Hypoxia-Inducible Factor 1-Alpha Stabilizers in the Treatment of Inflammatory Bowel Diseases: Oxygen as a Novel IBD Therapy?

Abstract

Despite the significant advances in the medical armamentarium for inflammatory bowel diseases [IBD], current treatment options have notable limitations. Durable remission rates remain low, loss of response is common, administration routes are largely parenteral for novel biologics, and medication safety remains a concern. This explains an ongoing unmet need for safe medications with novel mechanisms of action that are administered orally. In line with these criteria, hypoxia-inducible factor [HIF]-1α stabilizers, acting via inhibition of prolyl hydroxylase enzymes, are emerging as an innovative therapeutic strategy. We herein review the mechanism of action and available clinical data for HIF-1α stabilizers and their potential place in the future IBD treatment algorithm.

1. Introduction

Inflammatory bowel disease [IBD], a chronic condition characterized by inflammation of the gastrointestinal tract, may be classified as Crohn’s disease [CD] or ulcerative colitis [UC] and affects more than 6.8 million individuals worldwide.¹ Potential treatment options for IBD include immunomodulators, biologics and small molecules. Unfortunately, all currently available IBD medications have significant limitations, including incomplete induction rates [20–35%], limited long-term remission rates [15–60%] and high rates of complications progressing to surgery [10–20% in treated UC and 70–80% in treated CD patients].^2–11 Moreover, route of administration, specifically injections and infusions, and current safety profiles, which often include a risk of serious infections, malignancies or immunogenicity, may be daunting to patients.^12,13 Thus, there is a monumental need within the IBD community for additional oral treatment options with novel mechanisms of action and superior safety profiles.

To date, most of the research centred on IBD pathogenesis and treatment mechanisms have focused on the immune system. Immune cells have been identified to initiate and perpetuate inflammation, causing chronic gut tissue damage. Thus, all the current IBD therapies target various cytokines and chemokines implicated in the inflammatory process.¹⁴ However, all these medications face a therapeutic ceiling effect.¹⁵ To obtain superior remission rates, there is now a growing recognition that [1] multiple inflammatory pathways may need to be targeted at the same time to attain IBD remission and [2] non-immune cell-mediated tissue damage or immune/non-immune cell interactions may be contributing factors.¹⁴ Thus, the development of new therapeutic approaches will require investigation beyond the current paradigm and understanding of IBD pathogenesis.

2. Hypoxia-inducible factor-1α stabilizers

One such evolution in understanding inflammatory pathogenesis is the recognition of the role of the epithelial cell barrier, whether located in the lung, skin or gastrointestinal mucosa.^16–19 The epithelial cell barriers are designed to allow for the transportation of essential nutrients, fluids and waste while restricting the entry of harmful luminal antigens [Figure 1].²⁰ In inflammatory conditions, alteration in the epithelial cell barrier can occur due to diminished tissue oxygen levels, generating a hypoxic environment. However, functional, metabolic and vascular adaptions to this hypoxic environment ensue with the activation of the hypoxia-inducible factor [HIF] signalling pathway. HIF is characterized by heterodimeric transcription factors that are formed by HIF-α, which is oxygen-dependent, and HIF-β subunits. Well-known isoforms of HIF-α include HIF-1α and HIF-2α, which are both expressed in the cytoplasm and share structural and functional similarities, but produce different transcriptional outputs with regard to erythropoiesis, metabolism and angiogenesis.^16–18 For example, HIF-1α and HIF-2α have varying effects of inducing vascular endothelial growth factor [VEGF], an angiogenic factor that may have a critical role in tissue regeneration and repair.^21,22

Figure 1.

Structure of intestinal epithelial barrier in [A] a normal healthy cell indicating intact tight junction and homeostasis of the lamina propria and (B] setting of inflammation with barrier disruption, microbial translocation, lamina propria inflammatory infiltrate and hypoxia.

