Abstract
The management of localized ileocecal Crohn’s disease (CD) requires balancing escalation of medical therapy against early surgical resection. While clinical phenotyping guides initial management, standard assessments often fail to distinguish reversible inflammation from irreversible fibrosis. This limitation remains a major barrier to personalized care. This narrative review aims to review current evidence on clinical features, imaging modalities, and biomarkers relevant to intestinal fibrosis and to propose a phenotype-driven framework for treatment selection in localized ileocecal CD. To address this, we synthesized current literature on intestinal fibrogenesis, cross-sectional imaging (magnetization transfer MRI (MT-MRI), shear wave elastography (SWE)), and therapeutic outcomes. Evidence was integrated to develop a conceptual algorithm incorporating disease phenotype, quantitative imaging markers, and multidisciplinary decision-making. The proposed two-tiered framework begins with the Montreal classification (Tier 1). B1 phenotypes are directed toward early intensive treat-to-target medical strategies, while B3 phenotypes may require individualized surgical management, sometimes preceded by short-term medical optimization depending on complication severity. For the challenging B2 phenotype, we propose a precision medicine pathway (Tier 2) utilizing quantitative imaging, such as MT-MRI or SWE. This pathway is intended to help stratify B2 strictures into fibrosis-dominant, inflammation-dominant, or indeterminate phenotypes within a multidisciplinary decision-making process. A phenotype-driven approach incorporating clinical and imaging information may help inform therapeutic decision-making across the spectrum of ileocecal CD, particularly in the management of stricturing disease. The proposed framework provides a roadmap for prospective validation and future clinical trials aimed at optimizing treatment selection in stricturing disease.
Plain language summary
Crohn’s disease can behave in different ways. In some people, inflammation is the main problem. In others, bowel damage and scarring can narrow the intestine and cause a stricture. These different disease patterns may need different treatments, but in everyday practice it is often difficult to tell which process is dominant.
This review focuses on ileocecal Crohn’s disease, a common form that affects the end of the small bowel and the beginning of the large bowel. We summarize current evidence on how doctors assess disease behavior using symptoms, endoscopy, and cross-sectional imaging such as magnetic resonance enterography or computed tomography enterography. We also discuss newer imaging methods that may help estimate the amount of fibrosis, or scarring, inside a stricture.
Based on this evidence, we propose a two-tier framework. The first tier reflects current validated care: patients are assessed and classified using standard clinical, endoscopic, and imaging findings. The second tier is a proposed pathway for future research. In this pathway, quantitative imaging could be combined with multidisciplinary discussion to help distinguish inflammation-dominant disease, fibrosis-dominant disease, and cases that fall into a mixed or uncertain zone.
This approach is not yet ready for routine use and still requires prospective validation. However, it may provide a practical research roadmap for improving treatment selection, including when to intensify medical therapy and when earlier surgery may be more appropriate.
Introduction
The management of localized ileocecal Crohn’s disease (CD), a progressive condition often culminating in surgery for up to 70% of patients,1–3 has long been defined by a therapeutic dilemma: early surgical resection versus escalating medical therapy. 4 Historically, surgery was considered a last resort,4,5 a paradigm reinforced by concerns over postoperative recurrence. However, accumulating evidence has increasingly challenged this view, suggesting that early, elective surgery may offer favorable long-term outcomes for appropriately selected patients.6,7
This shift was catalyzed by the landmark LIR!C trial, 6 which established that early laparoscopic surgery was non-inferior to infliximab for quality of life and more cost-effective. Long-term data revealed that 48% of the infliximab group eventually required surgery versus only 26% of the surgical group needing subsequent anti-tumor necrosis factor (TNF) therapy. 8 These findings established clinical equipoise, reframing the question from a general comparison to a personalized one: which therapy is optimal for this specific patient?
This framework applies specifically to patients with isolated ileocecal CD (⩽40 cm involvement) without concurrent perianal fistulizing disease, mirroring LIR!C inclusion criteria. 6 Patients with extensive small bowel involvement (>40 cm) or active perianal disease require distinct management approaches beyond this scope. 9 While individual elements of this approach, such as Montreal phenotyping, advanced imaging modalities, and multidisciplinary discussion, are recognized in current guidelines, to our knowledge, these elements have not yet been integrated into a structured, testable algorithm specifically for the B2 stricturing subset. Current guidelines recommend differentiating inflammatory from fibrostenotic strictures but do not provide a standardized decision pathway for doing so.9,10 This framework is intended to address that gap. We present this framework with three objectives: (1) synthesize current evidence for phenotype-driven treatment selection, (2) propose a testable algorithm for future trials, and (3) provide flexible implementation strategies. Throughout this review, we distinguish between components that are already supported by current evidence and guideline-based practice and those that are presented as conceptual, author-proposed elements requiring prospective validation. Prospective validation is required before clinical adoption; we therefore present the illustrative PRECISION-CD trial concept as one possible validation strategy.
