Genetic screening in new onset inflammatory bowel disease during anti-interleukin 17 therapy: unmet needs and call for action

KEYWORDS:

• Genetic
• GWAS
• IBD
• IL-17
• psoriasis
• risk prediction
• screening

New onset inflammatory bowel disease (IBD) during anti-interleukin (IL)-17 therapy represents a rare but potentially dramatic adverse event (AE). A recent systematic review with meta-analysis by Yamada et al. found no difference in the risk of developing new onset IBD with anti-IL-17 agents compared to placebo [1]. The authors examined 38 randomized controlled trials (RCT) including 16,690 patients treated with anti-IL-17 drugs. A total of 12 cases of new onset IBD associated with anti-IL-17 agents were found in 5 studies, whereas no cases were observed in the placebo arms. The risk of diarrhea was also evaluated due to the possible underdiagnosis of IBD, yet no difference was found either (Mantel-Haenszel risk difference 0.0013, 95% CI: 0.0014–0.0041, p = 0.34). Although Yamada et al. demonstrated a low incidence rate of new onset IBD with anti-IL-17 agents in patients with existing inflammatory or autoimmune diseases (2.4 cases per 1000 patient-year), their findings should be interpreted with caution, as RCTs may not be the best setting to capture AEs, due to strict initial patient selection [1].

Indeed, a postmarketing study in Vigibase – a unique WHO global database of individual case safety reports – revealed almost one thousand cases, with no recovery in about one third, significant sequelae and even fatal outcomes. More specifically, 1129 gastrointestinal Individual Case Safety Reports (ICSRs) were reviewed, comprising 850 IBD (42.5% Crohn’s disease [CD], 31.9% ulcerative colitis [UC], and 25.6% undifferentiated IBD) and 279 colitis). Secukinumab (SEC, 83.6%) or ixekizumab (IXE, 16.3%) were identified as culprit drugs in 83.6% and 16.3% of total cases, respectively, whereas only one colitis was associated with brodalumab (0.1%). Most IBD and colitis cases occurred within 6 months from treatment initiation in both the SEC (68.8% and 73.5%, respectively) and IXE (100% and 66.7%, respectively) groups. There was no recovery from symptoms in 33.9% and 29.5% of IBD and colitis cases; and sequelae were reported in 5.4% and 4.2% of IBD and colitis cases, respectively. Eight intestinal perforations (0.9%) were observed, including five patients with SEC and three patients with IXE. Most perforations (75%) were experienced by patients treated for psoriasis. Fatal events occurred in four patients (1.2%) in the IBD group (3 after SEC and one with IXE) and two SEC-treated subjects in the colitis group (2.1%). Of the latter, only one was caused by intestinal perforation in hemorrhagic enterocolitis. The patient was on secukinumab for psoriasis vulgaris [2].

Moreover, several individual case reports and case series have been published in the literature [3–8]. Onac et al. reviewed the data of 306 patients (124 with ankylosing spondylitis and 182 psoriatic arthritis) commencing secukinumab, in 10 UK hospitals from 2016 to 2019. They observed that amongst patients who developed gastrointestinal adverse events 4 had definite IBD, 7 had probable IBD while 13 had possible IBD and that all definite IBD cases had concomitant ankylosing spondylitis. Of these, two had preexisting IBD and the other two were de-novo cases, with one requiring surgical care [9].

Summarizing, a risk spectrum appears to exist among anti-IL-17 drugs, with ixekizumab and particularly secukinumab presenting higher chances of associating with new onset CD, UC or indeterminate colitis, and brodalumab positioning on the safer side [2]. At the time of writing, however, it is unclear if these differences depend at least partly on prescribing tendencies, drug availability and recency.

From a pathogenetic point of view, IL-17, and IL-17A particularly, has a protective role in intestinal barrier maintenance. IL-17A-dependent regulation of the tight junction protein occludin limits excessive permeability and bacterial penetration in the event of injury. Consequently, inhibition of IL-17 may result in higher exposure to bacterial products and thus enhanced gut inflammation, leading to higher risk of IBD triggering in susceptible individuals. Unsurprisingly, drugs targeting IL-17A specifically (i.e. secukinumab) have a higher chance of disrupting the intestinal epithelial barrier. Protective IL-17A production was shown to be IL-23 independent [10]. Contrarily to IL-17, IL-23 signaling is involved in the maintenance and expansion of the T helper (Th) 17 lineage via a positive feedback loop that is crucial for the expansion of pathogenic pro inflammatory Th17 cells in CD [11]. Accordingly, IL-23 inhibition proves beneficial in IBD, without impairing gut barrier function: this may explain the lack of new onset IBD with this class of biologics.