When the HIF pathway is activated under hypoxic conditions, HIF transcription factors trigger alternative methods of energy generation and oxygen absorption to permit and foster cell survival and metabolism, leading to expression of multiple metabolic enzymes, angiogenic factors, vasoactive factors and extracellular matrix to enhance barrier function.²³ HIF is regulated by three prolyl hydroxylase enzymes [PHD] and asparaginyl hydroxylase factor inhibiting HIF [FIH], which under normoxic conditions utilize oxygen to hydroxylate HIF and prevent HIF-mediated transcriptional activation through proteasomal degradation.¹⁹ Conversely, in the setting of hypoxia, PHD cannot hydroxylate HIF-α subunits due to the lack of availability of molecular oxygen as a co-factor, thereby permitting the HIF-α subunit to stabilize and bind to HIF-1β in the nucleus to form transcriptionally active complexes [Figure 2]. Hence, HIF is believed to mediate an epithelial barrier protective effect once stabilized. To date, HIF upregulation has been assessed primarily in IBD and colitis.

Figure 2.

In normoxia [A], prolyl hydroxylase enzymes [PHD] utilize oxygen to hydroxylate hypoxia-inducible factors [HIF], leading to HIF-1α degradation and prevention of HIF-mediated transcriptional activation. In the setting of hypoxia, oxygen is not available as a co-factor, thereby allowing HIF-1α to stabilize and induce genes to foster cell survival and metabolism. GB004 inhibits PHD [B], preventing HIF-1α degradation and permitting for HIF-1α stabilization.

3. Preclinical Evidence

3.1. HIF effects in normal vs inflamed intestinal epithelium

The intestinal epithelium in healthy individuals is constantly subjected to transient phases of hypoxia due several factors, including oxygen diffusion into the lumen, epithelial metabolism, fluctuating oxygen demands and countercurrent exchange blood flow.²⁴ However, stimulation of the HIF signalling pathway allows HIF transcription factors to activate and control cell survival and metabolism, permitting the normal intestinal epithelium to maintain its functionality.²⁴ Conversely, the hypoxic environment is more pronounced in individuals with IBD due to several mechanisms that contribute to an ongoing inflammatory process. The epithelial cell barrier becomes compromised in the setting of increased apoptosis and altered expression and assembly of tight junction proteins. This results in leaks and crossover of luminal antigens, which triggers an inflammatory process, including production of inflammatory cytokines [Figure 1].²⁵ This is accompanied by an increase in activated immune cells including neutrophils, macrophages and dendritic cells, which are protected from apoptosis and produce additional pro-inflammatory cytokines, resulting in significant oxygen consumption.^24,26,27 Additionally, inflamed tissues may also contribute to vasculopathy, resulting in reduced blood perfusion and reduced oxygen supply.^27,28

The expression levels of HIF vary between healthy and inflamed intestinal epithelium.¹⁷ In normal intestinal tissues, HIF-1α and HIF-2α are present at low levels with low expression of VEGF identified in the surface and glandular epithelial cells. However, in CD, HIF-2α has been noted in epithelial cells, stromal fibroblasts and myocytes, with HIF-1α only expressed in epithelial cells and stromal fibroblasts, both with a mixed nuclear/cytoplasmic pattern. Conversely, in UC, both HIF-1α and HIF-2α were identified in surface and glandular epithelial cells and stromal fibroblasts as a mixed nuclear/cytoplasmic pattern. Overall, compared to normal intestinal tissues, an upregulation of HIF-α molecules has been noted in patients with active UC or CD.

The effect of HIF isoforms in IBD has been evaluated in several studies. One study with in vivo and in vitro methodologies identified the following: [1] mice with decreased HIF-1 expression experienced severe colitis symptoms [e.g. mortality, weight loss], whereas mice with HIF-1 overexpression experienced protective effects; and [2] HIF-1α played a role in regulating the expression of several barrier-protective genes, such as multidrug resistance gene-1, intestinal trefoil factor and CD73.²⁵ Another murine study identified an association between absence of HIF-1α and downregulation of autophagy in intestinal epithelial cells, exacerbation of colitis, increased levels of harmful gut bacteria, such as Escherichia coli and Bacteroides, and decreased levels of protective gut bacteria, such as butyrate-producing bacteria [Butyrivibrio] and Lactobacillus.²⁹ HIF-1α induced barrier-protective genes by repressing oxazolone- and 2,4,6-trinitrobenzene sulfonic acid [TNBS]-induced colitis whereas HIF-2α increased inflammatory responses in dextran sulphate sodium [DSS]-induced colitis.³⁰ Lastly, PHD1 and PHD3 deficiencies led to a protective effect of the intestinal epithelial cell barrier in mice.³¹ These findings collectively suggested HIF-1α stabilizer via hydroxylase inhibitors as a potential therapeutic approach in IBD.

