal calprotectin nonresponder; OUT, operational taxonomic unit; ROC, receiver operating characteristic; RS, fecal calprotectin responder.

analysis using machine learning, the influence of microbiota richness in predicting response to EEN became less important and certain bacterial taxa became more important predictors.

Although microbial signals identified could have prognostic value to screen patients whose gut inflammation will improve with EEN, and therefore personalize treatment options, their role in the primary disease pathogenesis is difficult to decipher within the current study. Randomized controlled trials can address such mechanistic questions where FCal responses to EEN will be measured after dietary or pharmacological manipulation of prognostic microbial species and metabolites identified in the present study, prior to EEN initiation. Based on the findings of this present study a potential intervention might be a diet low in precursors of phenylacetate production, such as a diet low in phenylalanine or protein. Nonetheless, common species previously associated with the pathogenesis of CD [27], such as Akkermansia

muciniphila, E. coli, and Veillonella, were not identified as predictors of EEN response in this study. It is also possible that we have observed different subsets of patients with different underlying microbial origins of CD, despite their similar immunity and disease phenotype. One subset of patients in which disease is driven by E. coli pathobionts which produce phenylacetate and in which EEN works by reducing their abundance, and another subset of patients where disease is driven by other butyrate-producing microbes such as Ruminococcaceae, for which EEN works by depleting fiber substrate they require for growth.

The main limitations of the current study include small sample size meaning that for some analyses statistical power may have been low, particularly when the original study was designed to test different primary outcomes, plus the lack of independent replication in larger studies. The results of these studies may also be relevant only to the local Scottish population of children with CD, treated for 8 wk with

Modulen IBD. Hence, replication of the current findings in a cohort of patients of different ethnic background and with use of other EEN feeds may be required before generalization of study findings can be made.

Using a multiomics approach, we identified pretreatment microbial species and diet-related metabolites associated with improvement in colonic inflammation during EEN. Should these microbial signals be replicated in independent multicenter research, this would open opportunities for personalized nutritional therapy in CD. Important dietrelated metabolites identified here can be measured quickly, noninvasively, and can be further modified with dietary, pharmacological, or other microbiome-modifying treatments prior to an EEN course.

Author contributions

The authors’ responsibilities were as follows – KG, RKR, RH: conceptualized the study; ML, VS, KGk, JH, SM: carried out patient recruitment and performed laboratory analysis; BN, AB, AM, JH, UZI, CQ: performed bioinformatics and statistical analysis; KG, RKR: supervised the project; KG, BN, AB: produced the first draft; and all authors: read and approved the final manuscript.

Conflict of interest

RKR reports speaker’s fees, travel support, and advisory boards: Nestle, AbbVie, Celltrion & Pharmacosmos. KG reports personal fees from Nutricia, research grants, and personal fees from Nestle Health Science, and personal fees from Dr Falk Pharma, Abbott, Servier, Mylan, and Baxter. The rest of the authors have no conflicts of interest to declare.

Funding

This research received funding from The Glasgow Children’s Hospital Charity, Nestle Health Science and The Leona M. and Harry B. Helmsley Charitable Trust. ML received a studentship from the Engineering and Physical Sciences Research Council (EPSRC) and the Nestle Health Science. UZI was funded by the Natural Environment Research Council (NERC NE/L011956/1) and supported by the EPSRC (EP/P029329/1 and EP/V030515/1). BN was partially funded by the Biotechnology and Biological Sciences Research Council (BB/ R006539/1).

Data availability

Anonymized data may become available to third parties after request to the corresponding author and for only for those patients who provided written consent for this specific aspect of study participation.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajcnut.2023.12.027.

References

[1] M. Logan, C.M. Clark, U.Z. Ijaz, L. Gervais, H. Duncan, V. Garrick, et al., The reduction of faecal calprotectin during exclusive enteral nutrition is lost rapidly after food re-introduction, Aliment. Pharmacol. Ther. 50 (6) (2019) 664 674, https://doi.org/10.1111/apt.15425.

