GUCY2C ligand replacement to prevent colorectal cancer

ABSTRACT

Despite advances in screening and prevention strategies, colorectal cancer (CRC) remains the second-leading cause of cancer-related death in the United States. Given this continued public health burden of CRC, there is a clear need for improved disease prevention. CRC initiates and progresses over decades, canonically proceeding via a series of stepwise molecular events that turn a normal epithelium into a dysfunctional epithelium, then subsequently into an adenoma, and finally an invasive adenocarcinoma. An emerging paradigm suggests that guanylyl cyclase C (GUCY2C) functions as a tumor suppressor in the intestine, and that the loss of hormone ligands for this receptor causes epithelial dysfunction and represents an important step in the disease process. In that context, GUCY2C ligand replacement therapy has been proposed as a strategy to prevent colorectal cancer, a translational opportunity that is underscored by the recent regulatory approval of the oral GUCY2C ligand linaclotide (Linzess™, Forest Laboratories and Ironwood Pharmaceuticals, Inc.).

KEYWORDS:

• Chemoprevention
• colorectal cancer
• guanylin
• guanylyl cyclase C
• linaclotide

Abbreviations

CRC	=	colorectal cancer
GUCY2C	=	guanylyl cyclase C
USPSTF	=	United States Preventive Services Task Force
ST	=	heat-stable enterotoxin
ETEC	=	enterotoxigenic E. coli
cGMP	=	cyclic GMP
CFTR	=	cystic fibrosis transmembrane conductance receptor
CIN	=	chromosomal instability pathway
FAP	=	familial adenomatous polyposis coli syndrome
MMR	=	mismatch repair
HNPCC	=	hereditary nonpolyposis colorectal cancer syndrome
CIMP	=	CpG island methylator phenotype
AOM	=	azoxymethane
IBS-C	=	constipation-predominant irritable bowel syndrome
CIC	=	chronic idiopathic constipation
IBD	=	inflammatory bowel disease
MPE	=	molecular pathologic epidemiology

Introduction

Colorectal cancer (CRC) remains a huge public health burden, representing the second leading cause of cancer-related death.¹ In addition to morbidity and mortality, CRC takes a large financial toll on health care expenditures, a cost which is projected to balloon to over 14 billion dollars annually in the next decade.² An old medical adage suggests that “A penny of prevention is worth a pound of cure.” Despite this age-old ideal of preventative medicine, actual development and implementation of preventative strategies has proven difficult. Some diseases have seen astounding success in preventing disease (such as cardiovascular disease where modifiable risk factors can prevent disease), while still others, notably cancer, have lagged behind. Colorectal cancer is relatively unique among cancers as there has been a successful prevention campaign put in place through endoscopic screening, but patient compliance (due to perceived inconvenience of colonoscopy) and effectiveness remain an issue.^3,4 Currently, CRC prognosis is driven by disease stage at presentation. Whereas advanced metastatic disease is incurable, localized disease offers the potential for definitive and curative surgery. Despite screening programs, nearly one fourth of patients present with metastatic disease. Thus, although these interventions have lowered mortality, the continued overall health burden of CRC demands still further efforts to prevent disease and save additional lives.⁵

Unfortunately, despite a litany of promising preclinical chemopreventative agents, translating preventative findings into clinical benefit has remained an elusive goal.⁵ In fact, only aspirin has emerged as a recommended chemopreventive agent against CRC by the United States Preventative Services Task Force (USPSTF), but its use is somewhat controversial in low-risk patients due to GI toxicity.⁶ As such, there is currently no FDA-approved, generally-accepted strategy for chemoprevention of CRC. In that context, the present review highlights the FDA-approved oral GUCY2C ligand linaclotide, an emerging candidate for chemoprevention. Leveraging a large body of preclinical data on safety and efficacy, linaclotide is now being translated into human chemopreventive studies.

