Exclusive enteral nutrition and infliximab combination therapy are superior to monotherapy in inducing endoscopic response in active adult Crohn’s disease

Abstract

Background:

Anti-tumor necrosis factor alpha (anti-TNFα) agents assume a significant role in the management of Crohn’s disease (CD). Exclusive enteral nutrition (EEN) is as effective as corticosteroids for inducing remission in children with active CD, but less effective in adults. Few studies have compared the relative effectiveness of these strategies in achieving endoscopic improvement and symptom relief.

Objectives:

We aimed to compare the efficacy of EEN, infliximab (IFX), and combination therapy in adult patients with CD.

Design:

Retrospective cohort study.

Methods:

Between March 2021 and June 2024, 270 adult patients with active CD were recruited and received EEN, IFX, or combination treatment for 14 weeks. Propensity-score (PS) matching was conducted to balance the baseline characteristics across the three groups. The fecal 16S rRNA sequencing and metabolomics analyses were applied to further analyze the underlying mechanism involved in combination and monotherapy.

Results:

Of the 270 reviewed patients, a total of 146 PS-matched adult CD patients who received EEN (n = 29), IFX (n = 56), and combination therapy (n = 61), respectively, were analyzed. The combination therapy of EEN plus IFX demonstrated greater improvement in endoscopic response at week 14 compared to both the EEN group (p = 0.021) and the IFX group (p = 0.032), but not in endoscopic remission. Combination therapy also exhibited a significantly higher clinical response rate compared with EEN (p = 0.027) and IFX (p = 0.038) monotherapy, but no significant difference in clinical remission. Furthermore, compared with IFX monotherapy, combination therapy increased the IFX trough level at week 14 (p < 0.001). Fecal 16S rRNA sequencing revealed that combination therapy successfully restored the abundance of Bacteroidetes and related genera. Targeted metabolomics further confirmed a significant increase in short-chain fatty acids (SCFA) and key metabolites (e.g., indolelactic acid) in fecal samples 14 weeks after combination therapy.

Conclusion:

Combination therapy is more effective than EEN or IFX monotherapy for short-term endoscopic and clinical response. Mechanistically, combination therapy reshapes the gut microbiota, restores levels of SCFAs, and increases key protective fecal metabolites, offering a promising strategy for the management of CD.

Plain language summary

Exclusive enteral nutrition and anti-TNF combination therapy in active adult Crohn’s disease

What is already known about this subject?

- Anti-tumor necrosis factor alpha (anti-TNFα) agents assume a significant role in the management of Crohn’s disease (CD).

- Exclusive enteral nutrition (EEN) is EEN is as effective as corticosteroids for inducing remission in children with active CD, but it shows inferior clinical efficacy in adults with CD.

- The combined utilization of these approaches and their efficacy in achieving endoscopic improvement, symptom alleviation, and the underlying mechanisms in adult CD have not yet been comprehensively elucidated.

What are the new findings?

- Compared with EEN or IFX monotherapy, combination therapy is more effective in inducing short-term endoscopic response and clinical response.

- Combination therapy markedly increases the serum trough levels of IFX and improves the nutritional status compared with IFX monotherapy.

- Mechanistically, combination therapy restores the abundance of Bacteroidetes and related genera, and enhances short chain fatty acids and key metabolites (e.g., indolelactic acid) in fecal samples.

How might it impact clinical practice in the foreseeable future?

- EEN and IFX combination therapy is effective in inducing short-term endoscopic response, clinical response, and improvement in nutritional status in adult CD patient. Optimizing the nutritional status of patients with CD plays a crucial role in the therapeutic management of the disease, especially when combined with biologic agents.

Introduction

Crohn’s disease (CD), a chronic idiopathic inflammatory bowel disease (IBD), frequently disrupts intestinal architecture, leading to fibrotic strictures, fistulas, abscesses, intestinal dysfunction, and impaired health-related quality of life.^1,2 The current perspective suggests that the Western lifestyle, including changes in dietary habits, industrialization, and urbanization, has been considered as one interpretation for this global increase in IBD. In particular, a Western dietary habit, high in protein and fat content and low in vegetables and fruits, contributes to the dysbiosis of the intestinal microbiota that is now identified as a key initiator of gut inflammation in patients with IBD.^1,3–5 Nutritional therapy, especially exclusive enteral nutrition (EEN), is recommended as the first-line induction therapy for active CD in children in North America and Europe.^6,7 More importantly, EEN has been shown to be effective in bringing active CD in children into remission as successfully as corticosteroids, with fewer side-effects and higher mucosal healing rates. ⁸ A recent cohort study further demonstrated that EEN therapy is more efficacious in achieving clinical remission and is associated with long-term steroid-free remission without the use of biologics or the need for intestinal surgery in pediatric CD. ⁹ Mechanistically, the fecal metagenome in children reveals that EEN significantly enhances the abundance of genes involved in spermidine/putrescine biosynthesis and the shikimate pathway. ¹⁰