1. [2] P.F. van Rheenen, M. Aloi, A. Assa, J. Bronsky, J.C. Escher, U.L. Fagerberg, et al., The medical management of paediatric Crohn's disease: an ECCOESPGHAN guideline update, J. Crohns Colitis. 15 (2) (2020) 171 194, https:// doi.org/10.1093/ecco-jcc/jjaa161.
2. [3] G.M. Douglas, R. Hansen, C.M.A. Jones, K.A. Dunn, A.M. Comeau, J.P. Bielawski, et al., Multi-omics differentially classify disease state and treatment outcome in pediatric Crohn's disease, Microbiome 6 (1) (2018) 13, https://doi.org/10.1186/s40168-018-0398-3.
3. [4] S. Kugathasan, L.A. Denson, T.D. Walters, M.O. Kim, U.M. Marigorta, M. Schirmer, et al., Prediction of complicated disease course for children newly diagnosed with Crohn's disease: a multicentre inception cohort study, Lancet 389
4. [5] A. Levine, N. Chanchlani, S. Hussey, T. Ziv-Baran, J.C. Escher, J.A. Dias, et al., Complicated disease and response to initial therapy predicts early surgery in paediatric Crohn's disease: results from the Porto group GROWTH study, J. Crohns Colitis. 14 (1) (2020) 71 78, https://doi.org/10.1093/ecco-jcc/jjz111.
5. [6] A. Ricciuto, M. Aardoom, E. Orlanski-Meyer, D. Navon, N. Carman, M. Aloi, et al., Predicting outcomes in pediatric Crohn's disease for management optimization: systematic review and consensus statements from the pediatric inflammatory bowel disease-ahead program, Gastroenterology 160 (1) (2021) 403 436.e26, https://doi.org/10.1053/j.gastro.2020.07.065.
6. [7] K. Gerasimidis, C.K. Nikolaou, C.A. Edwards, P. McGrogan, Serial fecal calprotectin changes in children with Crohn's disease on treatment with exclusive enteral nutrition: associations with disease activity, treatment response, and prediction of a clinical relapse, J. Clin. Gastroenterol. 45 (3)
7. [8] D. Turner, A.M. Griffiths, T.D. Walters, T. Seah, J. Markowitz, M. Pfefferkorn, et al., Mathematical weighting of the pediatric Crohn's disease activity index (PCDAI) and comparison with its other short versions, Inflam. Bowel Dis. 18
8. [9] N. Carman, D. Tomalty, P.C. Church, D.R. Mack, E.I. Benchimol, A.R. Otley, et al., Clinical disease activity and endoscopic severity correlate poorly in children newly diagnosed with Crohn's disease, Gastrointest. Endosc. 89 (2)
9. [10] G. Zubin, L. Peter, Predicting endoscopic Crohn's disease activity before and after induction therapy in children: a comprehensive assessment of PCDAI, CRP, and fecal calprotectin, Inflam. Bowel Dis. 21 (6) (2015) 1386 1391, https://doi.org/10.1097/MIB.0000000000000388.
10. [11] M.H. Mosli, G. Zou, S.K. Garg, S.G. Feagan, J.K. MacDonald, N. Chande, et al., C-Reactive protein, fecal calprotectin, and stool lactoferrin for detection of endoscopic activity in symptomatic inflammatory bowel disease patients: A systematic review and meta-analysis, Am. J. Gastroenterol. 110 (6) (2015) 802–819, https://doi.org/10.1038/ajg.2015.120, quiz 20.
11. [12] A. Levine, A. Griffiths, J. Markowitz, D.C. Wilson, D. Turner, R.K. Russell, et al., Pediatric modification of the Montreal classification for inflammatory bowel disease: the Paris classification, Inflam. Bowel Dis. 17 (6) (2011) 1314 1321, https://doi.org/10.1002/ibd.21493.
12. [13] M. Tsiountsioura, J.E. Wong, J. Upton, K. McIntyre, D. Dimakou, E. Buchanan, et al., Detailed assessment of nutritional status and eating patterns in children with gastrointestinal diseases attending an outpatients clinic and contemporary healthy controls, Eur. J. Clin. Nutr. 68 (6) (2014) 700 706, https://doi.org/10.1038/ejcn.2013.286.
13. [14] K. Gerasimidis, M. Bertz, L. Hanske, J. Junick, O. Biskou, M. Aguilera, et al., Decline in presumptively protective gut bacterial species and metabolites are paradoxically associated with disease improvement in pediatric Crohn's disease during enteral nutrition, Inflam. Bowel Dis. 20 (5) (2014) 861–871, https:// doi.org/10.1097/mib.0000000000000023.
14. [15] V. Svolos, R. Hansen, B. Nichols, C. Quince, U.Z. Ijaz, R.T. Papadopoulou, et al., Treatment of active Crohn's disease with an ordinary food-based diet that replicates exclusive enteral nutrition, Gastroenterology 156 (5) (2019) 1354 1367.e6, https://doi.org/10.1053/j.gastro.2018.12.002.
15. [16] J.D. Jaimes, A. Slavickova, J. Hurych, O. Cinek, B. Nichols, L. Vodolanova, et al., Stool metabolome-microbiota evaluation among children and adolescents with obesity, overweight, and normal-weight using 1H NMR and 16S rRNA gene profiling, PLOS ONE 16 (3) (2021) e0247378, https://doi.org/10.1371/ journal.pone.0247378.
16. [17] K. Gerasimidis,M.Bertz,C.Quince,K. Brunner,A.Bruce, E.Combet,etal.,The effect of DNA extraction methodology on gut microbiota research applications, BMC Res. Notes. 9 (2016) 365, https://doi.org/10.1186/s13104–016-2171-7.
17. [18] T. Rognes, T. Flouri, B. Nichols, C. Quince, F. Mahe, VSEARCH: a versatile open source tool for metagenomics, PeerJ 4 (2016) e2584, https://doi.org/ 10.7717/peerj.2584.
18. [19] B.J. Haas, D. Gevers, A.M. Earl, M. Feldgarden, D.V. Ward, G. Giannoukos, et al., Chimeric 16S rRNA sequence formation and detection in Sanger and 454pyrosequenced PCR amplicons, Genome Res 21 (3) (2011) 494 504, https:// doi.org/10.1101/gr.112730.110.

