With regards to the endoscopic approach, endoscopic balloon dilatation (EBD) is considered the preferred technique for selected CD strictures, as it has proven a high rate of short-term technical and clinical efficacy, with substantial long-term efficacy and acceptable rates of complication
[63][64]. EBD is a minimal-invasive procedure that consists in placing a radial expanding balloon dilator (available in an array of designs, lengths, and calibers) in the stenotic tract and inflating the balloon as needed
[65]. Experts have judged the following items relevant in practicing EBD: 18 mm as the maximal luminal diameter after dilatation in one or several sessions, a balloon inflation time of at least 1 min, and 5 cm as the maximum stricture length that should be dilatated
[6]. Furthermore, such a procedure has been defined successful when it is possible to pass an adult ileocolonoscope through a previously non-traversable stricture with a reasonable amount of pressure applied.
When medical or endoscopic therapy fails or is contraindicated, surgery should be considered. According to an expert consensus, in stricturing disease both surgical resection and strictureplasty are valid options, with similar safety, efficacy, and long-term recurrence rates
[66]. The preferred treatment of multiple fibrotic strictures of the small intestine, when technically feasible, are stricureplasties. According to the length and site of the stricture, multiple techniques for strictureplasty have been proposed, including Heineke-Mikulicz, Michelassi, and the Finney technique. In order to reduce fibrosis recurrence, multiple surgical strategies have been explored, including special anastomotic configurations, such as the antimesenteric functional end-to-end handsewn anastomosis, also known as Kono-S anastomosis
[67], mesentery and lymph node excision
[68], and the laparoscopic approach
[69].
3.3. Promising Anti-Fibrotic Therapy in CD
Increasing knowledge of the molecular mechanisms underlying intestinal fibrosis has enabled the identification of anti-fibrotic therapeutic targets. At present, although there is no therapy capable of treating or reversing intestinal fibrosis in CD, several pre- clinical studies have been conducted in vivo, ex vivo, and in vitro, with encouraging results. Herein, the most promising anti-fibrotic therapeutic targets known to date and the relevant target-specific molecules under investigation are summarized in Table 1.
3.3.1. Targeting TGF-β Pathways
The most promising target for anti-fibrotic therapy is TGF-β, the principal molecular mediator of fibrogenesis, and its signaling pathways.
-
Several studies on fibrosis of other tissues have shown that TGF-β1 production was strongly stimulated by the local activation of angiotensin II
[70][71][72], the main effector of the renin-angiotensin system, whose activity is increased in the colonic mucosa of CD patients
[73]. For this reason, it was assumed that angiotensin conversing enzyme (ACE) inhibitors and sartans (angiotensin II receptor antagonists), which typically act as anti-hypertensives, could also play a role in the process of intestinal fibrogenesis. The first ACE-inhibitor investigated was captopril, which showed to be effective in preventing colonic fibrosis in 2,4,6-trinitrobenzene sulfonic acid (TNBS)-induced colitis in rats. Its anti-fibrotic action has been assumed to derive from blocking TGF-β1 overexpression and/or from a direct down-regulation of TGF-β1 transcripts
[74]. Moreover, transanal administration of enalaprilat has been shown to be effective in preventing colonic fibrosis in a dextran sulfate sodium (DSS)-induced colitis model
[75]. More recently, losartan, an antagonist of the angiotensin II receptor, was investigated and exhibited a pleiotropic effect, reducing TGF-β1 concentration and significantly improving the macro- and microscopic scores of fibrosis in the colonic wall of rats
[76];
-
Based on the known antagonistic relationship between the TGF-β/Smad pathway and the peroxisome proliferator-activated receptor (PPAR)γ, a member of ligand-activated transcription factors of nuclear hormone receptor superfamily
