The clinical potential of etrolizumab in ulcerative colitis: hypes and hopes

Abstract

Antiadhesion molecules are effective and safe in patients with ulcerative colitis (UC). Etrolizumab, a monoclonal antibody targeting both α₄β₇ and α_Eβ₇, represents a promising therapy for patients with UC, since this novel mechanism of action may be effective in blocking leukocyte recruitment both at the vascular and at the mucosal level. Preliminary studies show that etrolizumab is effective in inducing clinical response and remission, and mucosal healing. Moreover, new predictors of response have recently been identified, opening the way to a tailored therapeutic approach. This review of the literature aims to present and discuss the most recent evidence on etrolizumab in UC, focusing on the clinical implications of the use of etrolizumab in UC.

Introduction

Ulcerative colitis (UC) is a chronic immune-mediated inflammatory bowel disease (IBD) that affects the colonic mucosa from rectum up to the caecum [Danese and Fiocchi, 2011]. A complex interaction between genetic background, intestinal microbiota, environmental factors and immune response pathways plays a key role in triggering and maintaining chronic inflammation in the colon, although the causative factors still remain unknown [Danese et al. 2014].

Currently, the therapeutic armamentarium to treat patients with UC remains poor. Steroids, 5-aminosalycilic acid compounds and immunomodulators, such as thiopurines or calcineurin inhibitors, represent a nonselective therapy for mild and moderate to severe disease [Dignass et al. 2012]. In the last decade, more specific therapies directed against selective targets, such as antitumor necrosis factor α (anti-TNF) monoclonal antibodies have been developed [Rutgeerts et al. 2005], although the majority of patients will not respond or will lose response during the maintenance phase [Gisbert and Panes, 2009; Danese et al. 2011], resulting in almost 30% of patients requiring colectomy over time [Langholz et al. 1992]. Moreover, despite their selectivity in targeting TNFα, anti-TNFs have systemic effects which can result in potential serious adverse events (AEs) [Sousa and Allez, 2015]. Recent data on effective therapeutic strategies, such as a combination of anti-TNF and immunosuppressants [Colombel et al. 2010; Panaccione et al. 2014], showed increased response rates and decreased immunogenicity. However, new therapeutic targets are still needed in UC to increase the chance of response and remission by lowering the risks of AEs to a minimum.

Antiadhesion molecules in UC

Integrins represent a family of α,β heterodimeric transmembrane receptors, which includes a combination of at least 24 different pairings of 18 α subunits and 8 β subunits [Thomas and Baumgart, 2012]. The α subunit determines the specificity of the integrin ligand, the β subunit is connected to the cytoskeleton and affects multiple signaling pathways [Barczyk et al. 2010; Thomas and Baumgart, 2012].

Leukocyte recruitment into the gut mucosa from the blood stream plays a key role in activating and maintaining chronic inflammation in IBD. Leukocytes roll across the endothelium, express integrins that firmly adhere to their specific mucosal and vascular ligands, called adhesion molecules (mucosal addressin cell adhesion molecule (MadCAM) and vascular cell adhesion molecule (V-CAM)), and then migrate into the inflamed tissue [Danese et al. 2005]. This multistep mechanism is mediated by selectins and chemokines that facilitate the recruitment and the crossing from the blood stream to the gut [Fiorino et al. 2010; Thomas and Baumgart, 2012; Cesarini and Fiorino, 2013; Danese and Panes, 2014].

The blockade of adhesion molecules has been shown to be effective in UC [Feagan et al. 2013], as well as in Crohn’s disease (CD) [Sandborn et al. 2005; Targan et al. 2007]. Natalizumab blocks α₄ integrins, whereas vedolizumab selectively blocks α₄β₇. The ENACT-1 and -2 trials showed that natalizumab was not superior to placebo at week 10 in patients with moderate to severe CD, but continuing natalizumab resulted in higher rates of sustained response (61% versus 28%, p < 0.001) and remission (44% versus 26%, p = 0.003) through week 36. Similar results were found in the GEMINI 2 trial, in which a total number of 1115 subjects with CD where randomized to receive vedolizumab or placebo. At week 6, no significant differences were found between the two groups. However, among patients who had a response and continued in the maintenance phase, 39.0% and 36.4% of those assigned to vedolizumab every 8 weeks and every 4 weeks respectively were in clinical remission at week 52 compared with 21.6% who received placebo (p < 0.001 and p = 0.004) [Sandborn et al. 2013]. In UC, vedolizumab was significantly more effective in inducing clinical response than placebo at week 6 (47.1% and 25.5%, p < 0.001). At week 52, clinical remission resulted in 41.8% of patients who continued to receive vedolizumab every 8 weeks and 44.8% of patients who continued to receive vedolizumab every 4 weeks compared with 15.9% of patients receiving placebo (p < 0.001) [Feagan et al. 2013].