The evolving therapeutic landscape
The therapeutic landscape for ileocecal CD has evolved significantly over the past three decades, refining both surgical and medical strategies. 11 Surgical management has shifted toward minimally invasive laparoscopic resections with conservative margins, significantly reducing morbidity.12,13 Furthermore, innovative techniques like the Kono-S anastomosis have been developed to specifically reduce the risk of postoperative recurrence. 14 However, the inherent risk of recurrence remains a significant challenge, highlighting that while surgery effectively addresses the local inflammatory burden, the underlying disease susceptibility persists. 15 Concurrently, the medical armamentarium has expanded significantly beyond anti-TNF agents to include newer biologics and small molecules targeting alternative pathways, such as anti-integrin (e.g., vedolizumab) and anti-interleukin (e.g., ustekinumab, risankizumab, guselkumab) therapies, as well as Janus kinase (JAK) inhibitors (e.g., upadacitinib).16–20 Crucially, these newer agents targeting the integrin or interleukin (IL)-12/23 pathways have demonstrated favorable safety profiles alongside high rates of treatment persistence and success, further strengthening the role of medical therapy. 21
This expansion of therapeutic options has intensified the debate, with recent real-world evidence highlighting the complexities of treatment selection. A 2023 nationwide Danish cohort study by Agrawal and colleagues found that early surgical resection was associated with a 33% lower risk of adverse outcomes compared to anti-TNF therapy, with nearly 50% of the surgery group remaining treatment-free at 5 years. 7 However, the interpretation of such observational data requires caution. A recent nationwide Swedish cohort study attempting to replicate these findings concluded that such comparisons are heavily limited by “confounding by indication,” where patients are selected for therapy based on underlying clinical characteristics and patient preference. 22 When applying stricter, LIR!C-like inclusion criteria, the observed benefit of early surgery was no longer statistically significant.6,22 This ongoing debate is also reflected in recent contemporary discussion of whether surgery or biologic therapy should come first in ileocolic CD, further supporting the view that treatment choice remains context-dependent rather than uniformly resolved. 23
Beyond real-world cohorts, landmark randomized trials have also reshaped our understanding. The PROFILE trial, for instance, provided compelling evidence for early intensive treatment. 24 Patients receiving upfront combination therapy achieved dramatically superior outcomes compared to conventional step-up management, with a substantial difference in sustained remission and a marked reduction in the need for urgent surgery, and significantly fewer adverse and serious adverse events. While the trial’s 17-gene biomarker failed to demonstrate clinical utility for guiding therapy, the results underscored a key principle: early, effective intervention improves outcomes, even if accurately predicting who needs it remains elusive.
These parallel advances underscore a critical reality: treatment efficacy is not uniform but is instead dictated by the patient’s underlying disease phenotype. This highlights a key therapeutic ceiling: potent anti-inflammatory agents do not reverse established, irreversible fibrosis, while surgery for purely inflammatory disease may constitute overtreatment. This review proposes a conceptual precision medicine framework that synthesizes existing evidence into a structured approach for treatment selection.
Literature search and selection strategy
This is a narrative review based on a purposive, non-exhaustive literature search designed to generate hypotheses and support development of a conceptual framework rather than to provide a comprehensive systematic synthesis. A literature search was conducted across PubMed, Embase, and the Cochrane Library databases for articles published between January 2010 and September 2025. Key search terms included “Crohn’s disease,” “ileocecal,” “surgery,” “biologics,” “fibrosis,” “stricture,” “magnetization transfer MRI,” “shear wave elastography,” “contrast-enhanced ultrasound,” and “biomarkers.” Foundational mechanistic studies and seminal articles published prior to 2010 were also included to provide historical and scientific context.
Inclusion and exclusion criteria
We purposively selected studies that informed the major biological, imaging, and therapeutic pillars of the proposed framework, with emphasis on randomized controlled trials, prospective and retrospective cohort studies, international consensus statements, and major clinical guidelines focusing on the management, pathophysiology, and imaging assessment of ileocecal CD. For studies evaluating advanced imaging modalities, particular priority was given to investigations with surgical histopathology as the reference standard, prospective designs where available, and direct relevance to fibrosis characterization or treatment selection in stricturing CD. Articles were excluded if they were case reports, non-human studies (unless establishing fundamental fibrogenesis pathways), or if they primarily focused on extensive small bowel involvement (>40 cm) or active perianal fistulizing disease, which fall outside the algorithm's scope.
Data synthesis and evidence grading
Given the narrative design of this review, literature was purposively sampled and evaluated to construct the biological and clinical pillars of the proposed framework. To evaluate the strength of the literature supporting this framework while remaining consistent with the narrative nature of this review, we performed a structured qualitative appraisal of the included studies. This appraisal considered factors such as study design, sample size, methodological rigor, and consistency of findings. The resulting qualitative assessment across key domains—including imaging modalities, surgical versus medical interventions, and biomarkers—has been integrated into the operational summaries presented in Tables 1 and 2.
Evidence base and phenotype-driven management strategy for ileocecal Crohn’s disease.
B1, inflammatory; B2, stricturing; B3, penetrating; HR, hazard ratio; MDT, multidisciplinary team; RCT, randomized controlled trial.
Decision pathway for B2 stricturing disease.
Numeric thresholds are illustrative and derived from single-center cohorts (MTR: n = 119; SWE: n = 25). See Limitations for details on evidence quality.
MT-MRI serves as the primary modality (AUC > 0.90); SWE is adjunctive due to conflicting data and its “uncertain” 2024 consensus rating. Expected outcomes are extrapolated from mixed populations and require prospective validation in stricturing disease cohorts. For the proposed validation protocol, see Supplemental Appendix 2.
AUC, area under the curve; CEUS, contrast-enhanced ultrasound; EBD, endoscopic balloon dilation; MDT, multidisciplinary team; MT-MRI, magnetization transfer magnetic resonance imaging; MTR, magnetization transfer ratio; SWE, shear wave elastography.