Guanylyl cyclase C

Linaclotide functions through engagement of its receptor, guanylyl cyclase C (GUCY2C) in the intestine. GUCY2C is a particulate guanylyl cyclase receptor expressed primarily in intestinal epithelial cells^7-9 and in the brain.^8,10-12 The GUCY2C receptor possesses several cognate ligands which are all structurally similar peptides. These ligands include the endogenous hormones guanylin and uroguanylin as well as exogenous ligands such as the heat-stable enterotoxins (STs) produced by enterotoxigenic E. coli (ETEC).^13,14 Importantly, linaclotide is structurally derived from naturally-occurring GUCY2C ligands,¹⁵ and like these other peptides has a similar ability to increase cyclic GMP (cGMP) levels through GUCY2C.¹⁶ In intestine, GUCY2C is localized to the apical brush border membranes, where it has a well-characterized role in fluid homeostasis.^8,13,17 Upon engagement of the extracellular ligand binding domain of GUCY2C by its ligands, the cytoplasmic catalytic domain catalyzes GTP into cGMP. Cyclic GMP acts as a second messenger to direct cell behavior, including through cGMP-dependent protein kinase (PKG) II, which phosphorylates the cystic fibrosis transmembrane conductance regulator (CFTR) inducing secretion of salt, and therefore water, leading to diarrhea.¹⁸ More recent evidence which is reviewed below has suggested that GUCY2C also functions as a tumor suppressor in the intestine.

Colorectal cancer initiation and progression

Colorectal cancer is thought to arise by a series of mutations that in a stepwise fashion transforms normal epithelium into a malignant tumor. In most cases of sporadic colorectal cancer, chromosomal instability (CIN) leading to biallelic APC loss in intestinal stem cells initiate transformation.¹⁹ Indeed, 80–90% of colorectal tumors harbor biallelic APC mutations,²⁰ and these are sufficient to cause disease.²¹ Consistent with this model, germline mutations in APC in humans result in the familial colorectal syndrome known as familial adenomatous polyposis coli (FAP).²² Canonically, APC functions as a key component of the destruction complex that targets β catenin for degradation.²³ In turn, this attenuates Wnt signaling and limits its oncogenic functions through downstream proto-oncogenes such as c-MYC and Cyclin D1.²³ Conversely, loss of APC function and dysregulation of Wnt signaling leads to adenoma formation.²³ Subsequently, mutations to established tumor suppressors and oncogenes, such as p53 and KRAS, drive progression from adenoma to invasive adenocarcinoma.²³

Mismatch repair (MMR) deficiency is another pathway that can lead to familial CRC (Lynch Syndrome; Hereditary Non-Polyposis Colorectal Cancer HNPCC) or, more rarely (∼15% cases), sporadic CRC in a pathway characterized by microsatellite instability.²⁴ These tumors originate through mutations in mismatch repair genes including MLH1, MSH2, MSH6 or PMS2.²⁴ Deficient MMR leads to an accumulation of somatic mutations that cause disease, including many of the molecular targets discussed above. Yet another pathway, the CpG island methylator phenotype (CIMP), silences tumor suppressor genes via promoter methylation, and functionally resembles MMR tumorigenesis.²⁵

Although colorectal cancer is traditionally considered a disease of irreversible mutational events, the paracrine hormone hypothesis of cancer highlights an alternative model of tumorigenesis in which the loss of the paracrine hormone guanylin potentiates tumorigenesis by silencing the GUCY2C tumor suppressor. Indeed, guanylin expression is universally lost early in colorectal tumorigenesis in an event which is conserved across species.^26,27 This observation suggests that the loss of GUCY2C ligands may represent an important mechanistic event in colorectal tumorigenesis. Interestingly, worldwide there is an inverse epidemiological association between the incidence of colorectal cancer and the risk of being colonized with ETEC expressing ST, consistent with a model in which chronic ETEC supplementation of GUCY2C ligands protect against the development of CRC.^28,29

