Inflammatory bowel disease: new therapies from antisense oligonucleotides

Abstract

Inflammatory bowel diseases (IBD) are chronic inflammatory conditions of the gastrointestinal tract encompassing two main clinical entities: Crohn’s disease (CD) and ulcerative colitis (UC). These disorders are characterized by various grades of tissue damage and development of local complications and extra-intestinal manifestations. The cause of IBD remains unknown but accumulating evidence indicates that both CD and UC arise in genetically predisposed individuals as a result of the action of multiple environmental factors, which ultimately trigger excessive and poorly controlled immune response against antigens of the luminal flora. Despite this realization, a full understanding of IBD pathogenesis is still out of reach and, consequently, treatment is far from optimal. However, in recent years, several pathways of intestinal damage have been delineated and the improved knowledge has contributed to the development of new therapies. Various approaches have been used to either inhibit the expression and/or function of inflammatory molecules or enhance counter-regulatory mechanisms. This review summarizes the available pre-clinical and clinical data for antisense oligonucleotides and oligonucleotide-based therapy to provide a comprehensive understanding of the rationale and mechanism of action of these compounds in IBD.

Key messages

• Preclinical studies and clinical trials show that antisense oligonucleotide (ASO)-based therapy could be of benefit in inflammatory bowel diseases.

• ASOs have an excellent safety profile.

• Technical issues emerged from clinical trials suggest that changes in drug formulation and/or route of administration could improve ASO efficacy.

Keywords:

• Antisense
• Smad7
• colitis
• ICAM-1
• NF-kB
• GATA3

Introduction

Inflammatory bowel diseases (IBD) are chronic, immunologically-mediated diseases of the gastrointestinal tract characterized by various grades of tissue damage and development of local complications (e.g. strictures, fistulae, cancer) and extra-intestinal manifestations [1,2]. Crohn’s disease (CD) and ulcerative colitis (UC) are the major IBD in humans and these disorders are clinically characterized by intermittent exacerbations and sometimes chronically active course [3,4]. The causes of CD and UC are unknown and therefore there is no curative therapy [2]. However, data emerging from epidemiological and experimental studies suggest that IBD are the result of interaction between genetic and environmental factors, which promotes an excessive immune response against components of the luminal flora [5–8]. In the last decades, the advent of sophisticated techniques of molecular and cellular biology and the possibility to sample biopsies from the patients for deeply evaluating the ongoing mucosal inflammation has largely advanced our understanding of the mechanisms that drive tissue damage in IBD. These progresses paved the way for the development of new therapeutic compounds, which can selectively target molecules supposed to play a major role in IBD pathogenesis. Blockers of inflammatory cytokines, such as TNF and IL-12/IL-23, and compounds interfering with the recruitment of inflammatory cells in the gut, such as natalizumab and vedolizumab, are classical examples of this new era of biologic drugs [2]. However, not all the patients respond to these drugs and the initial response to these treatments can wane over time. The use of some biologics can also enhance the risk of infections, neoplasias and new immune-mediated disorders [2,9,10]. Overall, these limitations have boosted new studies aimed at identifying further molecules, which can be targeted for therapeutic purposes, and developing compounds, which selectively modulate either effector or regulatory pathways in a safe way [2]. This review summarizes the available pre-clinical and clinical data for antisense oligonucleotides (ASOs) and oligonucleotide-based therapy to provide a comprehensive understanding of the rationale and mechanism of action of these compounds in IBD.