When interpreting conflicting data, particularly the divergent prospective results regarding Shear Wave Elastography (SWE) for fibrosis differentiation, we deferred to the highest level of expert consensus. Specifically, we aligned our framework with the 2024 international ultrasound consensus, which rated elastography as “uncertain.” Consequently, SWE is explicitly positioned as an adjunctive tool within our algorithm, subordinate to modalities with more consistent validation such as Magnetization Transfer MRI (MT-MRI).
Phenotype-based initial stratification: The foundation of precision
This section combines evidence-based elements that are already aligned with current clinical practice with conceptual extensions intended to support future framework development. In particular, Montreal-based stratification for B1 and B3 phenotypes reflects established clinical practice, whereas the proposed precision pathway for B2 stricturing disease is presented as a hypothesis-generating model.
The Montreal classification, which categorizes CD behavior into three distinct phenotypes, including B1 (inflammatory), B2 (stricturing), and B3 (penetrating), provides a standardized and clinically relevant framework. 30 Phenotype-based stratification fundamentally influences treatment response and aligns therapy with underlying pathophysiology. Evidence supports clear treatment pathways for B1 and B3, thereby efficiently reserving resource-intensive precision diagnostics for the B2 stricturing subset where optimal treatment selection remains genuinely uncertain. 10
The B1 phenotype, characterized by active inflammation without structural complications, is highly amenable to medical therapy.10,30 The PROFILE trial demonstrated that early intensive therapy dramatically reduces the need for urgent surgery compared to step-up approaches (Table 1), providing strong rationale for directing B1 patients toward optimal treat-to-target strategies. 24
The B3 phenotype involves full-thickness bowel destruction causing penetrating complications (fistulas, abscesses). 30 Management is individualized: clinically significant fistulas or large abscesses (>2 cm) generally require drainage and/or surgical resection, while smaller uncomplicated abscesses may allow conservative management with antibiotics as a bridge to biologics. 10
This initial stratification isolates the greatest clinical challenge: the B2 phenotype. These patients have developed fixed luminal narrowing causing obstructive symptoms. The critical question that standard assessment cannot answer is the composition of the stricture: what proportion is due to active, reversible inflammation versus established, irreversible fibrosis? This distinction has profound therapeutic implications, as a patient’s response to medical therapy is dictated by the stricture’s composition: an inflammatory stricture may resolve, while a predominantly fibrotic one will not. 10 The historical inability to distinguish these components has led to empirical treatment, an approach that results in a significant number of failures. Prospective trials have demonstrated both the promise and the limitations of medical therapy for B2 disease. For example, the CREOLE study showed that adalimumab was successful in 64% of patients with symptomatic small bowel strictures at 24 weeks, yet a significant 36% failed to respond. 26 Similarly, the STRIDENT trial highlighted that an intensified, treat-to-target drug therapy strategy was more effective than standard care, although a 28% failure rate in the standard arm still underscored the challenge in this population. 27 This evidence reinforces that while medical therapy is a powerful option for a majority of B2 patients, a substantial minority will not benefit, highlighting the urgent need for better patient stratification. Furthermore, a post-hoc analysis of clinical trials found that patients with non-passable strictures had clinical remission rates of only approximately 50% after 1 year of treatment. 28 While this demonstrates that medical therapy can be effective for a substantial subset of B2 patients, it also means that half of these patients may endure a year of ineffective, costly therapy. This 50/50 clinical uncertainty is the core impasse that drives the need for the precision medicine approach detailed in the following sections. The management implications for each Montreal phenotype are summarized in Table 1.
The precision medicine toolkit for stricturing disease
The tools discussed below include both evidence-informed components and investigational elements. They are reviewed here to support a conceptual framework for future validation, not to imply that all modalities are currently ready for routine treatment-directing use. For patients with stricturing (B2) disease identified through phenotypic stratification, applying a three-pillar precision medicine framework is proposed to move beyond empirical approaches and objectively determine the optimal treatment strategy.
Pillar 1: Understanding fibrosis pathobiology
Intestinal fibrosis is not a passive scar but an active, self-perpetuating process driven by excessive extracellular matrix deposition, particularly collagen. 31 While mechanisms such as mechanotransduction can drive fibrogenesis independently of active inflammation, 32 it is crucial to acknowledge that intestinal strictures are rarely purely fibrotic or purely inflammatory. Histological and molecular data indicate that active inflammatory pathways frequently persist long after a stricture is considered clinically fibrotic. Recognizing this persistent biological overlap explains why potent anti-inflammatory therapies often yield partial or mixed responses, and why they ultimately fail to reverse established fibrotic strictures.26–28 Therefore, the clinical challenge and the primary goal of the proposed quantitative imaging are not to establish a strict binary distinction, but rather to identify the dominant tissue component driving the mechanical obstruction.
Critically, no specific anti-fibrotic therapies have been approved. 33 However, the therapeutic landscape is evolving. AGMB-129, an oral GI-restricted transforming growth factor (TGF)-BRI (ALK5) inhibitor, showed promising interim data in the Phase IIa STENOVA trial, demonstrating target engagement with downregulation of both fibrotic (p = 0.0036) and inflammatory (p < 0.0001) pathways. 34 The development of such agents, which inherently target this overlapping pathobiology, underscores the critical need for quantitative tools to assess fibrosis, both for surgical stratification and for identifying candidates for emerging treatments.