Despite its extensive use in pediatric CD, EEN is less commonly employed in adults due to significant challenges related to tolerability, adherence, and efficacy.^11–14 The mainstay treatment strategy for active CD remains biologics, such as anti-tumor necrosis factor alpha (anti-TNF) monoclonal antibody (mAb) therapy (e.g., infliximab, IFX). Controlled clinical trials have demonstrated the efficacy and safety of anti-TNF mAb in the treatment of moderate to severe CD for inducing and maintaining clinical and endoscopic remission, reducing hospitalizations, relapse rates, and the need for surgery, while improving quality of life.^15–18 Additionally, there has been emerging evidence of alterations in the gut microbiota during IFX therapy in recent years. IFX therapy modifies the gut microbiota diversity in patients with active CD, characterized by a decrease in Firmicutes, Lachnospiraceae, and Blautia.^19,20 The fecal metabolomic analysis further reveals a significant association between butyrate and substrates involved in its synthesis and clinical remission following anti-TNF therapy. ²¹ However, emerging studies have shown that the introduction of anti-TNF treatment has not yielded the anticipated reductions in population rates of IBD-related hospitalizations or intestinal resections.^22–24 Thus, there is an urgent need for additional therapeutic options that can effectively alleviate both clinical and endoscopic disease activity in patients with IBD.

Currently, to the best of our knowledge, no existing evidence directly compares the effectiveness, complication profiles, and taxonomic characteristics of the gut microbiota between combination therapy with anti-TNF agents and EEN versus either anti-TNF or EEN monotherapy in adult patients with CD. We conducted an initial retrospective study to assess and compare the efficacy, safety, and metabolic profiles of IFX monotherapy, EEN monotherapy, and the combination therapy as induction treatments in adult patients with active CD.

Methods

Patients and study design

The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement. ²⁵ This study was conducted as an observational retrospective analysis and was approved by the Institutional Review Board of the Shanghai Tenth People’s Hospital of Tongji University (SHSY-IEC-5.0/23XJS2/P01). The informed consent was waived due to the retrospective nature of the study.

Demography

Between March 2021 and June 2024, 270 patients with active CD were reviewed at the Center for IBD Research, Shanghai Tenth People’s Hospital of Tongji University. All of the CD patients enrolled in this study were diagnosed according to established clinical, endoscopic, radiological, and pathological criteria. ²⁶ Patients were categorized according to age at disease onset (A1: <17 years, A2: 17–40 years, or A3: >40 years), location of disease (L1: ileal, L2: colonic, or L3: ileocolonic), and disease behavior (B1: nonstricturing and nonpenetrating, B2: stricturing, or B3: penetrating). ²⁷ The inclusion criteria involved: (1) adult patients (18 ⩽ age ⩽ 75); (2) confirmed diagnosis of CD for at least 3 months; (3) active CD, defined as a Crohn’s Disease Activity Index (CDAI) score of 150–450 points at enrollment ²⁸ ; (4) the endoscopic photographs, follow-up, and hospitalization data were available. The exclusion criteria involved: (1) history of intestinal surgery; (2) biologic therapy and enteral nutrition use within 8 weeks; (3) disease extent could not be evaluated; (4) unable to undergo conventional ileocolonoscopy or IBD unidentified; (5) demographic, disease characteristic, or medical history data were not available.

Assessment of the outcomes

All of the patients had regular follow-up and received an ileocolonoscopy evaluation and imaging examination. Ileocolonoscopy was conducted at screening and at week 14. The endoscopic images were reviewed and evaluated using the simple endoscopic score for CD (SES-CD; score range 0–56) by two experienced readers (L.C. and Y.S.). Endoscopic response is defined as a 50% reduction in the SES-CD score from baseline. Endoscopic remission is defined as an SES-CD ⩽4.^29,30 Clinical disease activity is assessed using the CDAI. ²⁸ Clinical response is defined as a 70-point reduction in CDAI from baseline. Clinical remission is defined as CDAI <150. The Nutritional Risk Score (NRS2002) was used to evaluate the risk of malnutrition. ³¹ The score <3 is defined as no nutritional risk, and a score ⩾3 is regarded as indicating a risk of malnutrition.

Study endpoints

The primary endpoint of this retrospective study was the proportion of patients achieving endoscopic response at week 14, defined as a ⩾50% reduction in the SES-CD from baseline. The secondary endoscopic endpoint was the proportion of patients achieving endoscopic remission (defined as SES-CD ⩽4) at week 14. Secondary clinical endpoints were the proportion of patients with a clinical response (defined as a ⩾70-point decrease in CDAI from baseline), clinical remission (defined as a CDAI score <150), and change in NRS2002 at week 14. Changes from baseline in SES-CD, CDAI, and NRS2002 scores were assessed. Exploratory endpoints included the changes in the biomarkers, including serum C-reactive protein (CRP), albumin (Alb), and fecal calprotectin (FCP) levels.

Therapeutic drug monitoring

The serum samples were obtained from all CD patients at week 14 after IFX therapy, which was administered according to a protocol as described below, and were sent to the Qingdao Bodhi Huisheng Medical Laboratory (Qingdao, China) for further analysis. The serum trough concentrations of IFX (s-IFX) were subsequently determined using a commercially purchased ELISA kit (Tarcine BioMed Inc., Beijing, China), following the guidelines provided by the manufacturer. All samples were analyzed in a single run, and the results were presented in μg/mL. Levels of free anti-IFX antibodies (ATI) were assessed concurrently with drug concentrations using an ELISA kit.