Several compounds with HIF-1 stabilization properties have been evaluated through in vitro and in vivo studies [Table 1]. One oral compound, GB004 [also known as Akebia-4924 or AKB-4924], emerged as a potential treatment option for patients. GB004 is a small-molecule PHD inhibitor with preference for stabilization of HIF-1α within the gut. In vitro and murine studies assessing its efficacy and safety are discussed below.

Table 1.

Cell line and murine studies evaluating hypoxia inducible factor compounds

Year	Author	Study design	Compound	Findings	Reference
2008	Cummins EP et al.	Murine study with mice treated with vehicle or DMOG IP for 2 days with colons excised and epithelial cells scraped and snap-frozen in liquid nitrogen. Colitis was also induced by DSS treatment and DAI was used to determine body weight, occult blood in faeces, and stool consistency/diarrhoea. Length of colon was measured	DMOG	DMOG showed stabilization of HIF-1 in epithelial-enriched colonic mucosal scrapings and increased NF-kappaB activity; DMOG rendered mice refractory for DSS-induced weight loss and DAI increases, along with less colon shortening and presence of normal faecal pellets in colon and overall reduced DSS-induced colon injury; overall levels of inflammatory markers [e.g. MPO, TNF-α, IL-6, IL-12] were lower in DMOG-treated mice; DMOG also showed accelerated recovery and less reduction in induction of colon epithelial cell apoptosis	Cummins EP, Seeballuck F, Keely SJ, et al. The hydroxylase inhibitor dimethyloxalylglycine is protective in a murine model of colitis. Gastroenterology 2008;134[1]:156–165. doi:10.1053/j.gastro.2007.10.012
2008	Robinson A et al.	Murine study with 10 mice with TNBS-induced colitis compared to control. FG-4497 20 or 40 mg/kg IP administered day before TNBS, 4 h prior to instillation of TNBS and on the day after TNBS administration. Outcomes evaluated after administration	FG-4497	Dosing of FG-4497 20 and 40 mg/kg respectively on day −1, day 0 and day +1 attenuated initial weight loss [p < 0.05 for both], and more significantly protected weight loss with the ongoing course of recovery [p < 0.025 for both]. Mice receiving 40 mg/kg FG-4497 showed less inflammatory infiltrate, more intact epithelium, and overall preservation of tissue architecture	Robinson A, Keely S, Karhausen J, Gerich ME, Furuta GT, Colgan SP. Mucosal protection by hypoxia-inducible factor prolyl hydroxylase inhibition. Gastroenterology 2008;134[1]:145–155. doi:10.1053/j.gastro.2007.09.033
2013	Keely S et al.	Murine study with TNBS colitis mice treated with AKB-4924 [HIF-1 isoform-predominant PHD inhibitor] 0.3, 1 and 5 mg/kg subcutaneously. Outcomes included inflammatory signalling, HIF-mediated epithelial barrier responses, and HIF-driven innate cell activity	AKB-4924	Mice treated with AKB-4924 experienced reduced weight loss, attenuated colon shortening, reduced tissue damage and decreased DAI. Additionally, the treatment group had augmented epithelial barrier function, leading to 50-fold reduction in serum endotoxin. Reduction in serum levels of IL-1B, IL-6 and TNF-α were also noted with treatment	Keely S, Campbell EL, Baird AW, et al. Contribution of epithelial innate immunity to systemic protection afforded by prolyl hydroxylase inhibition in murine colitis. Mucosal Immunol. 2014;7[1]:114–123. doi:10.1038/mi.2013.29
2013	Gupta R et al.	Murine study with TNBS or DSS-induced colitis treated with oral TRC160334. Treatment was initiated at 2 mg/kg/day 1 day prior to induction of colitis and continued until study termination in TNBS-induced colitis mice. Treatment in DSS-induced colitis mice was 2 and 5 mg/kg/day initiated at day 5 and continued until study termination	TRC160334	In both types of induced colitis, TRC160334 reduced DAI measured at day 2 and 4 [9.9% vs 33.6% in vehicle-treated mice]. Treatment also attenuated the severity and extent of colonic damage when compared to vehicle-treated animals and histologically, a significant reduction in severity of inflammatory and necrotic changes were noted. Percentage reduction in body weights was 13.1% vs 21.6% with TRC160334 compared to placebo. Survival rate following treatment was 42% vs 18% in vehicle mice	Gupta R, Chaudhary AR, Shah BN, et al. Therapeutic treatment with a novel hypoxia-inducible factor hydroxylase inhibitor [TRC160334] ameliorates murine colitis. Clin Exp Gastroenterol. 2014;7:13–23. doi:10.2147/CEG.S51923
2015	Marks E et al.	Murine study with TNBS-induced colitis vs vehicle-treated control animals. AKB-4942 5 mg/kg was administered every 48 h IP or orally. Prevention of colitis as measured by endoscopy, histology, barrier integrity, and immune profiling as well as potential off-target effects were assessed	AKB-4924	Mice with AKB-4942 IP treatment showed recovery within 3 days and recovered weight within 4 days. Both oral and IP treatment reduced IL-1B, TNF-α, IL-12p70 and IL-6. Reduced tissue inflammation and increased mucosal barrier function were noted. Histological disease improved in TNBS mice treated with PHD inhibitor. Oral administration of AKB-4924 was noted in ileal tissue but not heart or kidney. IP administration stabilized HIF in ileum, heart and kidney	Marks E, Goggins BJ, Cardona J, et al. Oral delivery of prolyl hydroxylase inhibitor: AKB-4924 promotes localized mucosal healing in a mouse model of colitis. Inflamm Bowel Dis. 2015;21[2]:267–275. doi:10.1097/MIB.0000000000000277
2021	Kim Y et al.	Cell studies + murine study with DNBS or DSS-induced colitis treated with CG-598 or AKB-4924 orally	CG-598	Mice treated with CG-598 had amelioration of intestinal inflammation and reduction in inflammatory lesions and pro-inflammatory cytokines. Barrier function was also noted to be boosted with expression of intestinal trefoil factor, CD73, E-cadherin and mucin	Kim YI, Yi EJ, Kim YD, et al. Local Stabilization of hypoxia-inducible factor-1α controls intestinal inflammation via enhanced gut barrier function and immune regulation. Front Immunol. 2021;11:609689. doi:10.3389/fimmu.2020.609689
2021	Goggins BJ et al.	Cell studies + murine study with TNBS-induced colitis treated with GB-004 5 mg/kg IP or vehicle solution. Disease activity was measured by endoscopy and colon were excised for histological analysis	GB-004	Cell study wise, wound closure and migration was noted along with functional blockade of certain integrins. In treated murine models, mucosal healing through induced epithelial integrin expression was noted	Goggins BJ, Minahan K, Sherwin S, et al. Pharmacological HIF-1 stabilization promotes intestinal epithelial healing through regulation of α-integrin expression and function. Am J Physiol Gastrointest Liver Physiol. 2021;320[4]:G420–G438. doi:10.1152/ajpgi.00192.2020