[77][78], the effect of a novel 5-ASA analog (named GED-0507-34 Levo), able to activate PPARγ, has been investigated. GED-0507-34 Levo showed improvement of intestinal fibrosis in DSS-induced chronic colitis in mice, reducing the activation of myofibroblasts and the expression of the main pro-fibrotic molecules including TGF-β, Smad3, IL-13 and connective tissue growth factor (CTGF)
[79]. Similarly, it has been shown that other PPARγ agonists, usually employed in the treatment of diabetes, such as troglitazone and rosiglitazone, may be useful in counteracting the fibrogenic process by suppressing TGF-β1-induced synthesis of collagen, fibronectin, and α-smooth muscle actin in human primary intestinal myofibroblasts
[80];
-
Another target signaling pathway induced by TGF-β1 but also by matrix stiffness is that of Rho/Rho chinase (ROCK)
[81]. The first ROCK inhibitors studied were CCG-1423, CCG-100602, and CCG-203971, which, by inhibiting RhoA signaling in myofibroblasts, induced a significant anti-fibrotic activity
[82][83]. These molecules, however, showed an unacceptable toxicity profile, especially with regard to cardiovascular side effects
[84]. For this reason, the effect of a locally acting ROCK inhibitor (AMA0825) was investigated. This molecule prevented and reversed intestinal fibrosis in vitro and ex vivo by diminishing TGF-β1-induced activation of myocardine-related transcription factor and p38 mitogen-activated protein kinase (MAPK) and increasing autophagy in fibroblasts, with a good tolerability profile
[85]. Combining AMA0825 with anti-inflammatory agents (such as anti-TNF-α) in vivo ameliorated inflammation but also prevented accumulation of fibrotic tissue, underscoring the importance of combination therapy;
-
Other compounds have been shown to downregulate the TGF-β signaling. These include cilengitide, which is an Arg-Gly-Asp (RGD)-containing αVβ3 integrin inhibitor, that is able to decrease TGF-β1 activation and development of fibrosis in chronic TNBS-induced colitis
[86]. More recently, anti-fibrotic intestinal efficacy has been proposed for two molecules approved for the treatment of idiopathic pulmonary fibrosis, namely pirfenidone and nintedanib
[87][88]. In particular, pirfenidone, an orally delivered pyridine derivative that suppresses TGF-β and TNF-α signals, inhibited, both in vivo and in vitro, intestinal fibroblast proliferation and motility and reduced collagen production through different TGF-β1 signaling pathways, including those of suppressor of mothers against decapentaplegic (Smad), phosphatidylinositol-3-kinase (PI3K)/AKT, MAPK, and mechanistic target of rapamycin (mTOR)
[89][90][91][92]. Therefore, this molecule is of great interest and has important therapeutic potential, but needs further studies to better clarify its mechanism of action, efficacy, and safety
[93]. No studies are yet available on the usefulness in intestinal fibrosis of nintedanib, a small oral molecule inhibitor of tyrosine kinase receptors, such as platelet-derived growth factor (PDGF), fibroblast growth factor (FGF), and vascular endothelial growth factor (VEGF) receptors. Finally, an anti-fibrotic action of maggot extract was described by downregulating the TGF-β1/Smad pathway via upregulation of nuclear factor erythroid 2-related factor 2 (Nrf2) expression
[94].
3.3.2. Targeting TIMP/MMP Balance
Intestinal fibrosis in CD is mainly due to the imbalance of deposition and degradation of ECM, regulated also by MMPs and TIMPs
[11]. Thalidomide, a molecule with anti-inflammatory activity and emerging as an alternative treatment for refractory CD
[95], has been shown to inhibit in vivo intestinal fibrosis by regulating TIMP/MMP protein balance and degradation of ECM
[96].
3.3.3. Targeting VEGF
The deposition of collagen causes chronic hypoxia, which in turn stimulates neo-angiogenesis through the upregulation of VEGF, thus favoring the deposition of further fibrotic tissue in a vicious circle
[29]. VEGF has been supposed to be a therapeutic target of fibrosis, and its blockade through a monoclonal antibody (bevacizumab) has been investigated as a possible anti-fibrogenic strategy. However, this molecule showed a significant increase in fibrosis-related inflammatory cytokines in vitro
[97] and, due to the possible side effects, could even worsen CD
[98]. For this reason, targeting neo-angiogenesis does not currently seem to be useful, but rather harmful. Studies with anti-VEGF agents on models of intestinal fibrosis are still lacking.