Sample collection

Fecal samples were collected from healthy controls (HC) and patients with active CD at baseline; these patients had not received any probiotics or antibiotics for 4 weeks prior to enrollment. In addition, fecal samples were collected from patients who achieved a clinical response at week 14 after treatment with EEN, IFX, or combination therapy. The stool samples were kept frozen before DNA was extracted or metabolomic profiling was determined.

Treatment

EEN (enteral nutritional powder AA, Elental^®; EA Pharma Co., Ltd, Tokyo, Japan) was administered via a nasogastric tube for 14 weeks. Elental is a meticulously blended powder mixture that contains a diverse range of amino acids, carbohydrates, vitamins, minerals, and a minimal amount of fat. It can be easily digested and absorbed through the digestive tract. Each plastic bottle contains 80 g of the product. All patients receiving enteral nutrition through a nasogastric tube underwent standardized training provided by experienced nurses on proper insertion and post-insertion care of the tube. The expected energy requirement was calculated based on factors including age, sex, height, weight, and an assumed “low active” physical activity level. IFX (Cilag AG, Schaffhausen, Switzerland) was administered via intravenous infusion at a dose of 5 mg/kg over 30 min on day 1 (week 0), followed by doses at weeks 2, 6, and 14, and every 8 weeks thereafter. The stool samples were kept frozen until DNA was extracted and metabolomic profiling was determined.

16S rRNA gene sequencing analysis

DNA was extracted from fecal samples using the QIAamp DNA Stool Mini Kit (Hilden, Germany) in accordance with the manufacturer’s protocols. The V3–V4 variable regions of the 16S rRNA genes were amplified using universal primers (343F: 5′-TACGGRAGGCAGCAG-3′; 798R: 5′-AGGGTATCTAATCCT-3′) for bacterial diversity analysis. PCR products were purified using Agencourt AMPure XP beads (Beckman Coulter, Brea, CA, USA) and quantified with the Qubit dsDNA Assay Kit (Cat. Q32854; Life Technologies, Carlsbad, CA, USA). End-pairing cycle sequencing was performed on an Illumina NovaSeq 6000 platform (Illumina Inc., San Diego, CA, USA) following concentration adjustment, with a read length of 250 bases per cycle. ³²

The raw data were analyzed using a method as established previously. ³³ Briefly, the primer sequences were removed using the Cutadapt software. The raw data from the previous step were subsequently processed through quality filtering, noise reduction, sequence splicing, and chimera removal based on the default parameters in QIIME2, employing DADA2 to obtain representative sequences and amplicon sequence variant (ASV) abundance tables. The representative sequences for each ASV were selected using QIIME2 and annotated taxonomically with the Silva database (v138).

Sample processing and gas chromatography-mass spectrometry analysis of the metabolomic profile

The metabolites were extracted by thoroughly mixing 30-mg stool samples with 800 μL of pre-cooled methanol (Sigma-Aldrich, St. Louis, MO, USA). Following centrifugation at 14,000 rpm for 15 min at 4°C, the supernatant was collected and dried. The extracts were subsequently mixed with a methoxyamine hydrochloride pyridine solution (Sigma-Aldrich) and subjected to derivatization. The derivatized samples were analyzed using an Agilent 7890B gas chromatography (GC) system connected to an Agilent 5977A mass selective detector system (Agilent Technologies Inc., Santa Clara, CA, USA). The DB-5MS fused-silica capillary column (30 m × 0.25 mm × 0.25 µm; Agilent J & W Scientific, Folsom, CA, USA) was employed to separate the derivatives at a constant carrier gas flow rate of 1 mL/min. An untargeted metabolomic analysis was conducted employing an Agilent 7890A/5975C gas chromatography-mass spectrometry system (GC-MS; Agilent Technologies Inc.). The GC-MS data were preprocessed using ChemStation software (Version E.02.02.1431; Agilent Technologies Inc.) and Chroma TOF software (Version 4.34; LECO, St. Joseph, MI, USA). Metabolites were annotated based on the untargeted GC-MS databases provided by Lumingbio and NIST. Principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA) were employed to visualize intergroup differences. Significant metabolites were selected based on the OPLS-DA model, applying a threshold of variable influence on projection values greater than 1 and p values less than 0.05, as determined by a two-tailed Student’s t-test.

Short-chain fatty acids (SCFAs) were extracted from 20 mg of fecal samples using 500 μL of water containing 10 g/mL hexanoic acid-d3 as an internal standard. After fecal samples were centrifuged at 15,000 rpm for 15 min at 4°C, the supernatants were mixed with an equal volume of ethyl acetate containing 5% sulfuric acid. Subsequently, the mixture was subjected to vortexing, centrifugation, and incubation at 4°C for 30 min. Standard mixtures of seven SCFAs (acetic acid, propionic acid, butyric acid, valeric acid, hexanoic acid, isobutyric acid, and isovaleric acid) were prepared using the same procedure.