Oligonucleotides-based strategy

ASOs are short (13–25 nucleotides) single-stranded nucleotides that target selectively with high specificity RNA sequences [11]. ASOs have effective drug-like properties and can regulate the fate of RNA molecules through multiple post-binding mechanisms [12]. These include modulation of RNA splicing, disruption of structures involved in the regulation of RNA stability, inhibition of RNA translation into protein, inhibition of 3-polyadenylation and catalytic degradation of target RNA molecules by the recruitment of RNase H1 [11,13]. Because of such mechanisms, the use of ASOs promotes the reduction of a given target RNA. However, there are novel antisense mechanisms, which can up-regulate gene expression (i.e. targeting of both upstream open reading frames on mRNAs and endogenous antisense transcripts themselves, modulation of mRNA nonsense-mediated decay pathways) [14]. Because of the unfavourable physicochemical properties of naturally occurring nucleic acids, the use of ASOs for therapeutic purposes is strictly dependent on strategies, which improve the absorption, distribution, metabolism, and elimination of these agents [15]. At low concentrations, ASO binds to trans-membrane receptors, whereas at high concentrations binding proteins are saturated and ASO is internalized by pinocytosis forming intracellular bodies [16,17]. ASO intracellular uptake is a tightly regulated process with the characteristics of active transport as it is inhibited by ATP depletion. Many nanoparticles, positively charged polymers, metal nanomaterials and other kinds of natural and synthetic polymers can be used to increase ASO uptake [18,19]. Following the in vivo administration, the naturally occurring phosphodiester linkage of ASOs can be rapidly degraded by nucleases, which are abundant in plasma and tissues [20]. Various chemical modifications have been used to limit nuclease-mediated ASO degradation and recent reviews have described in detail the chemical approaches to overcome the nuclease sensitivity of ASOs [19,21,22]. The substitution of sulphur for oxygen to generate phosphorothioate (PS) internucleotide linkages is the most common means to date to stabilize nucleic acid molecules against nuclease-mediated degradation [23]. In addition to increased nuclease resistance, PS linkages enhance the binding of ASOs to serum proteins (e.g. albumin) leading to prolonged circulation in plasma and decreased clearance by the kidney. In addition, PS oligonucleotides are able to activate RNase H enzymes and can be easily administered by different routes (e.g. subcutaneous, intravenous, intravitreal, topical, aerosol, oral or enema) due to their high solubility [23]. However, these molecules can induce sequence-independent, but length-dependent binding to various cellular proteins, such as fibronectin and laminin, and cause potentially serious off-target effects [24,25]. For this reason, additional modifications of the backbone sugar moieties, such as 2′hydroxyl methylation or methoxyethylation (2′OMe and MOE, respectively) or constraints such as the 2′-O, 4′-C methylene bridge in locked nucleic acids have been made to develop second- and third-generation ASOs [26]. Uncharged variations on the regular structure of oligonucleotides [e.g. phosphorodiamidate morpholino (PMO) and peptide nucleic acid (PNA)] have been also used [27]. This new generation of ASOs has increased affinity to target RNA, enhanced metabolic stability, and improved pharmacokinetic and toxicity profiles. Because of their unusual structure, PMOs do not interact with cellular proteins and tend to have fewer systemic toxicities compared with other ASOs. In addition to classic ASOs, whose activity is strictly dependent on RNAse H1, the PMOs and PNAs work through steric blocking [28]. Despite the potential of ASOs, significant hurdles remain in translating the promising preclinical results into effective therapy. One of the greatest limitations for these compounds is that they are rapidly cleared from the circulation so that only a fraction of the ASOs reaches the target tissues unless the ASOs are used at very high doses. Specific administration routes (e.g. intra-ocular, oral or rectal administration) can be used to enhance tissue accumulation of ASOs. For a more comprehensive discussion of these and more recent chemical modifications, the reader is directed towards excellent reviews [22,29,30]

Antisense oligonucleotide targeting ICAM-1 (alicaforsen)

The continuous recruitment of activated leucocytes from the bloodstream to inflamed gut tissue is one of the mechanisms involved in the amplification and maintenance of the pathological process in IBD [31]. These cells, which are activated in Peyer’s patches and isolated follicles, enter the circulation and eventually home back to the gut. Leucocyte adhesion and extravasation across the intestinal endothelium occur through a multistep process whereby circulating leucocytes are captured, roll, undergo activation, firmly adhere and finally transmigrate into the damaged tissue [32]. Firm adhesion and transmigration are mediated by the stable binding of integrins expressed on the leucocyte surfaces and members of immunoglobulin superfamily [i.e. intercellular adhesion molecule (ICAM)-1, ICAM-2 and VCAM-1] on endothelial cells [33]. Chemokines, produced within the inflammatory microenvironment, act as potent chemoattractants for their cognate receptors on the rolling leucocytes to promote their activation and migration across the endothelium [34].