DAI, disease activity index; DDS, dextran sulphate sodium; DMOG, dimethyloxaylglycine; IP, intraperitoneally; NF-kappaB, nuclear factor—kappaB; TNBS, 2,4,6-trinitrobenzenesulfonic acid.

3.2. In vitro studies

The effect of GB004 on barrier integrity using a human differentiated monolayer assay and in vitro immune cell co-culture system was explored.³² GB004 treatment was applied to human stem cell-derived epithelial monolayer platforms under normal conditions and those stimulated with tumour necrosis factor-α [TNFα] to induce barrier damage. Barrier integrity, H1F-1α genes, tight junction function and epithelial monolayer viability were assessed through several mechanisms, including transepithelial electric resistance, cell lysates and immunofluorescence staining. GB004 was found to preserve barrier integrity, promote tight junction function, and prevent TNFα-induced cell death by upregulating genes involved in barrier function and integrity and modulating immune cell function to reduce key mediators associated with inflammation, immunomodulation and tissue remodelling, such as IL-8, MCP-1, sIL-17F and collagen I.

The above findings were further supported by another study evaluating the effect of GB004 on gene expression, tight junctions and barrier integrity using intestinal epithelial cells and mouse epithelial organoids.³³ GB004 was found to induce HIF-1α-dependent gene expression in murine organoids, stabilize HIF-1α in intestinal epithelial cells, and prevent loss of barrier integrity in intestinal epithelial cells subjected to cytokines to induce barrier damage.

