1. Fibrosis Pathogenesis
Crohn’s disease can manifest in a variety of ways.
Multiple overlapping factors contribute to the formation of fibrostenotic lesions in Crohn’s disease. In summary, underlying genetic risk factors (in the form of single nucleotide polymorphisms (SNPs) identified in GWAS studies of Crohn’s disease) together with environmental stimuli interact to contribute to the activation of the inflammatory cascade that leads to the accumulation of fibrosis over time. Immune responses may be triggered within the mucosa and deeper layers of the wall following a breach of the intestinal epithelial barrier, allowing pathogens to enter. This breach is often caused by dysfunctional autophagy and unresponsive Paneth cells, a cause widely known to SNP mutations in NOD2 and ATG16L1, which constitute as the primary cause in around 30% of patients, but this number can vary based on the population
[19,20,21][1][2][3]. In the presence of numerous cytokines, particularly interferon gamma, interleukin-12, TGF-beta and interleukin-6, an elevated proinflammatory cascade is activated, which includes CD4 T-cells that can differentiate into pro-inflammatory CD4 Th
2 and Th
17 cells that drive inflammatory and fibrogenic processes. The reasons behind these are complex and involve a number of potential SNP mutations in toll-like receptors (TLRs) present on the surface of mononuclear cells, but these interactions are very complex and have been addressed in other studies
[22,23,24][4][5][6]. The invasion of specific bacteria and recruitment of inflammatory cells into the mesenteric fat increases cytokine production
[25][7], leading to aberrant adipocyte behaviour and the release of leptin, adiponectin and chemerin, with progression to creeping fat, with adipose tissue expanding and enveloping the inflamed regions of the small intestine, sometimes referred to as fat wrapping
[26,27,28,29][8][9][10][11]. Increasing endoplasmic reticulum stress from the accumulation of unfolded or misfolded proteins inside the affected cells increases signalling pathway activation and cytokine release. The overaccumulation of unfolded proteins leads to the dissociation of immunoglobulin-heavy-chain-binding protein (BiP) that leads to the activation of a number of interlocking pathways, namely, ATF6, IRE1 and PERK pathways
[30][12]. These respond to the presence of unfolded proteins and lead to the degradation of the excess mRNA, inhibit the translation processes, enhance autophagy, and, in extreme cases, lead to cell apoptosis. In Crohn’s disease, these pathways can be deregulated due to the presence of SNPs in critical genes as well as activation of a protein called anterior gradient homologue 2 (AGR2), which can prevent the inhibition of translation and, when released by the cells into the environment, acts as a chemoattractant of monocytes, leading to enhanced inflammation
[31,32,33][13][14][15].
Macrophages develop into the M2 macrophage phenotype, which adds to the cytokines already released into the intestinal wall
[15][16]. These cytokines activate fibroblasts and myofibroblasts, which generate collagenous fibrotic tissue with an associated extra-cellular matrix (ECM), often together with the in-growth of granulation tissue capillaries as part of a wound healing response
[34,35][17][18]. Some epithelial cells around the site of breach of the epithelial barrier may transform into mesenchymal-like cells in a process of epithelial-to-mesenchymal transition
[36,37,38,39][19][20][21][22]. Furthermore, some endothelial cells from either the granulation tissue capillaries or pre-existing vasculature may also undergo a similar process of endothelial-to-mesenchymal transition
[40,41][23][24]. Variable combinations of these changes result in fibroblast activation with ECM release, and increasing fibrotic tissue formation contributes to fibrostenosing lesion formation in Crohn’s disease.
2. Current Treatment Strategies
Currently, no specific anti-fibrotic therapy exists for CD, so treatment is currently aimed at managing the complications of stricturing disease. The CONSTRICT (Crohn’S disease anti-fibrotic stricture therapies) expert consensus group
[9][25] has published suggestions for diagnostic criteria for small bowel CD-related strictures. Magnetic resonance (MR) and computer tomography (CT) enterography are suggested as the preferred diagnostic tools for stricture assessment. Overall, MR enterography is favoured due to its lack of radiation exposure. Small bowel ultrasound (US) is another emerging imaging technique that appears to be able to accurately identify small bowel CD-related strictures
[42][26]. Endoscopic evaluation requires adequate bowel preparation, which can be challenging to achieve in patients with CD, with evidence that those with active CD experience more abdominal pain during bowel preparation, which is a predictor of poor preparation
[43][27]. When good bowel cleansing is achieved, strictures may be visualised endoscopically, but assessment is limited by its superficial mucosal views, lack of transmural assessment and inability to traverse the strictured bowel
[13][28].