3.3.4. Targeting FAP
The FAP protein, discussed previously, could be a unique therapeutic target as it is a marker of active fibroblasts
[99]. Thus, treatment directed against FAP would have high specificity and minimal side effects. Ex vivo treatment of stenotic tissues with anti-FAP antibody induced a dose-dependent decrease in collagen, particularly type I collagen, and TIMP-1 production, without altering MMP-3 and MMP-12 secretion
[28]. Another FAP inhibitor to be mentioned is talabostat mesilate (PT100), which in a murine model of pulmonary fibrosis showed an anti-fibro-proliferative effect
[100], but has not yet been studied for intestinal fibrosis.
3.3.5. Targeting EMT
Another therapeutic target of recent interest is EMT, the complex process in which epithelial cells lose their phenotypic and functional characteristics and develop mesenchymal features
[15]. In addition, it would appear that EMT may associate with intestinal fibrosis not only through direct production of myofibroblasts, but also through the release of crucial signals for myofibroblast differentiation
[101][102]. Recent data from animal models of renal, hepatic, and cardiac fibrosis have demonstrated the anti-fibrotic effect of recombinant human bone morphogenic protein-7 (rhBMP-7)
[103][104][105]. BMP-7 is a member of the TGF-β superfamily with the ability to counteract the pro-fibrotic action of TGF-β1. In vitro and in vivo studies have shown the effect of rhBMP-7 in inhibiting TGF-β1 induced EMT associated with intestinal fibrosis
[106]. In addition, a recent study showed that miRNA200b-containing microvesicles inhibited colonic fibrosis, thus suppressing the development of EMT by targeting zinc finger E-box binding homeobox (ZEB)1 and ZEB2
[107].
3.3.6. Targeting the Endogenous Cannabinoid System
The cannabinoid system comprises specific G-protein-coupled receptors (CB1 and CB2), a variety of exogenous (marijuana-derived cannabinoids) and endogenous ligands, and a machinery dedicated to endocannabinoid synthesis and degradation
[108]. One of the main endogenous CB1 and CB2 agonists is anandamide (AEA)
[109]. Given the evidence from experimental studies that the endocannabinoid system is involved in intestinal diseases and played a role in antagonizing fibrosis in chronic liver disease
[110][111], the in vitro effect of the AEA analogue methanandamide (MAEA) on CD strictured myofibroblasts was investigated
[112]. The CB2 agonist showed to reduce collagen production by strictured CD myofibroblasts and increase their migration ability. No further data are available about the anti-fibrotic role of cannabinoid receptor agonists.
3.3.7. Targeting IL-17
As mentioned above, the IL-17A is overexpressed in CD strictures and determines myofibroblasts production of collagen and TIMP-1 and reduction of their migratory ability
[25]. However, IL-17 contribution to IBD is still controversial
[113]. A recent study has demonstrated that treatment with the anti-IL17 antibody, in TNBS-induced intestinal fibrosis mice, not only significantly decreased profibrogenic cytokines (IL-1β, TGF-β1, and TNF-α) and intestinal inflammation, but also reduced fibrogenesis-related TIMP-1 and MMP-2 gene expression
[114]. Another recent study demonstrated a similar function of IL-17 in intestinal fibrosis, showing that IL-17-driven fibrosis is inhibited by Itch-mediated ubiquitination of hydrogen peroxide-inducible clone 5 (HIC-5)
[115]. However, previous clinical trials reported a contradictory effect of anti-IL-17 treatment in CD patients, as blocking IL-17 with specific antibodies (secukinumab and brodalumab) failed to relieve symptoms and even increased disease activity in active CD patients
[116][117]. The reason for this conflict could lie in the unclear role of IL-17 in the intestine immune homeostasis
[24]. Thus, more investigations on the effect of anti-IL-17 treatment in intestinal fibrosis and on the safety of this therapy are needed.