3.3. Animal data

One of the first studies evaluating the effect of AKB-4942 in mice with TNBS-induced colitis found that treatment led to reduced weight loss, attenuated colon shortening, reduced tissue damage and decreased disease activity indices when compared to vehicle.³⁴ Additionally, treatment led to decreases in intestinal permeability. AKB-4942 was also assessed in mice that spontaneously developed transmural CD-like inflammation in the terminal ileum, with treatment leading to significant decreases in histopathology, acute and chronic inflammation, villus distortion and inflammatory indices.

In another study involving mice with TNBS-induced colitis, oral AKB-4924 led to amelioration of inflammation, defined by weight gain, reduction in inflammatory cytokines such as IL-1β, TNF-α, IL-12p70 and IL-6, reduction in tissue inflammation and histological disease via endoscopic and pathological evaluation, and increased mucosal barrier function when compared to vehicle.³⁵ Additionally, HIF stabilization was noted, specifically in the ileal tissue, with higher levels of HIF and VEGF in nuclear extracts of mice treated with AKB-4942, indicating gut preference.

The gastrointestinal exposure relative to plasma and most effective dose following oral administration of GB004 in mice with and without colitis was evaluated.³⁶ GB004 1, 3, or 10 mg/kg was administered as a one time dose or once daily for 8 days. Anti-CD40 monoclonal antibody was administered intraperitoneally to induce colitis. GB004 demonstrated PHD inhibition in vitro and produced significantly higher serum concentrations in the colon than plasma irrespective of dose administered in both healthy and colitis mice. Moreover, a change in gene expression for NT5E, Cldn1 [both genes involved in barrier integrity], Bnip3 [member of the apoptotic Bcl2 family], and IL12b, Lcn2 and STc1 [all involved in inflammation] was noted. In addition to inducing genes important to mucosal healing, GB004 was also found to promote histological healing in mice with colitis compared to healthy controls.

4. Clinical Data

Given the findings from cell lines and murine colitis studies demonstrating significant reductions in disease activity, improvements in histological measures and greater concentration in gastrointestinal tissue, GB004 is undergoing evaluations to determine its efficacy and safety in humans. A phase I study determined that a single dose of GB004 in normal healthy subjects was well tolerated without any serious adverse events. To determine the safety and pharmacokinetic profile with multiple doses, a phase 1A multiple ascending dose [MAD] study at a single site in Canada enrolled and randomized healthy participants 18–55 years old to receive GB004 60, 120 or 240 mg or placebo once daily for 8 days.³⁷ Of the 42 study participants, most were white with a mean age of 47 years [range 27–55]. Primary endpoints included assessments of safety and tolerability parameters, as defined by adverse events or clinically significant changes in vital signs, electrocardiogram or laboratory findings, throughout the study and at day 15 of follow-up. Pharmacokinetic parameters and plasma levels of erythropoietin, VEGF and HIF-1α expression were determined by immunohistochemistry staining in colonic biopsies. Since HIF-1α stabilization can have systematic on-target effects, including erythropoietin and VEGF modulation, which are both physiological responses to hypoxia, assessment of plasma levels was an exploratory endpoint to determine the extent of gut selectivity with GB004. All subjects in each study group with the exception of 63% in the GB004 240-mg group received all planned doses. Adverse effects were comparable between GB004 [69%] and placebo [60%], with most on GB004 experiencing mild effects of dizziness [31%], headache [28%], diarrhoea [28%] or nausea [25%]. Two patients on GB004 discontinued the 240-mg dose due to vomiting and non-cardiac chest pain. Regarding pharmacokinetics and pharmacodynamics, GB004 was rapidly absorbed and rapidly eliminated from systemic circulation, with concentrations being three- to four-fold higher in colonic tissue vs plasma. While increases in HIF-1α expression were noted compared to baseline, no dose-related changes were observed with plasma levels of erythropoietin or VEGF, further highlighting the gut-predominant effect of GB004.

At present, GB004 is being clinically developed for treatment of UC. A phase 1B, double-blind, placebo-controlled study randomized adult participants with mild-to-moderate active UC (defined as total Mayo score of 3–12 with a centrally read Mayo endoscopic subscore ≥ 1 and presence of blood in stool plus Robarts Histopathology Index [RHI] ≥ 4 with neutrophils in the epithelium) to GB004 120-mg solution [n = 23] vs placebo daily [n = 11] for 4 weeks.³⁸ The primary endpoint was GB004 safety and tolerability. Additional endpoints included pharmacokinetic parameters and pharmacodynamic response. Of the 33 [97.1%] participants who completed the study, the mean age was 45.4 [12.71] years, 67.6% were male, 91.2% were white and the mean duration of UC diagnosis was 5.59 [6.1] years. Mean baseline Mayo score was 7.5 [2] and RHI score per sigmoid and rectum biopsy was 14.1 [18.63] and 17.4 [8.94], indicative of moderate UC. Approximately 39.1% on GB004 vs 27.3% on placebo experienced an adverse event, with nausea [21.7%] and dysgeusia [13%] being most common with GB004. One GB004-treated subject discontinued treatment after 21 days due to worsening UC, but this was deemed to be unrelated to study medication.