The nonselective blockade of integrins by natalizumab resulted in a significantly worse safety profile than vedolizumab, since fatal progressive multifocal leukoencephalitis (PML) can occur in patients treated with natalizumab. The estimated incidence of PML in this subgroup of patients is 11.1/1000 (or 1 in 90) patients [McGuigan et al. 2015], while no cases of PML have been reported to date in patients exposed to vedolizumab.

New molecules targeting integrins or adhesion molecules are currently in development in IBD with promising results [Bravatà et al. 2015].

Etrolizumab

Etrolizumab represents the next generation of antiadhesion molecules [Lin and Mahadevan, 2014]. It is a humanized monoclonal antibody that selectively binds the β7 subunit of both the α₄β₇ and α_Eβ₇ integrin heterodimers. This mechanism results in two different therapeutic approach, since it antagonizes the egress of lymphocytes by blocking the interaction between α4β7 and MAdCAM-1 at the vascular level, and also blocks the interaction between α_Eβ₇ and E-cadherin, potentially avoiding the retention of α_Eβ₇ ⁺ cells in the intraepithelial compartment [Vermeire et al. 2014]. The results from phase II trials show that etrolizumab can be effective in inducing and maintaining remission in UC.

Efficacy, safety and pharmacological profile of etrolizumab

A phase I and II trial have been conducted on etrolizumab in UC [Rutgeerts et al. 2013; Vermeire et al. 2014].

The phase I trial

The phase I study [Rutgeerts et al. 2013] was a randomized, placebo-controlled, double-blind within-cohort trial. Subjects with active moderate to severe UC, defined as a Mayo Clinic Score (MCS) at least 5 were enrolled. Etrolizumab was administered either by intravenous drip or by subcutaneous injection. Safety of etrolizumab was the primary objective; secondary objectives were pharmacokinetics (PK) characterization and immunogenicity evaluation. Furthermore, the protocol included additional exploratory objectives: evaluation of pharmacodynamics (PD) and potential clinical efficacy of etrolizumab.

Study design

The trial was composed of two consecutive stages: the single ascending dose (SAD) and the multiple dose (MD) stage. The MD phase started only after completion of SAD.

Patients were sequentially enrolled into the SAD stage, into five ascending dose cohorts, by a 4:1 randomization process. Only after completing the 15-day follow up for all the patients receiving any study dose was switching to the next dose planned. Etrolizumab was administered as a single dose (0.3 mg/kg intravenous, 1.0 mg/kg intravenous, 3.0 mg/kg intravenous, 10.0 mg/kg intravenous and 3.0 mg/kg subcutaneous) and patients were evaluated for 14 days following drug administration. Globally, 25 patients were enrolled (five in each cohort) and each patient was monitored over 24 h after drug administration.

Subsequently, as the last patient in the SAD stage had completed the follow-up period of 10 weeks after the completion of the SAD stage (highest dose), and if the safety analysis and PK profile were acceptable, the MD stage was opened. A total of 23 new patients were concurrently enrolled into four cohorts and randomized by a 4:1 ratio to receive placebo or a single dose of etrolizumab at different dosages (0.5 mg/kg subcutaneous, 1.5 mg/kg subcutaneous, 3.0 mg/kg subcutaneous, 4.0 mg/kg intravenous). Patients received three drug or placebo administrations at 4-week intervals. Globally, five patients received placebo, 1.5 g/kg and 4.0 mg/kg etrolizumab; four patients received the 0.5 and 3.0 mg/kg doses. A safety follow up was performed for 20 weeks after the final dose.