ICAM-1 is a transmembrane glycoprotein constitutively expressed on the surface of intestinal epithelial cells and vascular endothelial cells and its expression is increased in the inflamed colonic tissue of patients with CD and patients with UC [33,35]. Initial studies in experimental models of colitis showed that ICAM-1 inhibition was effective in dampening inflammation [36–38]. These results provided proof-of-concept for the subsequent development of a human ASO inhibitor of ICAM-1 by ISIS Pharmaceuticals for the treatment of IBD. ISIS 2302 (alicaforsen) is a PS ASO that selectively inhibits ICAM-1 RNA. In a pilot study, 20 patients with active, steroid-resistant/dependent CD were randomized to receive 13 intravenous infusions of alicaforsen or placebo for 26 days in a double-blinded manner [39]. At the end of treatment, 47% of alicaforsen-treated patients and 20% of placebo-treated patients achieved clinical remission [39]. However, these promising results were not confirmed by two subsequent double-blind, placebo-controlled clinical trials. In the first trial, alicaforsen was given subcutaneously to 75 steroid-refractory CD patients. The primary end point, steroid-free remission at week 14, was not reached [40]. In the second, placebo-controlled trial, 299 steroid-dependent CD patients were randomized to receive alicaforsen intravenously at the dose of 2 mg/kg three times a week for 2 or 4 weeks or placebo. Similar proportions of patients in the three groups achieved the primary end point (i.e. steroid-free remission) at week 14. The rates of response were similar in the treatment arms at the various time points [41]. Later on, ISIS pharmaceutical company announced that alicaforsen was not superior to placebo in inducing remission in 331 patients with moderate to severe CD in two phase III clinical trials. Therefore, the therapeutic development of alicaforsen in CD was eventually discontinued. The reason why alicaforsen failed in CD remains unknown. ICAM-1 is just one of the various molecules involved in leucocytes trafficking and, therefore, even in the absence of ICAM-1, other molecules could promote recruitment of activated leucocytes [42]. Another possibility is that, following the systemic administration, the concentration of alicaforsen accumulated within the gut tissue could not be sufficient to knockdown ICAM-1.

However, alicaforsen was revisited when an enema formulation was developed to deliver the ASO directly to the inflamed mucosa of UC patients or patients with pouchitis. In an initial, phase II randomized, double-blind, placebo-controlled trial, 40 patients with mild to moderate distal UC were enrolled and treated daily with 240 mg alicaforsen or placebo enema [43]. Alicaforsen enema formulation was effective in a dose-dependent manner and was well tolerated by the patients. A subsequent study compared the efficacy of alicaforsen with mesalamine enema. One-hundred and ninety patients were randomized to receive 120 or 240 mg alicaforsen enema or 4g mesalamine [44]. Although there was no difference in 6-week response rates among the three groups, the median duration of response to alicaforsen enema was longer than that to mesalamine, suggesting that alicaforsen has comparable induction efficacy to mesalamine but can produce a more durable response [44]. Similar results were seen in another phase II double-blind, placebo-controlled study in acute mild to moderate left-sided UC, where there was no difference in terms of clinical response between 240 mg alicaforsen enema and placebo but patients receiving the active drug had a significantly lower rate of relapse compared with placebo [45]. Pharmacokinetic observations showed that alicaforsen enema had a very low systemic absorption following rectal administration, whereas the ASO was present at high concentration in colonic tissue [46].

In an open-label, uncontrolled study alicaforsen enema was administered nightly to 12 patients with chronic unremitting pouchitis for 6 weeks [47]. The drug was well tolerated and effective in reducing clinical and endoscopic signs of inflammation. Altogether, these results led FDA to consider alicaforsen as an orphan drug, which can be prescribed as an unlicensed medicine in accordance with international regulation in patients with pouchitis and patients with left-sided UC.