Statistical analysis

Induction therapy in the EEN group was determined through shared decision-making between the physician and the patient; therefore, randomization was not applied in this group. To decrease selection bias resulting from the nonrandomized retrospective design, propensity-score matching (PSM) was conducted to balance the baseline characteristics across the three groups. The propensity score (PS) is the conditional probability of receiving an exposure given a vector of measured covariates. Given the smaller number of patients in the EEN group, we matched its PSs with those of the IFX group and the combination treatment group, respectively. In our study, PS for all patients was assessed by multiple logistic regression models applying the following baseline characteristics as covariates: age, sex, gender, CDAI score, disease location, and behavior. PS was generated using SPSS (version 22.0; SPSS, Chicago, IL, USA), R 2.12.1 (http://cran.r-project.org), and an R plug-in for SPSS.22. For the matched cohort analysis, patients treated with EEN were matched with patients treated with IFX and patients treated with combination therapy separately, according to the PS, using a 1:3 matching strategy to retain as many patients as possible, with a caliper width of 0.15.

Before and after PS matching, continuous variables were presented using standard descriptive statistics such as mean ± standard deviation. The Chi-square test or Fisher’s exact test was performed to compare categorical data, including patient sex and other clinical features. Continuous outcome measures were analyzed using the Student’s t-test or Mann–Whitney U test as appropriate. The statistical significance was defined as a p value less than 0.05. All statistical analyses were conducted using the Statistical Package for the Social Sciences version 22.0 software (SPSS Inc.).

Results

Patient characteristics

Of all patients with active CD (n = 270) enrolled in this study (Figure 1), 245 patients received either EEN (n = 29), IFX (n = 116), or a combination of EEN and IFX (n = 100), and 25 patients were excluded due to withdrawal of consent, loss to follow-up, adverse events, lack of efficacy or CD flare. After PSM, a total of 146 patients with active CD who had undergone induction therapy with EEN (n = 29), IFX (n = 56), or a combination of EEN and IFX (n = 61) in the matched cohort remained for further processing (Figure 1). The mean baseline age of the patients was 44 years (30–58 years), and the mean disease duration was 24 months (8–68 months). The clinical characteristics of the patients included in this study are presented in Table 1.

Figure 1.

Flowchart.

Table 1.

Baseline characteristic of all patients with active CD.

Group	EEN	Before match		After match		Before match		After match
		IFX	p Value	IFX	p Value	IFX + EEN	p Value	IFX + EEN	p Value
Female/male	10/19	32/84	0.464	18/38	0.828	36/64	0.881	21/40	0.996
Age	25.10 ± 7.90	29.82 ± 11.54	0.07	26.27 ± 9.43	0.805	29.29 ± 10.50	0.023	25.70 ± 7.59	0.583
Duration, months	19.53 ± 22.09	29.76 ± 35.86	0.044	28.78 ± 11.96	0.044	30.30 ± 29.90	0.013	29.57 ± 28.91	0.019
Background treatment
5-ASA (%)	15(51.72)	76(65.52)	0.169	32(57.14)	0.634	58(58.00)	0.548	33(54.10)	0.066
Steroids (%)	4(13.79)	17(14.66)	0.906	8(14.30)	0.951	18(18.00)	0.803	7(11.48)	0.756
Immunosuppressant
MTX (%)	0	7(6.03)	/	2(3.57)	/	5(5.00)	/	2(3.28)	/
AZA (%)	6(20.69)	15(12.93)	0.443	7(12.50)	0.499	11(11.00)	0.295	7(11.48)
Anti-TNF (%)	3(10.34)	11(9.48)	0.889	4(7.14)	0.686	9(9.00)	0.828	2(3.28)	0.4
CRP	34.23 ± 26.38	33.14 ± 42.31	0.198	28.78 ± 11.96	0.643	29.25 ± 34.56	0.024	30.58 ± 40.31	0.066
FCP	833.04 ± 380.53	749.05 ± 535.80	0.1	690.82 ± 383.19	0.082	728.95 ± 414.42	0.03	780.33 ± 566.26	0.115
Platelet	359.90 ± 129.01	307.45 ± 106.68	0.065	314.04 ± 110.98	0.146	310.11 ± 51.88	0.091	311.52 ± 55.22	0.149
Alb	34.24 ± 7.19	39.16 ± 5.63	0.0001	39.37 ± 3.40	0.619	40.04 ± 6.41	0.0001	40.54 ± 6.54	0.0001
Location a
L1 (%)	9(31.03)	48(41.38)	/	22(39.29)	/	22(22.00)	/	18(29.51)	/
L2 (%)	0	10(8.62)	/	0	/	19(19.00)	/	1(1.64)	/
L3 (%)	20(68.97)	58(50.00)	/	34(60.71)	/	59(59.00)	/	42(68.85)	/
Behaviors b
B1 (%)	26(82.76)	62(53.45)	/	41(73.21)	/	64(64.00)	/	49(80.33)	/
B2 (%)	3(10.34)	44(37.93)	/	13(21.43)	/	29(29.00)	/	9(14.75)	/
B3 (%)	2(6.90)	10(8.62)	/	2(3.57)	/	7(7.00)	/	3(4.92)	/
CDAI	200.51 ± 35.86	189.57 ± 40.17	0.05	190.06 ± 37.12	0.131	216.04 ± 60.36	0.373	205.57 ± 48.89	0.904
SES-CD	14.52 ± 6.71	13.26 ± 9.60	0.053	14.02 ± 7.18	0.619	16.23 ± 10.17	0.93	14.34 ± 9.54	0.368
BMI	18.37 ± 2.83	20.92 ± 3.89	0.001	19.54 ± 3.70	0.138	18.21 ± 2.23	0.531	18.41 ± 2.43	0.887
NRS2002	2.62 ± 0.49	1.61 ± 1.42	0.004	1.66 ± 1.15	0.044	2.31 ± 1.16	0.994	2.31 ± 1.13	0.888