The functional significance of this guanylin loss is underscored by studies done in knockout mice in which GUCY2C expression is eliminated. These mice exhibit epithelial dysfunction in the intestine characterized by defects in proliferation, migration, metabolic reprogramming, differentiation and genomic integrity along the crypt-villus axis.^30,31 These homeostatic processes that are corrupted when GUCY2C is silenced are the canonical pathways which are universally defective in all cancers in the “Hallmarks of Cancer” model.³² Consistent with this tumorigenic phenotype, GUCY2C knockout mice also display a marked increase in in vivo tumorigenesis in both an azoxymethane (AOM) carcinogen model of intestinal tumorigenesis as well as the genetic APC^min tumor model. Taken together, these findings suggest that the loss of the paracrine hormone guanylin is an initiating event in intestinal tumorigenesis, presenting a unique opportunity for oral hormone replacement to prevent CRC. This paradigm transforms our understanding of colorectal tumorigenesis from a disease of irreversible stepwise mutational events into one of reversible signaling events.

GUCY2C signaling prevents colorectal cancer

The paracrine hormone hypothesis suggests that colorectal cancer can be prevented by oral administration of exogenous GUCY2C ligands which replace the lost guanylin. Consistent with that hypothesis, GUCY2C signaling produces growth arrest of normal human intestinal cells and human colon carcinoma cells in vitro and ex vivo.^33-35 Mechanistically, GUCY2C activation was associated with reduced expression of markers of cell cycle progression including β catenin, cyclin D1, pRb, and increased expression of cell cycle inhibitors including p27. Functionally, GUCY2C signaling restricted transition through the G1/S interface, limiting cell cycle progression without inducing apoptosis.^33-35 These findings suggest that during colorectal carcinogenesis, GUCY2C persists as a dormant tumor-suppressing receptor following loss of its ligand(s), and that it can be re-activated through administration of exogenous ligand which rescues cell cycle regulation and restricts aberrant growth.

Translating these findings, several animal studies have provided pivotal proof-of-concept for administration of GUCY2C ligand to prevent colorectal cancer. One such study utilized oral administration of uroguanylin and demonstrated increased apoptosis and attenuated tumorigenesis in a mouse model of CRC.²⁸ Further studies utilized a genetic approach, generating transgenic mice to overexpress guanylin under the control of an inducible, intestinal-specific Cre recombinase.^36,37 Importantly, transgenic guanylin expression eliminated tumorigenesis in mice subjected to a carcinogen-obesity model of colorectal tumorigenesis.³² Further, obesity-associated epithelial dysfunction in the intestine was ameliorated in these animals, characterized by reversal of the observed defects in proliferation, migration, metabolic reprogramming, differentiation and genomic integrity along the crypt-villus axis.³² Additionally, guanylin transgenic mice exhibited persistent GUCY2C signaling over the lifetime of the animal, without obvious attenuation or desensitization.³⁷ Importantly, administration of GUCY2C ligand was well-tolerated; no adverse events were observed inside or outside of the GI tract, thus further underscoring the safety of GUCY2C ligand administration.^28,36,37 These studies provide compelling preclinical evidence to support the GUCY2C ligand linaclotide as a safe, effective and novel approach to prevent colorectal cancer.

Clinical translation of GUCY2C ligands for colorectal cancer chemoprevention

Although there are 3 similar peptide agonists of GUCY2C in clinical development [linaclotide (Linzess™, Forest Laboratories and Ironwood Pharmaceuticals, Inc.) and SP-304 (plecanatide) and SP-333 (Synergy Pharmaceuticals, Inc.)],^16,38,39 only linaclotide is currently approved by the FDA for treatment of constipation-predominant irritable bowel syndrome (IBS-C) and chronic idiopathic constipation (CIC or CC).⁴⁰ Administration of linaclotide in mice increased cGMP production and induced fluid secretion, consistent with GUCY2C activation.¹⁶ Further, linaclotide exhibited negligible bioavailability, confining its pharmacologic effects to the intestine and minimizing systemic side effects.¹⁶