3.3.8. Targeting IL-36
IL-36 is a member of the IL-1 superfamily and consists of three agonists and one receptor antagonist (IL-36Ra)
[21]. Endogenous agonists act as proinflammatory cytokines and the IL-36 signaling also promotes secretion of pro-fibrotic mediators. Thus, a potential role of IL-36R inhibition as a therapeutic strategy to treat pro-fibrotic disorders has been proposed
[118]. Antibodies against IL-36R were investigated in DSS or TNBS-induced mice colitis and showed to significantly reduce established fibrosis
[119]. Further studies are needed to ascertain the therapeutic potential of IL-36R signaling modulation in CD patients. A phase 2 trial is currently under way to evaluate the spesolimab (an anti-IL-36 receptor antibody) efficacy in patients with moderate-to-severe ulcerative colitis (NCT03482635).
3.3.9. Targeting TL1A
Accumulating evidence demonstrated the importance of TL1A in the pathogenesis of IBD and suggested a potential therapeutic role of TL1A blocking
[30][31]. More recently, anti-TL1A antibody injection showed to ameliorate intestinal fibrosis by inhibiting the activation of intestinal fibroblasts and reducing collagen deposition in the T cell transfer model of chronic colitis in mice
[31]. This effect may be related to the inhibition of TGF-1/Smad3 signaling pathway. A Phase 2a, multicenter, single-arm, open-label study demonstrated an acceptable safety profile for the anti-TL1A antibody (PF-06480605), which was effective in inducing endoscopic improvement in adults with moderate-to-severe ulcerative colitis
[120]. A phase 2b trial with this drug is in progress (NCT04090411).
3.3.10. Targeting Both TNF-α and IL-17
ABT-122 is a novel bispecific dual variable domain immunoglobulin targeting human TNF-α and IL-17. It has been demonstrated to be safe and effective in rheumatoid arthritis and psoriatic arthritis
[121][122]. The use of this molecule in immune-mediated intestinal diseases has recently been hypothesized, but no studies have yet been performed
[123].
3.3.11. Targeting AXL Pathway
AXL is a receptor tyrosine kinase that has been implicated in fibrogenic pathways involving myofibroblast activation
[124]. A recent study demonstrated a role of AXL pathway in models of intestinal fibrosis and suggested that the inhibition of AXL signaling through small molecule inhibitor (BGB324) could represent a novel target to antifibrotic therapy for intestinal fibrosis, inhibiting both matrix-stiffness and TGF-β1-induced fibrogenesis in human colonic myofibroblast
[125]. In addition, AXL inhibition sensitized myofibroblasts to undergo apoptosis.
3.3.12. Targeting NETs
The potential role of NETs in intestinal fibrosis has already been mentioned, although data in the literature are still scarce and sometimes conflicting. In some studies, CD inflamed ileum has shown high expression of NETs
[126][127], whereas in others no significant amount of NETs has been shown in CD when compared to ulcerative colitis
[128][129]. The key process promoting NET formation is H3-citrullination-mediated by peptidylarginine deiminase 4 (PAD4), and studies in mice on pulmonary fibrosis have shown a reduction in fibrosis by suppression of PAD4 and consequently of NETs
[130]. It has been suggested that PAD4/NETs inhibition may have a therapeutic role in CD as well
[34], however no studies have yet been performed.
3.3.13. Targeting miRNAs
miRNAs are increasingly studied as potential targets of anti-fibrotic therapies but no drugs targeting miRNAs are currently available in clinical practice. As said before, there is a significant down-regulation of the miRNA-29 family in the mucosa of CD strictured gut and it has been observed that the TGF-β1-induced collagen expression is reversed by exogenous overexpression of miRNA29b
[36]. In addition, the administration of miRNA200 has shown to partially protect intestinal epithelial cells from fibrogenesis in vitro, through the repression of ZEB1 and ZEB2 and the supposed inhibition of EMT
[38]. In the near future, miRNA modulation may provide interesting new therapeutic options.