Antisense oligonucleotide targeting NFκB

NF-κB is a transcription factor, which regulates the expression of multiple genes involved in a variety of biological processes and it is supposed to play a key role in immune-mediated diseases, cancer and infections, as well as in the control of immune homeostasis in the gut [48–50]. Pioneering studies in two experimental models of intestinal inflammation, namely the trinitrobenzene sulphonic acid (TNBS)-induced colitis and colitis developing in IL-10-deficient mice, documented up-regulation of the p65 subunit of NF-κB in inflamed colons [51]. It was also shown that mice given a specific ASO targeting the p65 subunit of NFκB exhibited reduced signs of colitis [51]. Knockdown of p65 with ASO was also beneficial in mice with DSS-induced colitis [52]. It was also demonstrated that NFκB was up-regulated in intestinal macrophages and endothelial cells of CD patients and inhibition of p65 in CD mucosal cells with the specific ASO reduced production of inflammatory cytokines [51]. Furthermore, knockdown of NFκB reduced both inflammation and fibrosis in a murine model of colitis-induced intestinal fibrosis [53]. In 2011, Tahara et al. used chitosan-modified poly (D,L-lactide-co-glycolide) nanospheres for the delivery of NF-κB ASO in the colon of mice [54]. This approach enhanced the stability of ASO and administration of the compound to mice ameliorated DSS-colitis. Despite these very promising pre-clinical results, no data about the use NF-κB ASO in human IBD have been yet published.

Antisense oligonucleotide targeting Smad7

In recent years, a large body of preclinical and clinical data have been accumulated to support the notion that amplification of the inflammatory process in IBD is sustained by defects in counter-regulatory mechanisms/factors [6]. One such a mechanism involves the anti-inflammatory cytokine transforming growth factor (TGF)-β1 [55]. Initial studies in mice with experimental colitis showed that lack of TGF-β1 activity exacerbated colitis, whereas increasing its activity reduced mucosal inflammation [56–58]. The Smad pathway mediates mainly the immune-regulatory role of TGF-β1. Binding of TGF-β1 to its heterodimeric receptor promotes the phosphorylation of Smad2 and Smad3 [59]. Once phosphorylated Smad2/3 binds to Smad4 to form a complex that translocates to the nucleus where it regulates target genes [60]. TGF-β1-associated Smad pathway is a tightly controlled phenomenon influenced by other intracellular proteins such as Smad7, which acts as a negative regulator of TGF-β1/Smad signalling. In inflamed tissue of IBD patients, TGF-β1 is produced by many immune and non-immune cells but the activity of the cytokine is impaired as indicated by reduced levels of Smad2/3 phosphorylation in IBD mucosa, a finding that has been associated to high Smad7 [61]. Indeed, knockdown of Smad7 in IBD LPMC with a specific ASO leads to TGF-β1-dependent decrease of inflammatory cytokine production [61]. These findings are in line with those generated in murine studies showing that Smad7 is up-regulated and TGF-β1-associated Smad signalling is impaired in the colons of mice with TNBS- and oxazolone-colitis and oral administration of Smad7 ASO restores TGF- β1 activity and dampens mucosal inflammation [62]. Pharmacokinetic studies in colitic mice showed that Smad7 ASO was taken up by epithelial cells and lamina propria mononuclear cells in the small intestine and colon, whereas there was no uptake in liver and spleen, suggesting that oral administration of the ASO associates with poor systemic bioavailability.