Pharmacokinetically, GB004 demonstrated rapid absorption and rapid clearance with low systemic exposure and high levels in colonic tissue compared to plasma. Pharmacodynamically, changes in HIF-1α expression in sigmoid colon biopsies occurred from baseline to day 28 with GB004 vs placebo (mean change of 11.43%, 95% confidence interval [CI] −5% to 27.9%), indicative of local target engagement. Reduction in inflammation with GB004 vs placebo was noted with mean percentage change from baseline in faecal calprotectin of −30.4% [95% CI −131.6% to 70.7%] and mean myeloperoxidase-positive cell proportions in sigmoid colon biopsies [–7.2%; 95% CI −20.1% to 5.6%]. Lastly, GB004 increased the concentration of faecal secretory immunoglobulin A, a barrier homeostasis and local immune defence biomarker, compared to placebo [87.16%, 95% CI −215.28% to 389.6%]. Although completely exploratory, efficacy outcomes at 28 days were also evaluated and are listed in Table 2. Overall, efficacy analyses were favourable for GB004 when compared to placebo.

Table 2.

Exploratory clinical outcomes at 28 days

Clinical parameter	Definition	Proportion meeting outcome [%]		Difference, 95% confidence interval [%]
		GB004	Placebo
Clinical response	Reduction in Mayo score of ≥ 3 points and ≥ 30% from baseline with accompanying decrease in rectal bleeding score ≥ 1 or absolute rectal bleeding score ≤ 1, among subjects with baseline rectal bleeding score > 0	30	18.2	11.8 [−24.9, 41.3]
Clinical remission	Mayo score ≤ 2, utilizing sigmoid endoscopic subscore among individuals with baseline sigmoid endoscopic subscore ≥ 1, with no individual subscore > 1	4.5	0	4.5 [−24.5, 23.3]
Resolution of rectal bleeding	Rectal bleeding subscore = 0, among subjects with baseline rectal bleeding subscore ≥ 1	57.1	36.4	20.8 [−18.4, 53.8]
Mucosal healing	Achieving improvement in endoscopic appearance and histological remission in same colonic segment	17.4	0	17.4 [−12.8, 38.8]
Histological remission	Sigmoid or rectum RHI score ≤ 3 with lamina propria neutrophils and neutrophils in epithelium subscores of 0 [among patients with corresponding baseline lamina propria and neutrophils in epithelium subscores > 0]	43.5	18.2	25.3 [−10.4, 53.5]
Endoscopic appearance improvement	Sigmoid or rectum endoscopic subscore ≤ 1 [if corresponding baseline endoscopic subscore > 1] or 0 [if corresponding baseline endoscopic subscore = 1]	17.4	18.2	0.8 [−34.5, 25.9]

MES, Mayo Endoscopic Score; RHI, Robarts Histopathology Index.

Based on the above findings, the study authors noted GB004 to be generally well tolerated with preferential localization to the gut issue. Moreover, trends with biomarkers and clinical activity hints to the potential efficacy of GB004 in improving mucosal healing and reducing inflammation in patients with UC.

In light of the above findings, the efficacy of GB004 in adults with mild-to-moderate active UC is currently being fully evaluated in the SHIFT-UC study, a phase 2 randomized, double-blind, placebo-controlled, multicentre trial [NCT04556383]. The intent of this study is two-fold, with a 36-week placebo-controlled period and a 24-week open-label extension period, aimed to evaluate the efficacy and safety of two doses of GB004 when administered concurrently with aminosalicylates, with or without systemic steroid use. Adults with UC diagnosed at least 3 months prior to the first dose of study medication and currently receiving treatment for UC and on a stable dose for at least 2 weeks prior to flexible sigmoidoscopy or colonoscopy with aminosalicyclate [e.g. mesalamine, sulfasalazine] as monotherapy or in combination with prednisone ≤ 20 mg/day [or equivalent], beclomethasone ≤ 5 mg/day or budesonide/budesonide multi-matrix ≤ 9 mg/day are eligible to participate. Those with prior biologic exposure for UC treatment, presence of bacterial or parasitic infection, prior treatment with tofacitinib, oral cyclosporine, sirolimus or mycophenolate mofetil within 8 weeks of day 1, and use of azathioprine or 6-mercaptopurine within 1 day of day 1 of the study are excluded. The primary endpoint is the proportion of participants with clinical remission at week 12. Additional outcome measures include the proportion of participants with clinical response, endoscopic improvement, mucosal healing and histological remission at week 12 and 36. This study is estimated to be completed by February 2023.