Several large-scale clinical trials supported the use of linaclotide in treating IBS-C, demonstrating the ability of the compound to improve colonic transit, improve abdominal pain and ameliorate pathologic bowel habits when compared to placebo.^41-47 Moreover, linaclotide's side effect profile was favorable, with the most common adverse event being diarrhea (in approximately 20% of patients), perhaps not surprising given GUCY2C's role in traveler's diarrhea. Other less common events included abdominal pain and distention, flatulence, gastroenteritis and headache.⁴⁸ However, none of these events were considered serious enough to discontinue development of the drug, and the FDA and international agencies approved its use in 2012.^41,48,49

SP-304, another GUCY2C ligand in clinical development, is an analog of the endogenous GUCY2C ligand uroguanylin.⁵⁰ Like linaclotide, SP-304 was designed to exert its effect in the intestine with minimal systemic exposure. In early phase IIa clinical trials, patients with chronic constipation exhibited a marked improvement in bowel habits compared with placebo.⁵¹ Yet a third ligand in development, SP-333, is structurally similar to SP-304 but possesses several amino acid substitutions which render it resistant to proteolysis. This resistance is predicted to increase the dose of active drug passing through the small intestine and into the colon, thus giving it potentially greater utility in treating colonic disorders.^52,53 SP-333 is therefore also currently in clinical trials for the treatment of inflammatory bowel disease (IBD), a disease with a typically large colonic component.^53,54

Given the established role for the loss of GUCY2C signaling in colorectal tumorigenesis, linaclotide and other GUCY2C ligands may provide a unique opportunity for cancer chemoprevention (Fig. 1).^{34,35,37,55-58} In that context, linaclotide is FDA-approved and possesses a favorable toxicity profile to support its use in healthy patients for disease prevention.

Figure 1. Preventing colorectal cancer by exploiting the paracrine hormone hypothesis of colorectal cancer. Colorectal cancer is a stepwise disease whereby a normal epithelium is transformed over a period of years into first an adenoma until finally becoming an adenocarcinoma. Importantly, guanylin expression is lost early and universally in this transformative continuum and contributes to disease progression as depicted in the diagram. Insets to this cartoon show adapted immunohistochemistry images (guanylin depicted in purple, nuclei/hematoxylin depicted in black) of normal human colonic epithelium as contrasted with human adenocarcinoma which demonstrate complete loss of guanylin staining in the adenocarcinoma (inset images adapted from Wilson etal. Biomarkers & Prevention, 2014). Importantly, GUCY2C ligand loss in colorectal cancer provides a therapeutic opportunity for chemoprevention through exogenous administration of GUCY2C ligands.

Given all of these factors, we have partnered with the Mayo Clinic and the National Cancer Institute (NCI)'s Division of Chemoprevention to undertake a phase 1 study to evaluate linaclotide as a chemopreventative agent for colorectal cancer (Linaclotide Acetate in Preventing Colorectal Cancer in Healthy Volunteers NCT01950403). The goal of this study is to evaluate the dose of linaclotide required to induce GUCY2C and increase cGMP levels in the colorectum of healthy volunteers. Secondary endpoints include measures of adverse events and toxicity as well as the activation of downstream tumor suppressive GUCY2C signaling. Tissue analysis will be conducted in biopsies collected via rectosigmoidoscopy, once at baseline and then again following 7 days of linaclotide treatment. Similar to the clinical path of NSAIDs/aspirin to chemoprevention, encouraging results from these studies must be confirmed in larger phase 2 and 3 trials that quantify actual tumor prevention over longer periods of follow-up.⁶