^aL1, ileum only; L2, colon only; L3, ileum and colon.

^bB1, inflammatory; B2, stricturing; B3, penetrating.

5-ASA, 5-aminosalicylate; Alb, albumin; AZA, azathioprine; BMI, body mass index; CD, Crohn’s disease; CDAI, Crohn’s disease activity index; CRP, C-reactive protein; EEN, exclusive enteral nutrition; FCP, fecal calprotectin; IFX, infliximab; MTX, methotrexate; NRS2002, Nutritional Risk Score; SES-CD, simple endoscopic score for CD; TNF, tumor necrosis factor.

Endoscopic disease activity

Overall, 49 (33.56%) out of 146 patients demonstrated an endoscopic response, and 35 (23.97%) achieved endoscopic remission by week 14. In the subgroup analysis, the combination therapy demonstrated a significantly greater improvement in endoscopic response compared to both the EEN (45.90% vs 20.69%, p = 0.021) and the IFX treatment (45.90% vs 26.79%, p = 0.032), while no significant difference was seen between the EEN and IFX monotherapy separately at week 14 (p = 0.605; Figure 2(a)). Furthermore, the endoscopic remission rates were higher in the combination therapy compared with both the EEN (29.51% vs 17.24%, p = 0.21) and the IFX monotherapy (29.51% vs 21.43%, p = 0.32), while there was no statistically significant difference between the EEN and IFX monotherapy (p > 0.05; Figure 2(b)).

Figure 2.

Endoscopic and clinical responses after 14-week treatment with EEN, IFX, or combination therapy. Endoscopic response (a), Endoscopic remission (b), clinical response (c), and clinical remission (d) following a period of 14-week treatment with EEN, IFX, and combination therapy, respectively.

EEN, exclusive enteral nutrition; IFX, infliximab; NS, no statistical difference; Mann–Whitney U test.

Clinical response and remission after induction therapy

The vast majority of patients with CD exhibited mild to moderate active disease, with a mean CDAI of 198.93 ± 42.49 at baseline (Table 1). Eighty-eight (60.27%, 88/146) of patients in all three groups exhibited a clinical response to treatment at week 14. In the subgroup analysis, the combination therapy demonstrated a significantly higher clinical response rate compared to both EEN (72.13% vs 48.28%, p = 0.027) and IFX monotherapy (72.13% vs 53.57%, p = 0.038; Figure 2(c)). The clinical remission rates from baseline to week 14 in the EEN, IFX, and combination therapy groups were 34.48% (10/29), 46.43% (26/56), and 50.82% (31/61), respectively (p > 0.05; Figure 2(d)).

Changes in secondary clinical endpoints and exploratory endpoints

FCP is a reliable biomarker that accurately predicts endoscopic activity and may replace routine endoscopic assessment for evaluating disease activity. Patients receiving combination therapy had significantly higher rates of FCP <250 mg/g and FCP <50 mg/g compared to those receiving EEN (p < 0.05) or IFX monotherapy (p < 0.05), respectively (Table 2). By 14 weeks post-therapy, patients in the combination therapy had higher levels of serum Alb than those in the IFX monotherapy (Table 2). Of note, we found that patients in the combination therapy group exhibited greater improvement in NRS2002 compared with IFX monotherapy (p < 0.05; Table 2).

Table 2.

FCP, CRP, Alb, and NRS2002 changes in CD patients receiving the EEN, IFX, and combination therapy at week 14.

	EEN	IFX	IFX + EEN	p (IFX + EEN vs EEN)	p (IFX + EEN vs IFX)	p (IFX vs EEN)
FCP
<50 μg/g	3/29	11/56	22/61	0.011	0.049	0.36
<250 μg/g	14/29	33/56	47/61	0.006	0.035	0.35
CRP				0.001	0.009	0.262
<1 mg/dL	15/29	36/56	52/9	0.001	0.009	0.262
Alb	45.04 ± 4.78	43.73 ± 5.32	45.73 ± 4.68	0.517	0.033	0.27
NRS2002 (scores)	0.66 ± 0.86	1.27 ± 1.14	0.70 ± 0.90	0.86	0.005	0.015

Alb, Albumin; CD, Crohn’s disease; CRP, C-reactive protein; EEN, exclusive enteral nutrition; FCP, fecal calprotectin; IFX, infliximab; NRS2002, Nutritional Risk Score.