Pillar 2: Quantitative imaging assessment
The second pillar provides the tools to non-invasively measure fibrotic burden, overcoming the limitations of traditional imaging, which cannot reliably distinguish stricture composition (detailed technical principles and suggested protocols for these modalities are provided in Supplemental Appendix 1). 35 Widely available modalities—Intestinal Ultrasound (IUS) and MR Enterography (MRE)—provide valuable initial clues.33,36 Both are considered “appropriate” and even “required” for diagnosis. The core features used to define a stricture on imaging are bowel wall thickening, luminal narrowing, and pre-stenotic dilation.33,36 However, differentiating the composition is difficult: while MRE and IUS can help identify the inflammatory component, both consensus groups state that currently no conventional imaging modality can accurately determine the presence or degree of fibrosis.33,36 These conventional signs are qualitative and lack the precision needed for definitive stratification, necessitating more advanced techniques.33,36 Recent systematic review evidence on IUS also supports its important role in defining and monitoring small bowel strictures, while highlighting persistent heterogeneity in definitions and limitations in distinguishing fibrosis from inflammation using conventional parameters alone. 37
Magnetization transfer MRI
MT-MRI directly quantifies tissue collagen content through magnetization transfer between bound macromolecular protons (collagen) and free water protons. 38 Prospective studies using surgical histopathology as the reference standard have reported promising diagnostic accuracy for MT-MRI in selected cohorts, with area under the curve (AUC) values exceeding 0.90 for the identification of moderate-to-severe fibrosis.35,38 A validation study by Fang et al. (using 119 bowel specimens) identified a magnetization transfer ratio (MTR) threshold of >31.25% for predicting predominant fibrosis (defined as fibrosis score ⩾2), achieving sensitivity 91.3% and specificity 92.3%. 35 However, these thresholds were derived from a single-center cohort and have not yet been validated in independent multicenter studies with standardized protocols. Subsequent independent validation by Li et al. confirmed MT-MRI’s diagnostic accuracy (AUC 0.919) for intestinal fibrosis characterization, although broader external validation and protocol harmonization remain necessary before threshold-based interpretation can be generalized across centers. 38
Despite its high accuracy, the primary limitation of MT-MRI is technical. The modality requires specialized sequences, and its main challenge is protocol standardization. Absolute MTR values are highly dependent on specific acquisition parameters (e.g., field strength, pulse sequences, and timing), meaning results cannot be quantitatively compared across different studies or scanner protocols. 39 Nevertheless, researchers suggest MT-MRI is a highly promising modality that could potentially be used to help differentiate fibrotic from inflammatory strictures in clinical practice.35,38,39
Shear wave elastography
SWE, a more accessible ultrasound-based technique, measures tissue stiffness in kilopascals (kPa) or shear wave velocity (m/s) and demonstrates correlation with histological fibrosis.40–42 For 2D-SWE, Chen et al. reported that mean stiffness values were significantly higher in severe fibrosis (23.0 ± 6.3 kPa) compared to moderate (17.4 ± 3.8 kPa) and mild fibrosis (14.4 ± 2.1 kPa), with a cutoff achieving 91.7% specificity and 69.6% sensitivity (AUC 0.822). 41 For point-SWE (pSWE), Ding et al. demonstrated that a velocity threshold of >2.73 m/s achieved 75% sensitivity and 100% specificity for detecting fibrosis, with pSWE outperforming other elastography modalities. 42
However, a critical challenge emerged in 2024: the international expert consensus on IUS concluded that elastography is currently “uncertain” for accurately differentiating fibrotic from inflammatory components, primarily due to lack of standardized methodology and operator dependency.36,40 This uncertainty was reinforced by recent prospective data from de Voogd et al., who found that SWE did not significantly differentiate stricture phenotypes when using surgical histopathology as the reference standard in their Western cohort. 43
Several factors may explain these discrepant findings. First, patient population differences between Asian and Western cohorts may affect tissue properties and disease phenotypes. 44 Second, technical heterogeneity across studies—including variations in SWE modality (2D-SWE vs point-SWE), acquisition protocols, operator experience, and thresholds used—introduces substantial variability. 42 Third, the physiological complexity of mixed inflammatory-fibrotic strictures presents a fundamental challenge: active inflammation causes tissue edema and hypervascularity, which may paradoxically decrease measured stiffness despite coexisting fibrosis.40,41 This inverse relationship may explain why contrast-enhanced ultrasound (CEUS), which directly assesses vascularity, performed better than SWE in the de Voogd study. 43 Fourth, the reference standard itself has limitations, as surgical histopathology represents only the resected segment and may not fully capture stricture heterogeneity. 45
Given the conflicting evidence and methodological limitations discussed above, SWE should be positioned as an adjunctive tool rather than a primary decision criterion. Within the present manuscript, SWE is therefore discussed as an investigational adjunct within the proposed framework rather than as an evidence-established standalone basis for treatment selection. A recommended pragmatic hierarchy for fibrosis assessment is: (1) MT-MRI as primary modality when available, due to its higher evidence quality, more direct collagen measurement, and consistent validation across studies; (2) SWE as alternative modality only when MT-MRI is inaccessible, with results interpreted cautiously and never as standalone criteria.
If using SWE, protocols should require three or more concordant measurements by a certified operator,41,42 with results presented to the MDT as “suggestive” not “definitive,” and interpreted as one input among multiple data streams including conventional imaging, endoscopy, clinical status, and prognostic factors.40,46 All cases should be tracked prospectively for local validation.