Interestingly, evidence also suggests that GUCY2C activation may have a role in preventing multiple forms of cancer outside the intestine. Mechanistically, this prevention is achieved through GUCY2C's role in regulating intestinal barrier function. Indeed, multiple studies have shown that silencing GUCY2C induces barrier deficiency, while its activation restores barrier integrity.^37,57 Thus, conditions which silence GUCY2C also are likely to result in a leaky barrier which allows environmental toxins and carcinogens to leak into the systemic circulation, promoting inflammation and DNA damage in other organs outside the intestine.³⁷ Further studies are necessary, but these findings represent an intriguing avenue for multi-focal chemoprevention with a well-tolerated, intestine-specific approach.

Identification of populations for GUCY2C-targeted prevention

One major challenge to developing chemoprevention strategies lies in identifying suitable at-risk populations for treatment. In addition to known genetic CRC syndromes (FAP and HNPCC), there are established environmental risk factors, including obesity (body weight, diet and physical activity level), advanced age, intestinal inflammation, cigarette smoking and alcohol abuse.⁵⁹ Beyond these established epidemiological factors, further work must be done to identify populations who would likely accrue the most benefit from linaclotide treatment. In that context, recent studies have raised the possibility that guanylin loss is an important molecular event in colorectal cancer associated with inflammatory bowel disease and obesity, both well-known clinical risk factors for CRC.^36,60,61

The paradigm of cancer prevention is expanding into molecular pathologic epidemiology (MPE). Whereas traditional epidemiologic studies attempt to link risk factors to disease (i.e. a clinical association), MPE adds additional depth by considering the molecular underpinnings of the disease itself.^62,63 This approach applies these epidemiologic findings to achieve more precise patient management. Importantly, loss of guanylin silencing GUCY2C in colorectal cancer is universal, suggesting that this event is involved in the pathogenesis of multiple molecular subtypes of CRC.²⁶ Despite this prevalence, additional studies are required to identify which specific molecular subtypes of CRC are most affected by GUCY2C silencing, thereby identifying which at-risk patients to target.

Conclusions and future directions

There has been a gradual shift in the balance of cancer research in recent years from a more treatment-centric approach to prevention. This change in strategy may perhaps reflect an improvement in our understanding of causative mechanisms of disease, or alternatively be attributed to the lack of improvement in clinical outcomes for most cancers in recent decades, despite billions of dollars in research spending.⁶⁴ This failure in approach is perhaps best exemplified by small molecule inhibitors for cancer treatment, which generally speaking have yielded only months of survival benefit despite costing billions of dollars to develop.^48,49,65 Although research into curative interventions is still essential, this shift in focus toward chemoprevention has the potential to provide substantial clinical benefit to at-risk patient populations (summarized in Fig. 1).

Currently, there is no FDA-approved, generally accepted strategy for such chemoprevention of CRC in the general population. However, recent advances in several preventative strategies, including linaclotide, have made the chemoprevention of CRC within striking distance of becoming a reality. Remarkably, activation of linaclotide's receptor, GUCY2C, eliminates tumorigenesis in mouse models of CRC, and these findings are currently being evaluated in human studies. Future studies should evaluate the efficacy of GUCY2C-directed chemoprevention in patients with known risk factors. In addition to chemoprevention of at-risk populations, linaclotide's favorable toxicity profile may even allow its use in the general population, a proposition which has previously been controversial in aspirin chemoprevention due to toxicity concerns.

Disclosure of potential conflicts of interest

SAW is the Chair (uncompensated) of the Scientific Advisory Board of Targeted Diagnostics & Therapeutics, Inc. which provided research funding that, in part, supported this work and has a license to commercialize inventions related to this work.

Funding

Supported by grants from NIH [CA170533] and Targeted Diagnostic and Therapeutics, Inc. ESB received an F30 Ruth Kirschstein MD-PhD Fellowship Award [CA180500]. SAW is the Samuel MV Hamilton Professor of Thomas Jefferson University.