These studies led to the development of a pharmaceutical compound named GED0301, and later on mongersen, containing Smad7 ASO. A phase I, open-label, single-centre, dose-escalating study was conducted in 15 active, steroid-dependent/resistant CD patients: three cohorts of patients received mongersen at the dose of 40 mg, 80 mg and 160 mg daily for one week [63]. The safety profile of the drug was very good as no drug-related adverse event was documented. Treatment associated with clinical benefit in all the patients. At day 28, clinical response was seen in all the patients and clinical remission was registered in 86% of the patients. At day 84, 60% of the patients were still in remission. The drug was barely detectable in one sample of one patient at a single time point confirming the poor absorption of the ASO following oral administration. As TGF-β1 is pro-fibrogenic [64], patients enrolled in phase I study were monitored for the development of strictures by ultrasonography [65]. At month 6, no patient developed strictures. In line with these data, we have recently shown that knockdown of Smad7 with the specific ASO reduces rather than increasing the occurrence of fibrosis in a mouse model of colitis-driven intestinal fibrosis [66].

Next, a phase II, multicentre, double-blind, placebo-controlled clinical trial was conducted in 166 patients with active, steroid-dependent/resistant CD [67]. Patients were randomized to receive placebo or 10, 40 or 160 mg mongersen for 2 weeks. The primary end-point of the study was clinical remission at the end of the 2 week-treatment maintained for at least another 2 weeks. At week 4, 55% and 65% of the patients receiving 40 mg and 160mg/day mongersen, respectively, were in clinical remission compared with 10% of patients receiving placebo, whereas there was no significant difference in terms of remission between patients receiving 10 mg/day mongersen (12%) and placebo. The study confirmed the safety profile of mongersen. Responders to mongersen had reduced serum levels of CCL20, a chemokine involved in the recruitment of immune cells to the gut wall during inflammation [68]. A subsequent phase II, multicentre, clinical trial demonstrated the efficacy of the drug in inducing endoscopic improvement in more than one-third of the patients. Moreover, this trial confirmed the clinical benefit of mongersen. Of the 63 active CD patients randomly assigned to receive 160 mg/day mongersen for 4, 8 or 12 weeks, 32%, 35% and 48%, respectively, were in clinical remission at week 12 [69]. However, a phase III clinical trial has been recently suspended due to an interim analysis documenting the lack of efficacy of mongersen.

DNAzyme targeting GATA3 as a novel therapy in ulcerative colitis

GATA3 is a transcription factor involved in T-cell development, which is both necessary and sufficient for Th2 cytokine gene transcription [70]. Recent experimental data support the notion that GATA3 plays a role in the development of colitis. Popp et al. showed that GATA3 RNA transcripts were higher in colonic tissue of UC patients than in controls and GATA3 levels in mucosal T lymphocytes from patients with active UC correlated with mucosal Th2 and Th9 cytokine expression [71]. In addition, GATA3 expression was increased in lamina propria T cells isolated from colitic mice compared with control animals and conditional GATA3 deficiency in T cells prevented experimental UC-like oxazolone-induced colitis [71]. These data are consistent with the demonstration that overexpression of GATA3 in T cells from transgenic mice augments DSS-induced colitis [72]. To further support the pathogenic role of GATA3 in the gut, colitic mice were given intrarectally anti-GATA3 DNAzyme [71]. DNAzymes are single-stranded DNA molecules that are characterized by their ability to specifically cleave RNA molecules after appropriate binding [73]. Mice treated with anti-GATA3 DNAzyme exhibited reduced expression of GATA3, produced less inflammatory cytokines and were less susceptible to oxazolone- and TNBS-induced colitis [71]. Moreover, anti-GATA3 DNAzyme attenuated inflammation in a TNF-independent model of colitis (TNF-R1/2 knockout mice) [71] raising the possibility that such an approach can be a valid option for dampening TNF-independent mucosal inflammation, which is supposed to sustain disease activity in UC patients refractory to TNF blockers.

SiRNA targeting STNM01 in Crohn’s disease

The pathogenesis of fibro-strictures, one of the major complications in CD [74], is not fully understood. However, fibrosis is supposed to be the late stage of an inflammation-driven process resulting in an altered balance between matrix deposition and degradation by specific enzymes [75]. Thus, enzymes controlling matrix remodelling could represent therapeutic targets in stricturing CD. Carbohydrate sulphotransferase 15 (CHST15) is a putative type II transmembrane Golgi enzyme, which catalyses sulphation of chondroitin sulphate (CS) to produce E-disaccharide units, which bind to various pro-inflammatory and pro-fibrotic mediators, adhesion molecules, receptors for advanced glycation end-product and pathogenic microorganisms [76]. Oversulfated CS enhances collagen deposition and binds to type V collagen, a subtype of collagen that accumulates in CD intestinal strictures [77]. CHST15 is increased in the colon of active CD patients [78]. Altogether, these observations support the potential involvement of CHST15 in intestinal fibrogenesis.