Safety evaluation was based on the National Cancer Institute Common Toxicity Criteria Adverse Event (NCI CTCAE), version 3.0. In the SAD stage, dose-limiting toxicity (DLT) was defined as any CTCAE grade of at least 3 occurring within 14 days of dosing, including UC exacerbations and grade 3 infusion reactions, whereas the maximum tolerated dose (MTD) was the dose level below the one causing a DLT in at least two patients. In the case of no more than one patient experiencing a DLT in each dose, the MTD was considered the highest administered dose (10.0 mg/kg). The development of PML symptoms and detectable serum John Cunningham Virus (JCV) DNA were also considered and neurological evaluations had to be performed for patients developing detectable JCV DNA or symptoms suggestive of PML. After completing the 20-week safety follow-up period, all the patients were further monitored for 2 years after the first investigational product administration to check for PML symptom development and white blood cell count abnormalities.

Safety

Globally, 15 patients prematurely discontinued the study. In the SAD phase, seven patients discontinued, five were treated with etrolizumab and two with placebo. In the MD stage, eight patients did not complete the study, seven received etrolizumab and one placebo. Etrolizumab was well tolerated: AEs were reported in 95% (36/38) and 100% (10/10) of etrolizumab-treated and placebo-treated patients respectively. No dose relationship with AE incidence was detected. Exacerbation of UC and headache were the most commonly reported AEs, in particular UC exacerbation was reported in 42% and 80%, and headache in 32% and 20% of patients treated with etrolizumab and placebo respectively. Serious AEs were reported in 18% (7/38) and 10% (1/10) of subjects receiving etrolizumab and placebo respectively, mostly UC exacerbations. Two of the patients undergoing colectomy for UC exacerbation experienced impaired wound healing. Both received a single dose of etrolizumab (1 mg/kg intravenous and 3.0 mg/kg subcutaneous), but a correlation between drug exposure and AE was hard to assess because of the presence of concomitant confounding risk factors (e.g. type 2 diabetes mellitus, obesity and hypoalbuminemia). In addition, two patients developed transient increased JCV DNA levels, without symptoms of PML or magnetic resonance imaging abnormalities.

Pharmacological characteristics of etrolizumab

Etrolizumab PK profile was linear and dose proportional. A slight trend of nonlinearity, often seen with immunoglobulin G, was observed at the lowest doses (<1.0 mg/kg) and could be explained by target-mediated clearance at lower doses. As a consequence, lower doses are characterized by a shorter mean t1/2 compared with higher ones (4.8–7.4 days at 0.3–1.0 mg/kg doses versus 10.6–13.2 days at 3–10 mg/kg doses), even if this finding was not associated with differences in area under the curve (AUC) or Cmax (exposure). Bioavailability of a single subcutaneous dose was approximately 67% at 3 mg/kg dose and the total systemic clearance (CLss/F) was 5.1–10.1 ml/day/kg after multiple doses. A moderate drug accumulation over time was associated with the repeated monthly dose (accumulation ratio for AUC at steady state was 1.2 fold for intravenous and 2.0 fold for subcutaneous). PK parameters support monthly subcutaneous administration.

Only two patients developed antitherapeutic antibodies (ATAs,) one in the lowest dose-group of the SAD stage population and one in the lower dose range in the MD stage. Antibody levels were near to the minimum reportable ATA level in both cases and they did not affect the PK profile.

Drug occupancy, which can be defined as the capability of a drug to bind the specific molecular target, on CD4+ lymphocytes following single or multiple drug administration was used to study the PD of etrolizumab. Etrolizumab occupied β₇ integrins on CD4 lymphocytes and there was a dose-dependent trend in duration of occupancy. Mean occupancy time was 2, 6 and 10 weeks after single intravenous doses of 0.3 mg/kg, 1.0 mg/kg and 10.0 mg/kg respectively without significant differences between the 3.0 mg/kg intravenous and the 3.0 mg/kg subcutaneous dose (median occupancy time 10 weeks).

After multiple doses, mean occupancy was about 2–4 weeks for the lowest doses (0.5 and 1.5 mg/kg subcutaneous) and 8–12 weeks for the higher doses (3.0 mg/kg subcutaneous and 4.0 mg/kg intravenous). An increase (from 10 to 12 weeks) was seen when comparing a single 3.0 mg/kg subcutaneous dose with three repeated 3.0 mg/kg subcutaneous administrations.