The combination therapy increases the serum trough level of IFX

The s-IFX and ATI levels were measured in all patients who received IFX at week 14 post-induction therapy. The s-IFX and ATI levels were then compared between the IFX monotherapy group and the combination therapy group. The mean s-IFX levels in the combination therapy were higher than those in the IFX monotherapy (p < 0.001; Figure 3). In addition, no statistically significant difference was observed in ATI between the two groups (p > 0.05).

Figure 3.

The serum trough levels of IFX at week 14 in the IFX monotherapy and IFX plus EEN combination therapy groups. The serum samples were obtained from IFX monotherapy and IFX plus EEN combination therapy at week 14, which was administered according to a protocol as described. The serum trough levels of IFX 14 weeks post-therapy.

EEN, exclusive enteral nutrition; IFX, infliximab; NS, no statistical difference; Mann–Whitney U test.

Safety

During the course of 14 weeks, the EEN treatment showed favorable safety. Two non-serious opportunistic infections (both candidiasis) and 1 case of tuberculosis were identified in the IFX monotherapy group during the study. Mild abnormal liver enzyme levels occurred in three patients in the IFX monotherapy, one patient in the EEN monotherapy, and one patient in the combination treatment (Table 3). Eight patients (14.29%) in the IFX monotherapy and seven patients (11.48%) in the combination therapy discontinued IFX after the induction phase due to worsening of CD. Overall, no cases of malignancy, death, congestive heart failure, allergic reaction, or other serious opportunistic infections were reported.

Table 3.

Safety data for all patients during the 14-week treatment period.

	EEN	IFX	IFX + EEN
Nasopharyngitis	1	0	0
Nausea	2	0	3
Abdominal pain	0	0	1
Opportunistic infections
Candidiasis	0	2	0
Tuberculosis	0	1	0
Hepatic-related adverse event	1	3	1

EEN, exclusive enteral nutrition; IFX, infliximab.

Reconfiguration of the gut microbiota during EEN, IFX, and combination treatment

Increasing evidence has shown that alterations in the compositions and functions of the gut microbiota are associated with the efficacy of EEN or IFX, particularly in pediatric patients with active CD. We then investigated the microbiota community composition and diversity of HC and CD patients at baseline and those who received EEN, IFX, or combination therapy and achieved a clinical response at week 14. After sequence processing and filtering, a total of 11,615 ASV representative sequences were obtained and annotated from samples of 19 HCs, 9 patients with active CD who had not received any treatment, 8 patients who received EEN monotherapy, 13 patients who received IFX monotherapy, and 8 patients who received combination therapy. Patients with active CD in the EEN group, the IFX group, and the combination group exhibited a marked decrease in bacterial diversity compared to HC at the Chao1, Observed species, ACE, and PD whole tree index levels (α-diversity; Figure S1(A)–(D)). PCA analysis of β diversity distance matrices demonstrated that the first two coordinates were able to separate samples between HC and CD groups; however, the differences were not significant among the active CD, EEN, IFX, and combination groups (Figure S1(E)).

The fecal 16S rRNA sequencing results of microbiota community compositions were further subjected to detailed analysis. The relative abundance of Bacteroidaceae and Bacteroides was significantly reduced in the pretreatment group of CD patients compared to HCs, but it was restored in patients following IFX monotherapy and combination therapy (Figure 4(a) and (b)). Conversely, the relative abundances of Gammaproteobacteria, Proteobacteria, and Enterobacteriaceae were higher in the pretreatment group of CD patients than in the HC group, but decreased by week 14 after IFX treatment (Figure 4(c)–(e)). Additionally, IFX and combination therapy exhibited higher relative abundances of Firmicutes compared to EEN monotherapy (Figure 4(f)).

Figure 4.

Taxonomic differences in the fecal microbiota of patients with CD receiving IEX, EEN, and combination therapy over a period of 14-week observation. Fecal samples were collected from CD patients and HCs after a 14-week treatment period and subjected to 16S rRNA gene sequencing analysis. The relative abundance of Bacteroidaceae (a), Bacteroides (b), Proteobacteria (c), Gammaproteobacteria (d), Enterobacter (e), and Firmicutes (f).

*p < 0.05. **p < 0.01. ***p < 0.001. ****p < 0.0001.

CD, Crohn’s disease; EEN, exclusive enteral nutrition; HC, healthy control; IFX, infliximab.

The aforementioned observations further demonstrated that patients with active CD exhibited a baseline intestinal microbial community characterized by reduced richness and evenness, as well as altered composition and structure. Meanwhile, EEN, IFX monotherapy, and combination therapy effectively restored the relative abundance of Bacteroidaceae and Bacteroides, whereas IFX and combination treatment concurrently reduced the relative abundance of conditionally pathogenic bacteria such as Proteobacteria, Gammaproteobacteria, and Enterobacteriaceae.