Contrast-enhanced ultrasound
While MT-MRI and SWE directly quantify fibrotic tissue properties (collagen content and stiffness, respectively), CEUS offers complementary information by assessing tissue vascularity.43,47 De Voogd et al. demonstrated that CEUS parameters, particularly the wash-in area under the curve (WiAUC), were highly accurate for identifying inflammatory phenotypes, whereas SWE was not significant in their cohort. 43 This likely reflects the modalities’ different targets: CEUS excels at detecting active inflammation via hypervascularity, 43 whereas MT-MRI and SWE are optimized for established fibrosis.35,38,41,42,47 In gray zone cases where SWE and MT-MRI yield intermediate values, CEUS may serve as a valuable adjunct to quantify the degree of active inflammation, helping to differentiate inflammation-dominant from truly mixed phenotypes. 43
Other techniques
Other techniques, such as Diffusion-Weighted MRI (DWI-MRI), have also been investigated.47,48 DWI-MRI measures restriction of water molecule diffusion, with lower Apparent Diffusion Coefficient (ADC) values suggested as indicators of fibrosis.47,48 However, preliminary data suggest it may be less accurate than MT-MRI (AUC 0.747 vs 0.919 in direct comparison). 38 Given this, and the need for further large-scale validation similar to other emerging techniques, this framework prioritizes MT-MRI and SWE as the primary modalities for fibrosis quantification.47,48
The use of validated thresholds, once established, would allow clinicians to objectively stratify strictures into fibrosis-dominant, inflammation-dominant, or mixed phenotypes, directly informing therapeutic decisions (Supplemental Table S1). 40
Pillar 3: Emerging molecular biomarkers
Emerging molecular biomarkers remain investigational for B2 disease. 49 Despite investigation of various candidates (Supplemental Table S2), none has demonstrated sufficient specificity to differentiate fibrotic from inflammatory strictures. 49 The PROFILE trial’s 17-gene signature showed no clinical utility. 24 Recent data on collagen formation markers (PRO-C3/C5/C6) show promise (AUC 0.91 for identifying stenosing disease), but require validation. 50 These modalities remain emerging tools that require prospective validation.49,51
Limitations of endoscopic assessment and the role of endoscopic balloon dilation
While quantitative cross-sectional imaging forms the backbone of transmural stricture assessment, endoscopy and tissue acquisition retain several crucial, complementary roles that are integral to a comprehensive evaluation. Primarily, endoscopic evaluation is essential for visually assessing the disease and obtaining targeted biopsies to exclude malignancy, provide histologic confirmation for grading fibrosis and inflammation, and allow for molecular analysis.10,45,52 However, a critical limitation of this approach is the superficial nature of endoscopic biopsies, which often fail to capture the full transmural extent of fibrosis characteristic of B2 disease. Endoscopic balloon dilation (EBD), in particular, can be employed for carefully selected patients. 10 However, its significant limitations must be acknowledged early on. A systematic review and meta-analysis by Bettenworth et al. highlighted that symptom recurrence occurs in approximately half of patients, and up to two-thirds may require re-intervention (repeat dilation or surgery) during follow-up. 53 This underscores that EBD often serves as a temporizing measure rather than a definitive cure. 53 Therefore, a comprehensive assessment requires thoughtful integration of these endoscopic findings and limitations with quantitative imaging data to inform the evidence-based clinical algorithm detailed below.
The integrated clinical algorithm: From evidence to practice
The algorithm presented below should be interpreted as a conceptual and hypothesis-generating framework. Some of its components are grounded in established evidence and current guideline-based practice, whereas others—particularly the integration of quantitative imaging thresholds into treatment stratification for B2 disease—are author-proposed elements intended for prospective validation rather than immediate routine clinical implementation.
The algorithm (Figure 1) begins with Montreal classification to stratify patients into B1, B2, or B3 phenotypes. 30 B1 patients receive optimal medical therapy following treat-to-target strategies. B3 patients require surgical consultation, with management tailored based on complication severity: complex fistulas or large abscesses mandate resection, while small simple abscesses may allow conservative management as bridge-to-biologics.9,10
Proposed phenotype-driven treatment framework for ileocecal Crohn’s disease. This two-tiered framework separates validated standard care (Tier 1, green) from investigational approaches (Tier 2, red). Tier 1 uses the Montreal classification to direct B1 patients to optimal medical therapy and B3 patients to surgical consultation. Tier 2 applies quantitative imaging (MT-MRI or SWE) for preliminary stratification of B2 stricturing disease. All B2 patients undergo comprehensive Multidisciplinary Team (MDT) discussion integrating multimodal data to inform consideration of intensive medical therapy for inflammation-dominant strictures, early surgical resection for fibrosis-dominant strictures, or hierarchical assessment for mixed/gray-zone cases. Figure 1 is intended to illustrate the conceptual structure of the proposed framework and should not be interpreted as a validated clinical decision pathway for current routine use. Numeric thresholds are illustrative, derived from limited cohort data, and should not be used in routine clinical decision-making without prospective validation.