PK and PD were then correlated to determine the minimal etrolizumab serum concentration that led to the full occupation of the β₇ integrin. Data show that drug serum concentration decreases when β₇ receptors returned unoccupied.

Clinical efficacy

Preliminary data about clinical activity were obtained by measuring the MCS both in the SAD and the MD stage. Clinical response was defined as a decrease in MCS of at least 3 points and 30% from baseline, together with a decrease in the bleeding subscore of at least 1, or with a bleeding absolute subscore of 0–1; clinical remission was defined as a MCS up to 2, with no individual subscore exceeding 1.

A trend for a decrease in MCS was observed in the low-dose treated groups of SAD stage, especially when the drug was administered subcutaneously. A similar trend was also seen in the MD stage, with a MCS decrease in the lowest range doses. Nevertheless, this trend was similar in placebo-treated patients. This could be explained by some differences in patient characteristics between placebo- and etrolizumab-treated patients: there was a greater proportion of patients with concomitant steroid therapy in the placebo group, whereas a higher proportion of patients previously exposed to anti-TNF, who can be considered as refractory subjects, were allocated to receive etrolizumab.

The phase II trial

The phase II trial [Vermeire et al. 2014] was a randomized, placebo-controlled study to evaluate etrolizumab as an induction treatment for patients affected by moderate to severe UC. Inclusion criteria were MCS at least 5 (or 6 in the USA), with an endoscopic subscore of at least 2 and a rectal bleeding subscore of at least 1.

Study design

Patients were randomly assigned (1:1:1) to receive subcutaneous administration of placebo or etrolizumab at low dose (100 mg at weeks 0, 4, 8 with a placebo injection at week 2) or etrolizumab at high dose (300 mg at weeks 2, 4 and 8) with a loading dose of 450 mg administered at week 0. Patients were stratified according to concomitant therapy with corticosteroids, immunomodulators, previous exposure to anti-TNF and study site.

The primary endpoint was clinical remission, defined as MCS up to 2 with no subscore greater than 1 at week 10.

Secondary endpoints were: clinical remission at week 6; clinical response (defined as a 3-point decrease and a 30% reduction in MCS and a decrease of at least 1 point in bleeding subscore or a 0–1 absolute bleeding subscore) determined at two time points: week 6 and 10; and achievement of both endoscopic subscore and rectal bleeding subscore of 0 at two time points: weeks 6 and 10.

Exploratory outcomes included changes from baseline in mucosal healing (defined as endoscopic subscore of 0 or 1), histological active disease severity score and PD evaluation, defined as β₇ occupancy at lymphocyte and colonic tissue levels.

A total of 124 patients were enrolled, but five were then excluded from analysis because of low endoscopic subscore at baseline, as determined by the central reader. Thirty-nine patients received etrolizumab at the lowest dose regimen, 39 etrolizumab at the highest regimen and 41 placebo.

Clinical efficacy

At week 10, a significantly higher proportion of patients receiving etrolizumab 100 mg [21% (95% confidence interval 7–39), p = 0.004] and etrolizumab 300 mg with loading dose [10% (0.2–2.4), p = 0.048] achieved clinical remission compared with those receiving placebo.

Differences in clinical remission at week 6 and in clinical response at week 6 and 10 did not reach statistical significance towards placebo. Five percent of placebo-treated patients were in remission 6 weeks after the first dose compared with 10% in the etrolizumab 100 mg group (p = 0.66) and 8% in the etrolizumab 300 mg group (p = 0.97). Clinical response at week 6 was achieved by 34% of patients treated with placebo compared with 39% (p = 0.27) and 38% (p = 0.68) of patients treated with etrolizumab 100 and 300 mg respectively. Clinical response at week 10 was observed in 29%, 33% (p = 0.83) and 31% (p = 0.90) of patients receiving placebo, etrolizumab 100 mg and etrolizumab 300 mg respectively.

At week 6, 2%, 8% (p = 0.96) and 3% (p = 0.59) of patients treated with placebo, etrolizumab 100 mg and 300 mg respectively had absolute values of endoscopic and rectal bleeding subscores equal to 0. At week 10, none of the placebo-treated patients, 10% (p = 0.16) and 8% (p = 0.19) of patients receiving etrolizumab 100 mg and 300 mg doses had both subscores equal to 0. This is a combined score and, since only 2% reached this goal, this value at week 10 is justified by the relapsing remitting behavior of the disease.