Targeted metabolomics features refer to the therapeutic outcomes of EEN and IFX combination therapy

We subsequently attempted to identify specific features of the metabolome as an indicative of the therapeutic outcomes of EEN and IFX combination therapy. Based on the alteration of microbiota community compositions between HC and CD patients who received EEN, IFX, and combination therapy, respectively, we postulated that EEN and anti-TNF therapy might be associated with changes in fecal metabolites, specifically SCFAs. In our exploratory cohort, we employed targeted metabolomics to investigate the alterations in acetic acid, propionic acid, butyric acid, valeric acid, hexanoic acid, isobutyric acid, and isovaleric acid among the five groups. The targeted metabolomics analysis demonstrated that the levels of acetic acid, propionic acid, butyric acid, and valeric acid were significantly reduced in the pretreatment group of CD patients compared to those in the HC group. The levels of acetic acid were restored after 14 weeks of treatment with IFX monotherapy and combination therapy of IFX and EEN (Figure 5(a)–(d)). In addition, the combination therapy of IFX and EEN restored the levels of propionic acid, isobutyric acid, and isovaleric acid (Figure 5(b), (f), and (g)), but not butyric acid, valeric acid, and hexanoic acid (Figure 5(c)–(e)).

Figure 5.

Alterations in SCFAs in feces of CD patients receiving IFX, EEN, and combination therapy. Fecal samples were collected from CD patients after a 14-week treatment and HCs, and subjected to targeted metabolomics. The concentrations of acetic acid (a), propionic acid (b), butyric acid (c), valeric acid (d), hexanoic acid (e), isobutyric acid (f), and isovaleric acid (g) were determined in fecal samples using GC-MS.

*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 according to non-parametric Mann–Whitney U test.

CD, Crohn’s disease; EEN, exclusive enteral nutrition; GC-MS, gas chromatography-mass spectrometry system; HC, healthy controls; IFX, infliximab; SCFAs, short-chain fatty acids.

We further identified distinct untargeted metabolomics features 14 weeks after EEN, IFX, and combination therapy. The concentrations of indolelactic acid, L-fucose, and monolaurin were significantly higher in combination therapy at week 14 compared to those in the pretreatment group of CD patients (Figure 6(a)). The levels of indolelactic acid and sinapic acid were also increased in combination therapy relative to IFX monotherapy (Figure 6(b)).

Figure 6.

Fecal metabolomics changes in CD patients 14 weeks after IEX, EEN, and combination therapy. Fecal samples were collected from patients with CD both before and after a 14-week treatment and were subjected to untargeted metabolomics analysis. The changes in fecal metabolomics between combination therapy and untreated CD patients (a). The alterations in fecal metabolomics between the combination therapy group and the IFX monotherapy group (b).

CD, Crohn’s disease; EEN, exclusive enteral nutrition; IFX, infliximab.

Taken together, our data indicate that SCFA levels are significantly reduced in patients with active CD but are restored after 14 weeks of combination therapy. Moreover, the fecal metabolome profiles further reveal that the combination therapy is associated with elevated levels of key protective fecal metabolites, such as indolelactic acid.

Discussion

This retrospective comparative study evaluates the short-term outcomes of induction therapy with EEN, IFX, and combination therapy in a PS-matched adult CD cohort. This study demonstrates that each of the therapy strategies can ameliorate clinical symptoms of CD, but combination therapy with EEN and IFX is superior to either EEN or IFX monotherapy for inducing endoscopic and clinical response without an increased risk for complications. Furthermore, combination therapy markedly increased the serum trough levels of IFX and improved the nutritional status compared with IFX monotherapy. The fecal 16S rRNA sequencing results demonstrated that EEN monotherapy and the combination of IFX and EEN therapy effectively restored the relative abundance of Bacteroidetes and related genera. The efficacy of the combination therapy was further substantiated through targeted metabolomics analysis, showing a significant restoration of SCFA levels and elevated concentrations of key protective fecal metabolites (e.g., indolelactic acid) after a 14-week treatment regimen.