3.3.14. Targeting Matrix Stiffness
Modifications of physical environment can affect myofibroblast behavior and survival
[41]. In vitro experiments showed that upon culture in a fibrotic environment, normal myofibroblasts increased the expression of MMPs, to counteract the mechanical force exerted by the matrix, by expressing increased levels of the collagen crosslinking enzyme lysyl oxidase (LOX), and inducing more ECM contraction
[131]. LOX inhibition completely restored MMP3-activity in CD stenotic myofibroblasts and prevented ECM contraction, allowing to consider LOX a potential anti-fibrotic agent.
3.3.15. Targeting Intestinal Microbiota
Given the increasing emphasis on the pro-fibrotic impact of the gut microbiota, many in vitro and in vivo studies have been carried out to assess the effect of probiotics and prebiotics on intestinal fibrosis
[132][133][134][135]. Among these, the most recent suggested that the soluble fraction of Vivomixx
® formulation was able to inhibit collagen-I and α-SMA expression in human colonic fibroblast by interfering TGF-β1/Smad2-3 signaling
[136]. All the data available are still preliminary and need to be confirmed and expanded.
Table 1. Therapeutic targets studied for intestinal fibrosis in Crohn’s disease.
TARGET |
AGENT |
MECHANISM |
MODEL |
TGF-β pathways |
Captopril |
↓ TGF-β1 expression and/or TGF-β1 transcript |
TNBS-colitis |
Transanal enalaprilat |
↓ TGF-β signaling pathway |
DSS-colitis |
Losartan |
↓ TGF-β1 expression |
TNBS-colitis |
GED-0507-34 Levo |
PPAR-γ activation |
DSS-colitis |
Troglitazone, Rosiglitazone |
PPAR-γ activation |
HIFs |
CCG-1423, CCG-100602, CCG-203971 |
ROCK inhibition |
CCD18-co HIFs |
AMA0825 |
ROCK inhibition |
DSS- and T-cell transfer-colitis, HIFs |
Cilengitide |
αVβ3 integrin inhibition |
TNBS-colitis |
Pirfenidone |
Smad, PI3K/AKT, MAPK, and mTOR signaling pathways inhibition |
HIFs, DSS-colitis, RIF |
Maggot extract |
↑ Nrf2 expression |
DSS-colitis |
TIMP/MMP balance |
Thalidomide |
Altered TIMP/MMPs balance and ECM degradation |
TNBS-colitis |
VEGF |
Bevacizumab |
↓ collagen deposition |
n.a. |
FAP |
Anti-FAP Ab |
FAP inhibition |
HIFs |
EMT |
rhBMP-7 |
EMT inhibition |
TNBS-colitis |
miRNA200b-containing microvescicles |
EMT inhibition |
TNBS-colitis, IEC-6 |
Endogenous cannabinoid system |
MAEA |
↓ collagen production and ↑ myofibroblasts migration |
Human organ culture biopsies, LPMCs, and HIFs |
IL-17 |
Anti-IL17 Ab |
↓ profibrogenic cytokines and MMP/TIMPs balance alteration |
TNBS-colitis |
IL-36 |
Anti-IL36R Ab |
↓ collagen production, MMPs, IL6 signaling, and EMT |
DSS- and TNBS-colitis |
TL1A |
Anti-TL1A Ab |
TGF-1/Smad3 signaling pathway inhibition |
T-cell transfer-colitis |
TNF-αand IL-17 |
ABT-122 |
n.a. |
n.a. |
AXL pathway |
BGB324 |
↓ matrix stiffness and TGF-β1-induced fibrogenesis |
CCD-18co, TNBS-colitis |
NETs |
PAD4 inhibitors |
↓ NETs-derived fibrosis |
n.a. |
miRNA |
miRNA29 |
↓ TGF-β1-induced collagen expression |
Human fibroblasts cultures |
miRNA200 |
↓ ZEB1 and ZEB2, EMT inhibition |
Intestinal epithelial cells |
Matrix Stiffness |
β-aminopropionitrile |
↑ MMP3 activity and ↓ ECM contraction |
HIFs |
Gut microbiota |
Probiotics and prebiotics |
Modulation fibrotic pathways |
Mouse and cellular models |