While techniques to identify the presence of a stricture are well established, assessing the composition (inflammatory versus fibrotic) is more challenging, and meta-analysis in this area is lacking
[13][28]. However, accuracy may improve when techniques that assess small bowel motility are routinely integrated into clinical practice
[44,45][29][30]. In surgical resections, most stricture specimens have both inflammatory and fibrotic components
[9][25].
If there is clinical suspicion of an inflammatory component of the stricture, anti-inflammatory therapy, typically with steroids acutely followed by anti-tumour necrosis factor (anti-TNF) medication, is indicated. A recent prospective and observational trial, the CREOLE study, evaluated the efficacy of adalimumab in 97 patients with symptomatic small bowel strictures. In this group, 64% achieved success at 24 weeks (did not require other anti-inflammatory drugs or endoscopic/surgical intervention), and at follow-up (median time 3.8 years), 45.7 ± 6.6% of those who had achieved success at 24 weeks remained in prolonged success at four years
[46][31]. There is growing interest within the clinical community in the ‘window of opportunity’ in CD
[47][32], which describes the concept of commencing aggressive treatment early in the disease course before bowel damage occurs, as well as the concept of treatment to target (such as mucosal healing) and tight monitoring of both drug levels and biochemical markers of inflammation, such as faecal calprotectin.
In addition to maximising the efficacy of currently established treatments, new medical therapies are being investigated, including mesenchymal stem cells
[48][33] and small molecules with antifibrotic properties
[49][34], although it will be some time before they reach clinical practice.
Endoscopic intervention generally involves either balloon dilatation, endoscopic stricturotomy or stenting. It is generally suitable to consider endoscopic therapy for short (<5 cm), straight strictures without penetrating complications or deep ulceration
[50,51][35][36]. Endoscopic balloon dilatation (EBD) is the most commonly used technique, with technical success rates (defined as the ability of the endoscope to pass through the stricture following dilatation) of over 80%
[52][37] and high rates of short-term symptomatic improvement
[51,52][36][37]. However, the risk of complications, including luminal perforation, is around 3%, and this risk increases with increasing stricture length
[50,52][35][37]. Symptomatic recurrence is also high, and there is likely to be a requirement for further dilatation or surgery
[50][35].
Endoscopic stricturotomy is a newer technique that is not yet widely practised, and is associated with a lower perforation rate and lower stricture recurrence rates compared to balloon dilatation, but with a higher bleeding risk
[53][38]. Endoscopic stenting with a self-expanding metal stent (SEMS) is another potential strategy for the management of CD-related strictures, but further studies are needed to assess the effectiveness of this against more well-established techniques. A randomised trial of 80 patients in 19 Spanish centres demonstrated a similar safety profile for SEMS versus EBD, but EBD was found to be more effective overall
[54][39]. Surgery can, in some circumstances, be an appropriate first-line management option for the management of CD-related strictures. Indications for a surgical approach over endoscopic management include fistulating disease, abscess formation or significant pre-stenotic dilatation proximal to the stricture. First-line management should also be considered if there is any concern regarding malignancy.
The main surgical options include resection and stricturoplasty. The choice of which technique to use depends on, amongst others, the site, length and number of strictures. The European Crohn’s and Colitis Organisation (ECCO) and the European Society of Proctology (ESCP) published consensus guidelines on surgery for CD in 2018. The guidelines recommend that, where technically feasible, stricturoplasty should be considered first-line surgical management, particularly when long segments of the small bowel are affected (to minimise the risk of short-bowel syndrome), in recurrent strictures at ileocolic anastomotic sites, and where there is no complicating factor such as an associated abscess
[55][40].
In localised stricturing ileo-caecal Crohn’s disease, first-line surgical management is with either a laparoscopic ileo-caecal resection or stricturoplasty, unless there are perforating complications, in which case resection is required
[55][40].
There remains a high risk of recurrence following surgical intervention, and thus the risk of further surgeries also remains; with repeated resections, eventually, there is a risk of short-bowel syndrome, with major implications for nutrition and quality of life.
In the BACARDI study
[56][41], a risk model was suggested to identify those most at risk of requiring surgery and those who may benefit from ongoing medical management or endoscopic intervention for stricturing CD.
However, none of these options provide definitive, recurrence-free treatment for Crohn’s fibrosis; thus, the need for therapies specifically targeting the pathways leading to fibrosis is urgently required. A recent paper by Lin et al.