Several randomized controlled trials have shown that IFX is effective in inducing and maintaining clinical remission in patients with IBD.^15,34–37 Similarly, EEN plays a well-established role in the treatment of CD, particularly in pediatric patients. It also appears to be the first clinically validated nutritional intervention and first-line treatment strategy for active pediatric CD.^6,7,38 Several meta-analyses and randomized controlled trials have demonstrated that there is no statistically significant difference between EEN and oral corticosteroids in terms of inducing clinical remission among pediatric patients with active CD.^39–44 There have also been several studies demonstrating the effectiveness of EEN in adult patients with CD.^45,46 However, lower overall efficacy has been observed in adult CD studies using EEN in the induction of remission. Accumulating evidence has proven that EEN shows lower clinical efficacy in adult CD patients compared to pediatric patients, with response rates of 60%–85% in the latter.^47,48 While these discrepancies may be attributed to differences in disease severity, they are more likely secondary to variations in treatment compliance. Although most adult studies on EEN permit oral intake of formula, those CD patients who require nasogastric tube placement for EEN appear to have comparable efficacy to pediatric studies.^49,50 Both systematic reviews highlight EEN as effective for inducing CD remission, while emphasizing limitations such as poor tolerability and the need for personalized, solid-food alternatives to improve long-term adherence.^51,52 Recent evidence demonstrates that combination therapy with biologics and EEN is a promising approach for the treatment of CD.^53,54 Wang et al. ⁵³ primarily reported clinical and endoscopic outcomes, whereas Song et al. ⁵⁴ investigated the pharmacokinetics and body composition associated with combination therapy using biologics and EEN, compared with biologic monotherapy. In our retrospective comparative study, we specifically included a pure EEN monotherapy arm in adults, which provides a clearer understanding of the distinct and additive contribution of EEN when combined with IFX. In our study, 48.28% of the patients in the EEN monotherapy group achieved a clinical response at week 14, while only 20.69% exhibited an endoscopic response and 17.24% achieved endoscopic remission at week 14. Nonetheless, combination therapy with EEN and IFX showed greater efficacy in both endoscopic and clinical response compared with EEN or IFX monotherapy. Furthermore, the combination therapy showed the highest ratio of FCP <250 mg/g compared with the EEN and IFX monotherapy, suggesting that combination therapy is superior to EEN or IFX monotherapy in inducing clinical response.

It has been reported that adequate serum trough concentrations of IFX are positively associated with endoscopic and clinical remission. Conversely, low s-IFX levels and the presence of ATI have been correlated with a loss of response to IFX therapy.^55–57 In this study, s-IFX levels were significantly lower in the IFX monotherapy group compared with the combination treatment group at week 14, but ATI showed no obvious difference between the two groups. The combination therapy also demonstrated significant improvement in the serum levels of Alb and a reduction in NRS2002 scores compared with IFX monotherapy at week 14. The serum level of Alb, which also serves as a key biomarker for achieving clinical remission in patients with active CD,^58,59 was found to be positively correlated with s-IFX concentration. The combination therapy of EEN and IFX may exert a protective effect by increasing s-IFX—this increase is potentially mediated through elevated Alb levels and improved nutritional status.

The changes in the gut microbiome and metabonomics have been documented in numerous immune-mediated disorders, including IBD.^21,60–62 EEN emerges as a potent microbiome-targeted therapy for IBD, with recent studies linking its efficacy to patient-specific microbial shifts.^63,64 By enriching butyrate-producing taxa like Lachnospiraceae and reducing pro-inflammatory Enterobacteriaceae, EEN restores mucosal barrier function and dampens immunostimulatory potential, offering a non-pharmacological alternative to corticosteroids in pediatric CD. ⁶³ Despite the utilization of diverse methodologies and sampling conditions in previously published studies, we presented a cohesive dataset derived from CD patients recruited within the same clinical setting. Furthermore, there have been limited metabolomics studies conducted on CD patients who received combination therapy involving EEN and IFX. The fecal 16S rRNA sequencing results in our study revealed that the combination therapy of IFX and EEN promotes the restoration of Bacteroidetes and related genera, while a concurrent reduction in the relative abundance of opportunistic pathogenic bacteria was seen in patients with active CD who achieved clinical response at week 14. The targeted metabolomics further demonstrated a significant restoration of SCFA levels and elevated concentrations of key fecal metabolites, such as indolelactic acid, after combination therapy. Indole and its derivatives act as ligands for the aryl hydrocarbon receptor, inducing localized IL-22 production. This cytokine, secreted by T helper cells, neutrophils, and innate lymphoid cells, maintains intestinal homeostasis by enhancing immune defense and tissue repair.^65–67 These findings in the intestinal microbiome may partially explain the enhanced efficacy of combination therapy with EEN and IFX. In terms of safety, there were no serious adverse drug reactions. Abnormal liver enzyme levels were observed in three patients in the IFX group, one patient in the combination treatment group, and one patient in the EEN group. None of the clinically meaningful liver diseases were identified. It is noteworthy that there were no reported instances of rectal abscesses, Clostridium difficile-associated colitis, or severe infusion reactions.

The current study has several limitations. First, the retrospective characteristic of this study was unable to entirely preclude selection bias. Although PSM was conducted to control for potential confounding factors, it is possible that unmeasured or unknown variables, including patient choice, may contribute to patient stratification beyond the extent of PSM. Second, only 146 patients were included in this study, including 29 receiving EEN, 56 receiving IFX, and 61 receiving combination treatment. Thus, a fair comparison would require a prospective, randomized controlled trial with long-term follow-up. Apart from all of the above limitations, this study presented the initial short-term outcomes of induction therapy using EEN, IFX, and their combination in adult patients with active CD.

Conclusion

The current study demonstrates that the combination treatment of EEN and IFX as an initial induction therapy leads to high rates of endoscopic and clinical response, improved nutritional status, and increased s-IFX. The analysis of the intestinal microbiome further demonstrates that combination therapy results in reconfiguration of gut microbiota, significant restoration of SCFA levels, and elevated concentrations of key protective fecal metabolites. More prospective studies are required to further investigate the long-term efficacy and safety of combination treatment.