Looking at possible predictors for response and remission, patients taking steroids, not taking immunomodulators and anti-TNF naïve were more likely to achieve clinical remission at week 10.

Pharmacological characteristics and immunogenicity

Anti-etrolizumab antibodies were detected in 4 of the 81 patients treated with active drug (5%). One patient had detectable antibodies before receiving etrolizumab. However, no association was found between antibody formation and PK parameters.

Etrolizumab reached β₇ occupancy on circulating CD8+ β₇ lymphocytes (p=0.0006 and p < 0.0001 with 100 and 300 mg versus placebo), CD4+ β₇ T lymphocytes (p = 0.071 and p = 0.0008 with 100 and 300 mg versus placebo) and CD19+ β₇ B lymphocytes (p < 0.0001 for 100 and 300 mg versus placebo). Furthermore, mucosal β₇ occupancy was significantly higher with etrolizumab versus placebo both at week 6 (p = 0.0043 and p = 0.0021 with 100 and 300 mg) and at week 10 (p = 0.0081 and p = 0.0087). All these data confirm the preliminary results from the phase 1 trial.

Safety of etrolizumab

Finally, etrolizumab showed a good safety profile. AEs were reported in 61% (etrolizumab 100 mg), 48% (etrolizumab 300 mg) and 72% (placebo) of treated patients, whereas serious AEs were reported in 12% (both in etrolizumab and placebo groups) and 5% (etrolizumab 300 mg group). Although the frequency of different AEs was similar in the three study groups, higher rates of influenza-like illness (7% versus 0% and 2%), arthralgia (15% versus 5% and 9%) and rashes (7% versus 3% and 2%) were observed in patients receiving etrolizumab 100 mg compared with those receiving 300 mg or placebo; however, the severity of all these events was judged as mild or moderate.

Further clinical studies on etrolizumab in UC

To date, six phase III trials and one phase II trial are ongoing in moderate to severe UC, both on anti-TNF naïve and on anti-TNF exposed patients [ClinicalTrials.gov identifiers: NCT01461317; NCT02100696; NCT02118584; NCT02136069; NCT02163759; NCT02165215; NCT02171429; NCT02171429]. All published or ongoing trials are summarized in Table 1.

Table 1.

Clinical trials on etrolizumab in ulcerative colitis (UC).