Abbreviations: Ab, antibody; CCD-18Co, noncancerous colon fibroblast; DSS, dextran sulfate sodium; ECM, extracellular matrix; EMT, epithelial-mesenchymal transition; EndMT, endothelial mesenchymal transition; FAP, fibroblast activation protein; HIF, human intestinal fibroblast; IEC, intestinal epithelial cell; IL, interleukin; LPMC, lamina propria mononuclear cell; MAEA, methanandamide; MAPK, mitogen-activated protein kinase; mTOR, mechanistic target of rapamycin; miRNA, micro ribonucleic acid; MMP, matrix metalloproteinase; n.a., not available; NET, neutrophil extracellular trap; Nrf2, nuclear factor erythroid 2-related factor 2; PAD4, peptidylarginine deiminase 4; PI3K, phosphatidylinositol-3-Kinase; PPAR, peroxisome proliferator-activated receptor; rhBMP-7, recombinant human bone morphogenic protein-7; RIF, radiation-induced intestinal fibrosis; ROCK, Rho/Rho chinase; Smad, suppressor of mothers against decapentaplegic; TGF, transforming growth factor; TIMP, tissue inhibitor of metalloproteinase; TL1A, TNF-like cytokine 1A; TNBS, 2,4,6-trinitrobenzene sulfonic acid; TNF, tumor necrosis factor; VEGF, vascular endothelial growth factor; ZEB, zinc finger E-box binding homeobox; ↑, increase; ↓, decrease.
4. Major Challenge for Anti-Fibrotic Agents Development
There are numerous pitfalls in identifying anti-fibrotic drugs for CD. First, experimental fibrosis in cellular and animal models does not necessarily resemble human fibrosis
[137]. Cells may behave differently in vitro and in vivo and single cell studies often do not reproduce the complex in vivo cellular network. For this reason, 3D models are under development to reproduce the natural microenvironment as closely as possible to that in vivo
[138][139]. Moreover, the targeted molecules often represent a small component of the complex molecular maze underlying the fibrotic process. The target molecules may even have multiple functions on the intestinal tissue, with the risk of targeting processes implicated in physiological tissue remodeling, resulting in negative effects. In addition, to date there are neither fibrosis biomarkers, nor diagnostic tools that can be used to identify and quantify the overall fibrotic burden in CD patients, especially in the early stages, when anti-fibrotic therapy may be mostly effective
[140][141]. Finally, there is an urgent need of end points that can be used to assess the efficacy of anti-fibrotic agents in clinical trials. For this reason, several groups of renowned IBD experts have reached expert consensus on this matter
[6][142]. In particular, a core set of 13 end-points (i.e., complete clinical response, long-term efficacy, sustained clinical benefit, treatment failure, radiological remission, normal quality of life, clinical remission without steroids, therapeutic failure, deep remission, complete absence of occlusive symptoms, symptom-free survival, bowel damage progression, and no disability) were considered critical
[142]. The combination of improved clinical, endoscopic and/or radiological features seems appropriate to define a successful treatment
[6]. The need for intervention within 24-48 weeks from medical therapy has been proposed as the most accurate end-point to assess anti-fibrotic agents in pharmacological trials.
Despite the urgency for anti-fibrotic therapy and the numerous molecules identified as potential anti-fibrotics in CD, no phase III clinical trial is currently ongoing or recruiting (according to
ClinicalTrials.gov, as of 19 January 2022).