[57][42] highlights the developed anti-fibrotic therapies that already exist for conditions affecting the lung, kidney, skin and liver, and identifies several potential therapies that may have a transferrable role in managing Crohn’s fibrosis. These include targets such as growth factor modulators, inflammation modulators, intracellular enzymes and kinases, extracellular matrix (ECM) modulators, renin-angiotensin system (RAS) modulators and 5-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase inhibitors. While these may present promising opportunities for further study, a greater understanding of the gut at a cellular level, particularly in relation to fibrosis, is essential to develop novel therapeutic options for CD specifically.
3. Single-Cell Sequencing
Conventional research approaches investigating Crohn’s disease rely on the superficial assessment of endoscopic mucosal biopsies or on the large amount of data generated by surgical resection specimens. However, their ability to decipher the complex cellular networks and environmental topography, which may help identify new therapeutic targets, is limited. Single-cell techniques have provided new tools for investigating cellular changes in tissues, including heterogeneity of cellular composition, the differential abundance of cell (sub-)populations, along with alterations in cell states between normal and diseased conditions. A number of single-cell sequencing technologies have been designed for the investigation of the genome, transcriptome, epigenome (chromatin accessibility, DNA methylation, histone modifications and chromosome structure) and proteome (surface marker expression). In addition, spatial transcriptomics allows the correlation of such data with the two-dimensional location of cells within tissue sections
[65,66,67,68][43][44][45][46].
Recent studies related to Crohn’s disease have focused on single-cell transcriptomics (scRNA-seq), gut microbiome and dysbiosis
[69][47]. It has been established that Crohn’s disease usually involves pre-existing genetic polymorphisms that interact with environmental triggers. A number of studies have explored the most common microbiota composition present in Crohn’s disease patients and found organisms such as
Escherichia,
Shigella or
Atlantibacter species
[70][48]; however, this often correlates with the patient’s original geographical disease location
[71][49]. Certain bacterial strains such as
Clostridium innocuum have been shown to be involved in deep tissue penetration, invading the mesenteric fat surrounding ileal tissue, which has been proposed to lead to pro-inflammatory and pro-adipogenic responses, contributing to the formation of “creeping fat” or “fat-wrapping”
[25][7].
The role of the immune system in Crohn’s disease biology, specifically that of pro-inflammatory CD4 T-cells, has been studied by scRNA-seq, with a recent focus on CD8 T-cells and populations of natural killer T (NKT) type II cells identified in Crohn’s disease
[72][50]. The presence of CD8 T-cells that express surface markers CD39
+ and PD-1
+ was associated with disease progression, with the exhaustion of CD39
+ PD-1
+ CD8 T-cells correlating with remission
[73[51][52],
74], although the mechanisms for this require further investigation
[75][53]. The use of scRNA-seq for studying mononuclear phagocyte populations in the lamina propria of the intestines has shown changes in macrophage and dendritic cell subtypes during inflammatory bowel disease inflammation
[76][54].
Furthermore, a recent study by Mukherjee et al.
[77][55] examining scRNA-seq in stricturing disease showed increased fibroblast heterogeneity, particularly in the mucosa and submucosa, in areas of stricture compared to non-strictured bowel. This suggests that upregulation of fibroblast-specific markers in areas of stricturing—for example, Cadherin-11, a profibrotic cell surface receptor expressed in these fibroblasts—may play a major role in the development of fibrosis.
4. Spatial Analysis
For fibrosing CD, the ability to spatially characterise the dynamic pathogenesis at a cellular level will be critical in facilitating targeted drug design. Single-cell sequencing technologies have the power to redefine disease mechanisms as seen in hepatic fibrosis; therefore, unbiased gene expression analysis may identify rare cell types, transforming the understanding of CD-related fibrosis. Specifically, novel highly multiplexed microscopy techniques that establish gene expression at cellular resolution across histological sections can reveal the spatial organisation of detailed cellular activity within tissues, thus enabling these cell types to be located within their functional groups. The collection of spatial data is crucial to precisely pinpoint the differences in the severity of the disease progression, depending on the location of its expression. This type of detailed spatial analysis is possible when the gut location is precisely recorded during tissue specimen collection.
Inherent anatomical variations, the effects of chronic disease and previous surgery make it challenging to accurately determine the origin of tissue samples, limiting data interpretation and clinical translation
[87][56]. This has been recognised within the global Human Cell Atlas (HCA) initiative
[88][57] and in the gut context with the development of the Human Gut Cell Atlas (HGCA) promoted by NIH-funded HuBMAP and Helmsley Trust HGCA programmes. A roadmap for the human developmental cell atlas
[88][57] and a proposal for a focused gut-specific Common Coordinate Framework (CCF) have been published
[89][58]. In brief, a common coordinate framework is being established to provide a mechanism for data integration and analysis that will allow appropriate cross-study analysis and comparison.