Patients with stricturing (B2) disease, defined by the presence of bowel wall thickening (>3 mm), luminal narrowing, and pre-stenotic dilation (⩾2.5 cm) as per the 2025 Society of Abdominal Radiology consensus recommendations, 25 enter the precision medicine pathway. This precision medicine pathway is centered around a Multidisciplinary Team (MDT) discussion that thoughtfully integrates multiple data sources. The primary objective tool for this discussion is quantitative imaging (MT-MRI or SWE) to assess stricture composition. The MDT then synthesizes these imaging results with crucial endoscopic findings (such as stricture passability and the presence of deep ulcers) and the patient’s clinical status to guide the next step. Within this proposed framework, quantitative imaging thresholds are presented as illustrative decision-support inputs for multidisciplinary discussion rather than validated standalone triggers for treatment assignment. For example, imaging findings suggestive of a predominantly fibrotic phenotype, including higher MT-MRI-derived values reported in selected cohorts, may support consideration of earlier surgical resection within MDT review, particularly when accompanied by a non-passable stricture or severe obstructive symptoms.10,35 Conversely, imaging findings suggestive of a predominantly inflammatory phenotype may support a trial of intensive medical therapy with close monitoring.10,39 When MT-MRI is inaccessible, SWE may be used as an adjunctive tool, with a value >2.73 m/s being suggestive of fibrosis, though this modality should not be used as a standalone criterion, and results must be interpreted with caution. 42
The most challenging clinical scenario occurs when quantitative imaging yields indeterminate results (e.g., MTR 28.7–31.25%) or conflicting findings between modalities, suggesting comparable inflammatory and fibrotic components.35,39 Rather than defaulting to empiric therapy, a structured four-step hierarchy guides MDT decision-making to integrate multiple data streams systematically.
First, anatomic suitability for endoscopic intervention should be assessed, as this typically overrides other considerations. First-line therapy with EBD may be considered for strictures meeting favorable criteria, including a length less than 5 cm, accessible location, non-angulated course, and the absence of complications such as deep ulceration or associated abscesses.33,54 This approach directly addresses the mechanical component while preserving future medical and surgical options, with optimal outcomes reported for lesions shorter than 4 cm. 55 Critically, as EBD only addresses the mechanical obstruction and not the underlying biological activity, successful dilation should be immediately followed by the initiation or optimization of intensive medical therapy (e.g., advanced biologics/small molecules) to maintain patency and reduce the high risk of recurrence.9,10
Second, if EBD is anatomically unsuitable, validated prognostic models should be used to stratify long-term surgical risk. The BACARDI model, incorporating prestenotic dilation, elevated CRP, and prior anti-TNF failure, provides evidence-based risk assessment. 56 Early surgical resection should be considered for patients presenting with multiple adverse factors, including prestenotic dilation exceeding 3 cm, persistently elevated inflammatory markers, and documented biologic failure.10,33 Conversely, patients with few risk factors may undergo time-limited intensive medical trials with predefined reassessment at 12–24 weeks. 10 Third, when prognostic risk remains intermediate, adjunctive CEUS can refine stratification by quantifying active inflammatory activity. 43 High CEUS enhancement (e.g., wash-in area under curve >38 dB) suggests substantial reversible inflammatory component, supporting intensive medical therapy trial. 43 Low CEUS enhancement in the setting of indeterminate structural findings suggests predominantly chronic changes with limited inflammatory activity, favoring surgical consideration. 43
Finally, objective data must be integrated with patient-specific factors through shared decision-making. Relevant considerations include age and comorbidities, prior surgical history, career stage and life planning, tolerance for medication burden versus surgical risk, and personal values regarding treatment trade-offs.9,10
When earlier steps yield conflicting signals, patient preference becomes decisive. This hierarchical framework is designed to structure rather than replace clinical judgment, moving from anatomic constraints through validated risk assessment and biological characterization to individualized patient preferences.
For patients who proceed with a trial of medical therapy, a clear response assessment strategy is mandatory. 10 Repeat quantitative imaging could be performed at or around 24 weeks, a timeframe consistent with the new international guidelines for MRE response assessment. 57 A significant radiologic improvement, now specifically defined by this consensus as a >25% improvement in MaRIA score or a decrease of ⩾1 point in the simplified MaRIA score, would justify continuing medical therapy. 57 Conversely, a lack of objective improvement or any evidence of worsening would warrant a prompt surgical referral to avoid futile therapy and disease progression, a common outcome in stricturing disease.10,26,33
Implementation considerations and health economics
The principles underlying this framework, specifically phenotypic stratification, objective assessment of stricture composition, and multidisciplinary decision-making, align with current guideline recommendations and can be adopted incrementally.9,10 Tier 1 phenotyping represents current standard of care and can be implemented immediately at all centers. 25 In contrast, the integration of advanced quantitative imaging into treatment selection should currently be viewed as a proposed enhancement to multidisciplinary assessment rather than an established implementation standard. Advanced imaging modalities can be incorporated as additional data for MDT discussion, with quantitative results interpreted as one of multiple inputs into clinical decision-making. This tiered approach allows centers to enhance their current practice with objective assessment tools while prospective validation is underway.