5. Clinical Implications

The Phase I studies provide a promising perspective of GB004. A thorough clinical drug development programme is necessary, but if successful this drug class would offer a significant and advantageous upgrade to the medical armamentarium in IBD. In addition to its novel mechanism of action, gut-preferential profile, oral administration and potentially promising safety profile, GB004 could fulfil many of the unmet needs present with current IBD therapies.

Current treatment options for mild-to-moderate IBD consist of oral or rectal aminosalicylates [e.g. sulfasalazine, mesalamine, balasalazide] and corticosteroids [e.g. prednisone, budesonide]. Due to treatment efficacy, aminosalicylates are used as first-line for induction and maintenance of remission in patients with mild-to-moderate UC; unfortunately, 17–75% of patients treated with aminosalicylates will experience a partial or incomplete response to treatment, warranting additional treatment escalation or change.^39–42 In mild-to-moderate CD, although use of aminosalicylates has not been shown to consistently induce remission or achieve mucosal healing in active disease, this medication continues to be used. Similarly, not every patient with CD will experience or maintain a response to aminosalicylates.^43,44 A previous study identified younger age, endoscopic activity at diagnosis, extensive colitis, early need for corticosteroids, elevated inflammatory markers and non-adherence to be factors predictive of aminosalicylate failure; however, not all of these factors are easily modifiable.⁴⁰ Another treatment option for mild-to-moderate IBD includes corticosteroids. However, despite their effect in relieving symptoms and inducing remission, corticosteroids have not been shown to slow disease progression and can cause serious adverse effects, including osteoporosis, glaucoma, high blood pressure and increased risk of infections, with repeated or long-term use.

For patients warranting treatment change or escalation, the next step is treatment with biologics or small molecules. However, this transition may be daunting to patients given that, compared to aminosalicylates, biologics and small molecules are more expensive, require a higher level of care coordination including prior authorization and navigation of specialty pharmacies and infusion centres, are administered primarily intravenously or subcutaneously, and carry more serious adverse effects. In light of this, patients with mild-to-moderate IBD may perceive the risks and inconveniences of biologics and small molecules as outweighing their treatment efficacy. A previous study demonstrated that patients perceive biologics to be invasive and limiting, corticosteroids and surgery being more beneficial than biologics.⁴⁵ Unfortunately, these perceptions led to an underutilization of biologics and small molecule treatment in a population that would greatly benefit from them. In this clinical scenario, HIF-1α stabilizers could either be a first-line option, combined with 5-ASA or steroids, or fit into the space of 5-ASA non-responders or partial responders as a second-line option. More treatment options for mild-to-moderate IBD remain an unmet need and an oral and safe option for these patients is desirable.

An additional space that has been identified is the optimization of IBD treatment to overcome partial or lack of treatment response with biologics. According to one 5-year follow-up study involving 310 patients with IBD treated with either adalimumab or infliximab, treatment intensification due to partial response was required in 28.9% of patients with CD and 37.1% with UC, and treatment discontinuation due to inappropriate disease control was noted in 40.7% of CD and 40.5% of UC patients.⁴⁶ Another contributing factor to variable biologic response is prior exposure and/or failure of anti-TNF therapies. Subsequent studies have demonstrated that patients previously exposed to anti-TNF therapies generally do not respond well to subsequent non-anti-TNF biologics used, such as vedolizumab or ustekinumab. For example, in the GEMINI 1 trials, >40% of patients with UC experienced prior anti-TNF failure, with a post hoc analysis showing that rates of clinical response at week 6 varied between 53% for anti-TNF-naïve patients to 39% with prior anti-TNF failure.⁴⁷ Generally, for non-responders, a transition to an alternative agent within the same medication class is recommended if drug levels are inadequate or if anti-drug antibodies are detected. Alternatively, if drug levels are adequate, transition to an alternative treatment class is advised.⁴⁸ For partial responders, dose optimization of current biologics and/or, in certain cases, addition of other medications such as immunomodulators is recommended to recapture response.⁴⁸ An additional approach, although currently in very early phases, is the use of dual biologic or small molecule therapy to target multiple disease-mediating mechanisms and collectively attain clinical response and remission.⁴⁹ However, these approaches are not always successful or feasible due to insurance or cost limitations. Thus, variable response to biologic therapy is prevalent in clinical practice and remains a significant concern due to limited treatment options and reduced treatment response and efficacy with subsequent biologic exposure. HIF-1α stabilizers may be an option as a combination option for biologics either initially or in the case of partial responses due to complementary mechanisms of action, targeting non-immune cells.