1. [20] B.J. Callahan, P.J. McMurdie, M.J. Rosen, A.W. Han, A.J. Johnson, S.P. Holmes, DADA2: high-resolution sample inference from Illumina amplicon data, Nat. Methods. 13 (7) (2016) 581 583, https://doi.org/10.1038/ nmeth.3869.
2. [21] C. Quast, E. Pruesse, P. Yilmaz, J. Gerken, T. Schweer, P. Yarza, et al., The SILVA ribosomal RNA gene database project: improved data processing and web-based tools, Nucleic Acids Res 41 (Database issue) (2013) D590 D596, https://doi.org/10.1093/nar/gks1219.
3. [22] J.B. Oksanen, G. Friendly, M. Kindt, R. Legendre, P. McGlinn, D. O'Hara, et al., Vegan: Community Ecology Package, R package version (2020) 25–27.
4. [23] X. Chen, H. Ishwaran, Random forests for genomic data analysis, Genomics 99
5. [24] A. Liaw, M. Wiene, Classification and regression by random Forest, R. News. 2
6. [25] M.A. Murphy, J.S. Evans, A. Storfer, Quantifying Bufo boreas connectivity in Yellowstone National Park with landscape genetics, Ecology 91 (1) (2010) 252 261, https://doi.org/10.1890/08-0879.1.
7. [26] J.P. Segain, D. Raingeard de la Bletiere, A. Bourreille, V. Leray, N. Gervois, C. Rosales, et al., Butyrate inhibits inflammatory responses through NFκB inhibition: implications for Crohn's disease, Gut 47 (3) (2000) 397 403, https:// doi.org/10.1136/gut.47.3.397.
8. [27] K. Gerasimidis, K. Gkikas, C. Stewart, E. Neelis, V. Svolos, Microbiome and paediatric gut diseases, Arch. Dis. Child. 107 (9) (2022) 784–789, https:// doi.org/10.1136/archdischild-2020-320875.
9. [28] M. Logan, K. Gkikas, V. Svolos, B. Nichols, S. Milling, D.R. Gaya, et al., Analysis of 61 exclusive enteral nutrition formulas used in the

1. management of active Crohn's disease-new insights into dietary disease triggers, Aliment. Pharmacol. Ther. 51 (10) (2020) 935 947, https:// doi.org/10.1111/apt.15695.
2. [29] V. Svolos, K. Gkikas, K. Gerasimidis, Diet and gut microbiota manipulation for the management of Crohn's disease and ulcerative colitis, Proc. Nutr. Soc. 23
3. [30] K. Gkikas, M. Logan, B. Nichols, U.Z. Ijaz, C.M. Clark, V. Svolos, et al., Dietary triggers of gut inflammation following exclusive enteral nutrition in children with Crohn's disease: a pilot study, BMC Gastroenterol 21 (1) (2021) 454, https://doi.org/10.1186/s12876-021-02029-4.
4. [31] H.K. Armstrong, M. Bording-Jorgensen, D.M. Santer, Z. Zhang, R. Valcheva, A.M. Rieger, et al., Unfermented beta-fructan fibers fuel inflammation in select inflammatory bowel disease patients, Gastroenterology 164 (2) (2023) 228–240, https://doi.org/10.1053/j.gastro.2022.09.034.
5. [32] C. Palmela, C. Chevarin, Z. Xu, J. Torres, G. Sevrin, R. Hirten, et al., Adherent-invasive Escherichia coli in inflammatory bowel disease, Gut 67 (3)
6. [33] K.L. Kolho, K. Korpela, T. Jaakkola, M.V.A. Pichai, E.G. Zoetendal, A. Salonen, et al., Fecal microbiota in pediatric inflammatory bowel disease and its relation to inflammation, Am. J. Gastroenterol. 110 (6) (2015) 921 930, https://doi.org/10.1038/ajg.2015.149.
7. [34] K. Diederen, J.V. Li, G.E. Donachie, T.G. de Meij, D.R. de Waart, T.B.M. Hakvoort, et al., Exclusive enteral nutrition mediates gut microbial and metabolic changes that are associated with remission in children with Crohn's disease, Sci. Rep. 10 (1) (2020) 18879, https://doi.org/10.1038/s41598-02075306-z.