Focusing on one of IL-17 inhibitors most popular indications, patients with psoriasis may also possess a higher risk of developing IBDs per se, independently of biologic treatment. Psoriasis and IBD have a strong, common genetic background, with 13 shared predisposition genes, as recently demonstrated by several Genome-Wide Associations Studies (GWAS) [12–14]. Chromosomal loci 6p22 (CDKAL1), 16q (NOD2), 1p31 (IL-23 R), and 5q33 (IL12B) represent some of the most important and extensively studied susceptibility loci shared by the two conditions Other susceptibility loci that have been reported for both diseases include: 20q13, 19p13, 6p21, and 5q31 [15]. Indeed, psoriasis and IBD appear to be strictly related, with a high degree of overlap in genetic substrate, disease risk factors and inflammatory pathogenetic cascades. Thus, it is difficult to ascertain if the mentioned AEs represent actual new onset IBD or flaring of preexisting but silent disease [16].

Other than for psoriasis, psoriatic arthritis and ankylosing spondylitis, employment of anti-IL-17 drugs has been advocated for rosacea and hidradenitis suppurativa, with preliminary reports showing favorable results [17,18]. In light of the possible new indications on the horizon and considering the foreseeable broadening of patient base for anti-IL-17 drugs in the near future, it is imperative to appropriately screen and monitor patients for what could be a life-changing AE. Furthermore, a single case of new onset IBD during anti-IL-17 therapy generates both direct and indirect costs which should be scrutinized closely, together with those of the discontinued anti-IL-17 medication [19]. Ideally, such AE-related expenditures would be offset by an effective screening method even if offered to a large segment of anti-IL-17 candidates.

In current clinical practice, prevention of IBD onset or exacerbation is based on the avoidance of IL-17 inhibitors in patients with first degree family members with positive IBD history or baseline and periodic calprotectin monitoring. Fecal calprotectin (FCP) is considered a marker of intestinal inflammation and has a recognized role in patients with known IBD, being useful in monitoring mucosal healing and predicting clinical relapse or disease recurrence[20]. As far as concerns patients without a known diagnosis of IBD, FCP may prove useful in distinguishing between inflammatory and noninflammatory conditions, but only if gastrointestinal symptoms are already present[20]. Fauny et al. confirmed that FCP could be used in rheumatological patients showing digestive symptoms, to decide in which cases a colonoscopy is indicated[13]. A specific cutoff has not yet been identified. Furthermore, FCP levels can be elevated in patients with active psoriasis and psoriatic arthritis compared with healthy controls [21–24]. A further limitation is that FCP assays differ and results can vary greatly depending on the manufacturer [21].

For these reasons, while certainly representing a noninvasive option, FCP has a limited role in asymptomatic individuals. Nevertheless, FCP dosing in asymptomatic patients eligible for anti-IL-17 drugs is suggested by different authors [10,23,25] before treatment commencement, together with a careful search for IBD family history and gastrointestinal symptoms, always keeping in mind that IBD can also occur in asymptomatic patients without a positive familiar record [10,26]. According to a recent review by Fauny et al., normal FCP values should authorize treatment with anti-IL-17 drugs (a threshold of 250 µg/g was proposed) [23,24]; if FCP is elevated (>250 µg/g) a gastroenterological evaluation is indicated to assess the need for complementary procedures (e.g. colonoscopy) [23]. Although there are no guidelines regarding FCP assessment at baseline and during follow-up for asymptomatic patients treated with anti-IL-17 drugs, different authors suggest this noninvasive test, based on their personal experience [2,16,17]. Unfortunately, this approach may not always prevent IBD onset and self-maintenance. On the other hand, FCP could be of help if gastrointestinal symptoms arise, enabling early treatment interruption and gastroenterological referral for further diagnostic testing [20].

Conversely, as a future perspective, genetic screening may hold promise in a priori risk assessment[27].