STNM01 is a synthetic, double-stranded RNA oligonucleotide directed against CHST15. In mice with acute and chronic DSS-colitis, silencing of CHST15 reduced inflammation and collagen deposition in colonic tissue [79]. This was associated with improvement of histological and endoscopic scores. Later on, a phase 1, randomised, double blind, placebo-controlled, clinical trial evaluated the safety of STNM01 in patients with CD [80]. During a colonoscopy, 18 CD patients with mucosal lesions refractory to conventional therapies received a submucosal injection of 2.5, 25 or 250 nM STNM01 (three patients per group) or placebo (nine patients). STNM01 was not detected in the circulation from 5 min to 24 hours after the injection. No drug-related adverse events were registered and STNM01 was well tolerated by all the patients. One month after the injection, the intestinal expression of CHST15 RNA was reduced, and six out of nine CD patients receiving STNM01 showed improvement in endoscopic score compared with placebo-treated patients. Furthermore, histological analysis of intestinal sections showed that STNM01 reduced the extent of intestinal fibrosis. The small number of patients enrolled in the study does not allow to make any statistical analysis and some of the patients treated with STNM01 had minimal endoscopic lesions at baseline. It is also noteworthy that the therapeutic target of the study was a single mucosal lesion, in which STNM01 was injected; therefore, it remains unclear whether such an approach can be adopted in patients who have extensive disease.

Oligonucleotides targeting TLR-9

Toll-like receptors (TLRs) are proteins that recognize structurally conserved molecules derived from microbes and activate various intracellular pathways involved in the control of innate immune responses [49,81]. TLR-9 is mainly expressed on antigen presenting cells, such as dendritic cells, macrophages and B cells and identifies bacterial DNA sequences with a high content of unmethylated cytidine-phosphate-guanosine (CpG) [81,82]. TLR-9 is over-expressed in colonic lesions of UC patients and its levels significantly correlate with the severity of intestinal inflammation and expression of inflammatory cytokines [83]. Pre-clinical studies showed that the TLR-9 agonist DIMS0150 (Kappaproct), a 19 base single-stranded, partially modified synthetic oligonucleotide, induced the production of anti-inflammatory cytokines, such as IL-10 and type I interferons, which have been previously described as inducers of response to steroids in specific cell types [84–87]. Indeed, DIMS0150 increased steroid sensitivity in cells isolated from steroid-resistant UC patients [88]. A proof-of-concept, randomized, double-blinded study was carried out in 34 active steroid dependent and/or resistant UC patients, who received a single rectal dose of DIMS0150 (30 mg) or placebo [88,89]. Improvement of clinical symptoms was observed one week after a single dose of DIMS0150 and treatment associated with sustained clinical response or remission. DIMS0150 demonstrated a very favourable safety profile with no significant differences in adverse events between the treatment group and placebo group. In a further study, eight patients with severe active UC received a single intracolonic administration of DIMS0150 during colonoscopy and this was associated with durable beneficial effects [90]. In a subsequent phase III study, 131 UC patients refractory to TNF blockers and immunosuppressive drugs were treated with either two rectally administered 30 mg doses of DIMS0150 or placebo at baseline and week 4 (91). In this study, the primary end-point (i.e. induction of clinical remission) was missed. However, various secondary end-points, such as symptomatic remission, endoscopic measures of disease activity and reduction in colectomy rate were achieved. The drug was safe and well-tolerated by all patients and no serious adverse event was registered [91].