Successful implementation requires overcoming significant practical barriers related to technology, cost, and specialized expertise.33,40 Key strategies include developing consensus-based imaging protocols to standardize techniques like SWE, establishing formal training and certification pathways to ensure quality,40,46 and leveraging teleradiology to bridge geographic gaps in expertise. 58 This need for standardization is further supported by recent systematic evidence demonstrating marked variability in radiologic stricture definitions and performance parameters across studies. 59 Clinically, structured support through dedicated multidisciplinary case conferences, aided by decision support tools, would be valuable to integrate these new data streams into routine practice.60,61 In resource-limited settings where MT-MRI or SWE are unavailable, conventional CTE or MRE features suggestive of fibrosis can serve as a pragmatic alternative to guide MDT discussions. 33
A tiered implementation strategy is proposed to address resource variation across centers, ensuring that the framework is adaptable to different healthcare settings. This approach can be envisioned in three levels:
Tier 1: Foundational Phenotyping. At the most basic level, all centers can immediately implement rigorous Montreal classification. 30 This initial stratification alone provides significant clinical value by appropriately directing straightforward B1 (inflammatory) and B3 (penetrating) cases, representing a marked improvement over a purely empirical approach.62,63
Tier 2: Intermediate Objective Assessment. Community or regional hospitals can integrate more accessible advanced ultrasound modalities as a powerful intermediate step. A multimodality approach, potentially combining CEUS to assess inflammatory activity (with low enhancement suggesting chronicity) and SWE to estimate tissue stiffness (with values >2.73 m/s indicative of fibrosis),39,64,65 could offer valuable objective data. While acknowledging the ongoing need for SWE standardization, these techniques can serve as objective tie-breakers in clinically ambiguous B2 cases.40,66 For instance, a very high stiffness value would strongly favor a surgical referral by suggesting dominant fibrosis, even without MT-MRI. 41 Conversely, a low stiffness value would support a trial of intensive medical therapy. 27
Tier 3: Comprehensive Precision Medicine. Major academic and research centers can implement the full precision toolkit, including advanced modalities like MT-MRI and emerging molecular diagnostics.
Table 3 summarizes this tiered implementation approach, allowing centers to adopt the framework incrementally based on available resources and expertise.
Paradigm shift and tiered implementation for the precision medicine framework.
This table contrasts current guideline-based management with the proposed precision enhancements across five key clinical domains. Tier 1 elements align with existing international guidelines and can be implemented immediately.9,10 Tiers 2 and 3 represent proposed enhancements that require prospective validation. The illustrative PRECISION-CD trial concept (Supplemental Appendix 2) outlines one possible pathway for prospective validation.
ACG, American college of gastroenterology; B1, inflammatory phenotype; B2, stricturing phenotype; B3, penetrating phenotype; CEUS, contrast-enhanced ultrasound; CRP, C-reactive protein; CTE, computed tomography enterography; ECCO, European Crohn’s and Colitis Organisation; MDT, multidisciplinary team; MRE, magnetic resonance enterography; MT-MRI, magnetization transfer magnetic resonance imaging; PRO-C3, N-terminal propeptide of type III collagen; SWE, shear wave elastography.
Preliminary economic modeling suggests the framework is unlikely to impose additional cost burden. Base-case projections indicate potential net savings of approximately $19,000–27,000 per patient over 5 years, while conservative worst-case analysis demonstrated cost-neutrality at 3–4 years (Supplemental Tables S3 and S4).6,24,67,68 These projections rely on proxy data from non-B2 populations and require formal cost-effectiveness analysis in stricturing disease cohorts. Detailed assumptions and sensitivity analyses are provided in Supplemental Tables S3 and S4.
To conclude this section, a summary comparison highlighting how this new framework represents a paradigm shift from current clinical guidelines is provided in Table 3.9,10
Limitations and the imperative for prospective validation
To provide full transparency on the current evidence level for each component of this framework, we have provided a detailed evidence quality assessment (Supplemental Table S5). In brief, the framework comprises validated components that are already aligned with current practice (Montreal B1/B3 stratification), moderately validated tools requiring further confirmation (MT-MRI for fibrosis in Asian cohorts), and elements requiring prospective evaluation (SWE per 2024 consensus, and the complete integrated algorithm). Because literature selection was purposive rather than exhaustive, the review remains vulnerable to selection bias in the identification and weighting of key studies, particularly in rapidly evolving areas such as advanced imaging. Potential harms, such as the risk of delaying therapy due to imaging logistics or misinterpretation of indeterminate cases, must be mitigated through efficient clinical pathways and robust MDT processes. This last concern is especially important because attempts to distinguish inflammation-dominant from fibrosis-dominant strictures currently rely on tools that remain incompletely standardized. Quantitative thresholds, particularly for MT-MRI and especially SWE, are derived from relatively limited cohorts and may vary across centers, protocols, operators, and patient populations. Consequently, a predominantly fibrotic stricture could be misclassified as inflammatory, leading to prolonged ineffective medical therapy and delayed definitive intervention, whereas a predominantly inflammatory stricture could be overcalled as fibrotic, potentially prompting unnecessary or earlier-than-needed surgical referral. This risk is particularly relevant in mixed or indeterminate strictures, where biological overlap and technical variability may further reduce classification accuracy. These uncertainties reinforce that the proposed framework should be interpreted as a conceptual and hypothesis-generating model rather than a routine clinical decision tool at the current stage. Borderline or indeterminate cases should therefore be interpreted within multidisciplinary discussion, and prospective outcome tracking will be essential for future refinement and validation. A possible next step would be prospective evaluation of this framework against standard care in patients with B2 disease, with long-term outcomes such as surgery-free survival serving as key endpoints; an illustrative trial concept is outlined in Supplemental Appendix 2.
Generalizability and global applicability
A significant consideration for the proposed framework is its generalizability, as the foundational evidence, from the LIR!C trial to the Danish cohort, is derived predominantly from North American and European populations.6,7 The applicability of this model to other ethnic groups and healthcare systems, particularly in regions like the Asia-Pacific where the incidence of CD is rapidly rising, remains an unvalidated but critical assumption. 44
In resource-constrained settings, local adaptation may be necessary; however, any modification of quantitative thresholds should be prospectively validated within the target healthcare context. For instance, where access to biologic therapy is severely limited, and surgery represents a lower-cost alternative, earlier surgical consideration at a lower MTR threshold (e.g., >28.7% rather than >31.25%) could be explored as a context-specific strategy, pending prospective validation.35,39
Therefore, a critical future direction is the initiation of large-scale, prospective studies to validate this precision medicine pathway in multi-ethnic cohorts. Establishing regional research consortia, particularly in Asia, Latin America, and Africa, to conduct such validation studies is essential.