Of note, oxygen has been previously utilized in IBD management in the form of hyperbaric oxygen therapy [HBOT]. This therapy is administered via a non-conventional treatment approach, in which individuals breathe 100% oxygen while in a pressurized chamber. This results in increased oxygen tension in plasma and enhanced diffusion of oxygen into tissues, thereby reducing hypoxia. Previous systematic reviews and meta-analyses have found HBOT to result in clinical remission rates of 31–88% depending on disease phenotype and minor adverse events, such as barotrauma and inner ear congestion.^50,51 Despite being a different approach, these findings support that therapies which address hypoxia, such as HIF-1α stabilizers, could play a role in IBD management.

In summary, some of the unmet needs with current IBD treatment include: [1] additional treatment options for mild-to-moderate IBD, [2] an effective treatment option for mild-to-moderate CD, [3] a treatment option that can be used to delay or prevent immediate escalation from aminosalicylates to biologics/small molecules, [4] an additional treatment to be used in combination with aminosalicylates or biologics/small molecules to optimize treatment response and efficacy, and [5] additional treatment options that are administered orally without extensive or serious adverse effects. Currently, GB004 is positioned to be a treatment option, which is well tolerated orally without many serious adverse effects due to its gut preferential profile, for mild-to-moderate UC and has potential clinical effects that would permit for use either as monotherapy or in combination with aminosalicylates or biologics. While additional studies are needed to assess GB004’s efficacy and safety in the treatment of mild-to-moderate CD and moderate-to-severe IBD, its unique and targeted mechanism of action may permit GB004 to have a unique niche overall in the IBD treatment algorithm.

6. Conclusion

HIF-1α stabilizers have emerged as a therapeutic area of interest for IBD due to several features including oral administration, targeted and unique mechanism of action, and to date, a mild adverse effect profile. Studies thus far have shown GB004, an oral small molecule that inhibits PHD enzymes and stabilizes HIF-1α, to be generally well tolerated with preferential localization to the gut tissue and with positive effect on barrier function genes. While additional clinical trials are ongoing, if approved for clinical use, GB004 has a significant potential to meet many unmet needs in the IBD population.

Funding

None.

Conflict of Interest

S.B. is a consultant to Bristol Myers Squibb. F.R. is consultant to Agomab, Allergan, AbbVie, Boehringer Ingelheim, Celgene, Cowen, Falk Pharma, Genentech, Gilead, Gossamer, Guidepoint, Helmsley, Index Pharma, Jannsen, Koutif, Mestag, Metacrine, Morphic, Origo, Pfizer, Pliant, Prometheus, Receptos, RedX, Roche, Samsung, Takeda, Techlab, Theravance, Thetis, UCB and received funding from the National Institute of Health, Helmsley Charitable Trust, Crohn’s and Colitis Foundation, Rainin Foundation, UCB, Boehringer-Ingelheim, Pliant, Morphic, BMS and 89Bio.

Author Contributions

All authors have made substantial contribution to all of the following: [1] the conception and design of study, or acquisition of data, or analysis and interpretation of data; [2] drafting the article or revising it critically for important intellectual content; and [3] final approval of the version to be submitted. S. Bhat and F. Rieder contributed to the conception and design of the study, acquisition of data, analysis and interpretation of data, as well as drafting the article and revising the article critically for important intellectual content. The authors had no writing assistance while drafting the article. The manuscript, including related data, figures and tables, has not been previously published and the manuscript is not under consideration elsewhere.

Data Availability Statement

No new data were generated or analysed in support of this research.