GWAS studies have identified more than 200 risk loci for IBD. More specifically, many of the candidate genes linked to IBD susceptibility are involved in activation of T-, B-, and NK-cells, epithelial barrier function and other processes linked to host-microbial interactions [28,29]. Some of the newly discovered candidate target genes, including LY75 (Lymphocyte Antigen 75), CD28 (CD28 Molecule), CCL20 (C-C Motif Chemokine Ligand 20), NFKBIZ (NFKB Inhibitor Zeta), AHR (Aryl Hydrocarbon Receptor) and NFATC1 (Nuclear Factor Of Activated T Cells 1), also regulate specific aspects of the T cell response [28]. The autophagy pathways have been associated to CD pathogenesis since the identification of IRGM (Immunity Related GTPase M) [30], PTPN2 (Protein Tyrosine Phosphatase Non-Receptor Type 2) [31] and ATG16L1 (Autophagy Related 16 Like 1) genes [30]. Recently, ATG4B (Autophagy Related 4B Cysteine Peptidase) has been attributed a central role in this process, emphasizing the importance of autophagy in CD pathogenesis [28]. Also, the increased expression of multiple integrin genes, such as ITGA4 (Integrin Subunit Alpha 4), ITGA8, ITGAL and ICAM1 (Intercellular Adhesion Molecule 1) increases IBD risk [32].

From a translational medicine perspective, the implementation of this knowledge could enable the design of targeted polygenic risk scores in the near future.

In this regard, Chen et al. recently compared four risk prediction models for IBD on a large case-control cohort genotyped for 909,763 GWAS SNPs or 123,437 SNPs on a custom designed Immunochip (polygenic score, best linear genomic prediction, elastic-net regularization and a Bayesian mixture model). One of these methods (the Bayesian mixture approach) had the best prediction performance on average. CD patients with higher risk scores demonstrated clinical characteristics typically associated with a more severe disease course, including ileal involvement and earlier age at diagnosis. The proof-of-concept analyses by Chen et al. demonstrated that the power of genomic risk prediction for IBD is mainly due to strongly associated SNPs with considerable effect sizes. Furthermore, genomic best linear unbiased prediction (GBLUP) with multiple genomic relationship matrices (MGRM) and Bayesan analyses (BayesR) could accelerate genetic advances in both IBD pathogenesis and risk stratification. Finally, although BayesR was confirmed as the best method in terms of prediction accuracy, additional improvements could be achieved using GBLPU and low-density panels with the inclusion of a small number of highly significant pre-selecting whole-genome sequence (WGS) variants [33].

1. Expert opinion

For research purposes, genetic screening before anti-IL-17 therapy initiation could offer a new setting for validation of the mentioned risk scores, eventually improving risk stratification and providing personalized treatment planning. Vice versa, genetics of reported cases of new onset IBD concomitant to anti-IL17 treatment should be studied. Such patients represent in fact rather unique scenarios featuring a known environmental factor in the context of a classic multifactorial disease.

At the time of writing, HLA-C*06:02 is the sole genetic allele with solid evidence as predictor of both susceptibility and treatment response in psoriasis. However, pharmacogenomic studies are on the rise and preliminary results seem particularly promising [34]. In this respect, the study by Morelli et al. is paradigmatic. The authors analyzed the genetics of a cohort comprising 62 patients with psoriasis on secukinumab therapy. Of the 417 SNPs taken into consideration, 8 SNPs in HLA-C and upstream region (rs13207315, rs6900444, rs12189871, rs12191877, rs4406273, and rs10484554), including HLA-Cw6 classical allele (rs1131118), and 3 in MICB-DT (rs9267325), DDX58 (rs34085293) and TYK2 (rs2304255) genes were associated with an excellent response to secukinumab. Importantly, a subgroup of super-responder patients was defined according to rs34085293 or rs2304255 SNP status [34].

Future research will need to aim to global, patient-tailored approaches, integrating not only efficacy but first and foremost safety genetic predictors. This should enable us to maximize benefits on a single patient level and even more so to optimize biologics prescription in a pharmacoeconomic perspective, avoiding the costs of biologic failure and/or related AEs.

Declaration of Interests

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial relationships or otherwise to disclose.

Additional information

Funding

This paper was not funded.