Successful models from other medical fields, such as teleradiology partnerships that facilitate expert interpretation of complex imaging across different regions, could be adapted to accelerate the adoption and standardized use of quantitative techniques like SWE and MT-MRI. This endeavor goes beyond simply improving access to technology; it is about ensuring that the biological principles and quantitative thresholds at the heart of this framework are universally applicable. True global health equity in CD management will only be achieved when personalized treatment strategies are validated for, and tailored to, the specific populations they are intended to serve.
Furthermore, regional research consortia, in addition to conducting clinical validation, might be tasked with developing locally relevant economic analyses. These models must incorporate local costs for diagnostics, surgery, and available medical therapies to determine if the precision medicine pathway offers economic value within that specific healthcare context. This parallel economic validation is proposed to provide policymakers and local healthcare leaders with the evidence needed to justify investments in the infrastructure and training required to adopt this framework.
Conclusion
For clinicians managing localized ileocecal CD, treatment decisions must integrate disease phenotype with an assessment of stricture biology. While established pathways exist for inflammatory (B1) and penetrating (B3) phenotypes, stricturing disease (B2) remains the most challenging scenario, where distinguishing inflammation-dominant from fibrosis-dominant strictures is critical yet poorly standardized.
This review proposes a two-tiered precision medicine framework to address this gap. Tier 1 utilizes the validated Montreal classification to direct inflammatory and penetrating phenotypes to established pathways. Tier 2 introduces quantitative imaging (MT-MRI or SWE) to objectively characterize stricture composition in B2 disease, guiding treatment toward early surgery for fibrosis-dominant strictures and intensive medical therapy for inflammation-dominant disease.
Until prospective validation is available, this framework should be viewed as a structured research roadmap rather than a clinical decision algorithm. The illustrative PRECISION-CD trial concept represents one possible pathway for generating the evidence needed for future clinical adoption.
Supplemental Material
sj-docx-1-tag-10.1177_17562848261446459 – Supplemental material for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework
Supplemental material, sj-docx-1-tag-10.1177_17562848261446459 for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework by Chuong Dinh Nguyen, Luan Minh Dang, Thong Duy Vo, Hoang Huu Bui, Partha Pal, Taku Kobayashi and Byong Duk Ye in Therapeutic Advances in Gastroenterology
Supplemental Material
sj-docx-2-tag-10.1177_17562848261446459 – Supplemental material for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework
Supplemental material, sj-docx-2-tag-10.1177_17562848261446459 for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework by Chuong Dinh Nguyen, Luan Minh Dang, Thong Duy Vo, Hoang Huu Bui, Partha Pal, Taku Kobayashi and Byong Duk Ye in Therapeutic Advances in Gastroenterology
Supplemental Material
sj-docx-3-tag-10.1177_17562848261446459 – Supplemental material for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework
Supplemental material, sj-docx-3-tag-10.1177_17562848261446459 for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework by Chuong Dinh Nguyen, Luan Minh Dang, Thong Duy Vo, Hoang Huu Bui, Partha Pal, Taku Kobayashi and Byong Duk Ye in Therapeutic Advances in Gastroenterology
Supplemental Material
sj-docx-4-tag-10.1177_17562848261446459 – Supplemental material for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework
Supplemental material, sj-docx-4-tag-10.1177_17562848261446459 for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework by Chuong Dinh Nguyen, Luan Minh Dang, Thong Duy Vo, Hoang Huu Bui, Partha Pal, Taku Kobayashi and Byong Duk Ye in Therapeutic Advances in Gastroenterology
Supplemental Material
sj-docx-5-tag-10.1177_17562848261446459 – Supplemental material for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework
Supplemental material, sj-docx-5-tag-10.1177_17562848261446459 for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework by Chuong Dinh Nguyen, Luan Minh Dang, Thong Duy Vo, Hoang Huu Bui, Partha Pal, Taku Kobayashi and Byong Duk Ye in Therapeutic Advances in Gastroenterology
Supplemental Material
sj-docx-6-tag-10.1177_17562848261446459 – Supplemental material for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework
Supplemental material, sj-docx-6-tag-10.1177_17562848261446459 for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework by Chuong Dinh Nguyen, Luan Minh Dang, Thong Duy Vo, Hoang Huu Bui, Partha Pal, Taku Kobayashi and Byong Duk Ye in Therapeutic Advances in Gastroenterology
Supplemental Material
sj-docx-7-tag-10.1177_17562848261446459 – Supplemental material for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework
Supplemental material, sj-docx-7-tag-10.1177_17562848261446459 for Phenotype-driven management in ileocecal Crohn’s disease: a narrative review with a proposed research framework by Chuong Dinh Nguyen, Luan Minh Dang, Thong Duy Vo, Hoang Huu Bui, Partha Pal, Taku Kobayashi and Byong Duk Ye in Therapeutic Advances in Gastroenterology
Footnotes
Acknowledgements
None.
Declarations
Guarantor of the article
Luan Minh Dang.
AI use disclosure
The authors acknowledge the use of Gemini 3.1 Pro (Google) for English language editing assistance and support in preparing the figure layout and visual abstract. All scientific content, interpretations, and conclusions are solely those of the authors, who take full responsibility for the final manuscript.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