More recently, a multicentre, open-label, phase IIa clinical trial was carried out in UC patients, who received another TLR-9 modulator (BL-7040) [92]. The drug was orally administered to 22 UC patients for 5 weeks (12 mg for the initial 3 weeks and 40 mg for the subsequent 2 weeks). Sixteen patients completed the study protocol; six patients discontinued the study due to either adverse events or UC exacerbation. Clinical remission was achieved in two patients and clinical response in half of the patients who completed the study. Furthermore, mucosal neutrophils and IL-6 levels were significantly reduced in responders compared with non-responders. The drug was well tolerated and all registered adverse events were mild with only one serious adverse event not related to treatment [92]. Although these results are preliminary, TLR-9 modulation seems to be a promising approach in UC treatment strategies.

Pre-clinical evidence for additional antisense oligonucleotides

Additional ASOs have been investigated in experimental models of colitis. A specific anti-CD40 ASO that inhibits the interaction between CD40 and CD154, an event crucial in antigen presentation and supposed to contribute in the IBD pathogenesis [93,94], was effective in suppressing inflammation in rats and mice with TNBS-colitis [95]. The ASO targeting the mucosal addressin cellular adhesion molecule (MAdCAM)-1, an adhesion molecule that mediates leucocyte trafficking to the intestine, improved inflammation in mice with TNBS-colitis [96]. Moreover, an ASO targeting macrophage migration inhibitory factor (MIF), a pro-inflammatory cytokine [97,98], ameliorated experimental colitis in mice [99]. In this case, the delivery of the ASO targeting MIF to the gut was obtained with the use of schizophyllan, a polysaccharide of the glucan family that increases stability and is effectively taken-up by macrophages following binding to dectin-1 [100]. As the development of antibodies against anti-TNF represents the main cause of secondary failure to these drugs in IBD [101], ASOs targeting TNF could represent a valid alternative to overcome the loss of response to anti-TNF. In the case of TNF-α, the antisense approach was conceived for the delivery of the drug to specific cell subtypes, such as macrophages [102]. In one case, the ASO targeting TNF was formulated as a nano-complex, based on galactosylated low molecular weight chitosan that is directly taken-up by activated macrophages and was successfully used in mice with TNBS- and T-cell transfer colitis [103]. Another group combined TNF ASO with a cationic glucomannan phytagel that was effectively phagocytized by macrophages and reduced inflammation in DSS-colitis [104]. However, as macrophages do not represent the only cell source of TNF in inflamed intestine of IBD patients, these compounds are unlikely to fully substitute monoclonal antibodies currently used in daily practice.

Conclusions

The better understanding of the risks of off-target nucleotide binding and the use of sophisticated strategies to eliminate more toxic molecules have largely improved the process of oligonucleotide development. Nonetheless, no ASO developed for IBD patients has yet reached the market. For some ASOs, such as ASOs against CD40, MAdCAM-1 or TNF clinical studies in patients are needed to verify whether such treatments can be useful for attenuating the intestinal inflammation in IBD. The benefit seen in preclinical studies and initial clinical trials in IBD patients for additional ASOs, such as mongersen, alicaforsen and anti-NF-kBp65 was not confirmed by large clinical trials. Although it remains unclear why these treatments failed, it is conceivable that selection of specific subsets of patients or technical issues emerging during the scale-up process may have contributed to generate negative results. There is, however, the possibility that changes in the drug formulation and/or route of administration can allow better delivery and persistence of the active compound in inflamed tissue, with the downstream effect of enhancing the activity of the compound. To date, the results obtained in clinical trials witness the excellent safety profile of ASOs, suggesting that such treatments could be more helpful than other drugs carrying on high risk of adverse events in the maintenance phases of the disease rather than induction of remission. Future studies will help clarify these issues as well as establish whether synthetic oligonucleotides, which stimulate regulatory mechanisms (e.g. TLR-9 agonists), are effective in dampening the IBD-associated detrimental immune response.

Disclosure statement

GM has filed a patent related to the treatment of inflammatory bowel diseases with Smad7 antisense oligonucleotides, whereas IM has no conflict of interest.