Reference	Phase	Design	Study population	Objectives	Active drug	Comparator	Main results
Rutgeerts et al. [2013]	I	Multicenter, randomized, double-blind within cohort trial Stage I: single ascending dose (SAD) study Stage II: multiple dose (MD) study	Moderate to severe UC (MCS ⩾ 5) Disease duration ⩾ 12 weeks	Primary objective Safety and tolerability Secondary objectives Pharmacokinetics Initial immunogenicity assessment Exploratory objectives Pharmacodynamics Clinical activity	SAD 0.3 mg/kg IV 1.0 mg/kg IV 3.0 mg/kg IV 10.0 mg/kg IV 3.0 mg/kg SC MD 0.5 mg/kg SC 1.5 mg/kg SC 3.0 mg/kg SC 4.0 mg/kg IV	Placebo	No difference in adverse events incidence between etrolizumab and placebo Linear PK SC bioavailability 67% Low immunogenicity Good capability of binding β7 integrins Trend (not significant) for a decrease in MCS
Vermeire et al. [2014]	II	Multicenter, randomized, double-blind study	Moderate to severe active UC	Primary objective Clinical remission at week 10 Secondary objective Clinical remission at week 6 Clinical response at week 6 and 10 Endoscopic subscore and rectal bleeding subscore of 0 at weeks 6 and 10 Exploratory objectives Changes in mucosal healing Histological active disease severity score PD evaluation (β7 occupancy at lymphocyte and colonic tissue levels)	Etrolizumab 100 mg SC at weeks 0, 4 and 8, placebo SC at week 2 Etrolizumab 420 mg SC at week 0, 300 mg SC at weeks 2, 4 and 8	Placebo	Clinical remission at week 10: 21% (p = 0.004) etrolizumab 100 mg versus placebo; 10% (p = 0.048) etrolizumab 420 + 300 mg versus placebo No statistically significant difference for secondary objectives Good safety profile
ClinicalTrials.gov identifier: NCT02136069	III	Multicenter, randomized, double-blind, double-dummy, parallel group study	Moderate to severe UC Anti-TNF naïve N = 720	Primary Sustained clinical remission (remission sustained at each of week 10, 34 and 54 assessments) Secondary Remission at week 10 and 54 Clinical response at week 10 Sustained clinical response (clinical response at week 10, 30 and 54)	Etrolizumab 105 mg SC every 4 weeks until week 54	Infliximab	Ongoing
ClinicalTrials.gov identifier: NCT02163759 NCT02171429	III	Multicenter, randomized, double-blind, double-dummy study	Moderate to severe UC Anti-TNF naïve N = 350	Primary Induction of remission compared with placebo at week 10 Secondary Induction of remission compared with adalimumab	Etrolizumab 105 mg SC every 4 weeks	Adalimumab Placebo	Ongoing
ClinicalTrials.gov identifier: NCT02100696	III	Multicenter, double-blind, placebo-controlled study	Moderate to severe UC Anti-TNF exposed N = 800	Primary Remission at week 14 Remission maintenance among patients in remission at week 14 Secondary Remission at week 14 and 66	Etrolizumab 105 mg SC every 4 weeks	Placebo	Ongoing
ClinicalTrials.gov identifier: NCT02118584	III	Part I: open-label extension study Part II: safety monitoring	Part I Patients previously enrolled in phase III controlled studies Part II Patients not eligible or not willing to continue with drug administration N = 1850	Primary objectives Long-term efficacy determined with MCS (up to 7 years) Incidence of adverse events (up to 7 years)	Etrolizumab 105 mg SC every 4 weeks	NA	Ongoing
ClinicalTrials.gov identifier: NCT02165215	III	Multicenter, double-blind, placebo-controlled study	Moderate to severe UC Anti-TNF naïve N = 350	Primary objective Remission at week 62 Secondary objectives Maintenance of remission at week 62 Clinical remission at week 62 Clinical response at week 62 Improvement in endoscopic appearance of the mucosa at week 62 Endoscopic remission at week 62 Steroid-free remission at week 62 Histologic remission at week 62 Quality of life	Etrolizumab 105 mg SC every 4 weeks	Placebo (only in the maintenance phase)	Ongoing
ClinicalTrials.gov identifier: NCT02118584	II	Open-label, extension study	Patients enrolled in phase II controlled study Nonresponder at week 10 Responder at week 10 and subsequent flare N = 116	Primary objective Adverse events and serious adverse events Secondary objectives Changes in vital signs and laboratory parameters Discontinuation due to adverse event Injection site reactions and hypersensitivity Infectious complications Immunogenicity	Repeated SC injections (dose not reported)	NA	Active, not recruiting

Remission: MCS up to 2 with individual subscores up to 1 and a rectal bleeding subscore of 0. Clinical response: MCS decrease of at least 3 points and a 30% reduction from baseline, with a decrease of at least 1 point in rectal bleeding subscore or an absolute rectal bleeding subscore of 0 or 1. Mucosal healing: endoscopic subscore of 0 or 1.

MCS, Mayo Clinic Score; IV, intravenous; NA, not applicable; PK, pharmacokinetics; SC, subcutaneous; TNF, tumor necrosis factor.

The potential role of etrolizumab in UC

The development of antiadhesion molecules in UC will dramatically revolutionize the management of UC in the near future. New drugs targeting integrins and adhesion molecules are currently in development and they will offer more therapeutic chances to patients with UC [Danese et al. 2014]. Based on the successful results showed by vedolizumab [Feagan et al. 2013], this mechanism of action will be a valid alternative to current therapies, especially in patients who are not eligible or do not respond to anti-TNFs, as well as the first choice in patients who are candidates for a biological treatment. However, etrolizumab cannot be considered just as an alternative to vedolizumab since it has a different mechanism of action, which potentially adds more value in the control of inflammation at the intestinal level (Figure 1). In fact, it shares with vedolizumab the blockade of α4β7 integrin, which is highly gut selective and warrants a good safety profile [Danese et al. 2014]. Moreover, the concomitant blockade of α_Eβ₇-E-cadherin interaction avoids the adhesion of intraepithelial T cells to the epithelial cells [Cepek et al. 1993; Karecla et al. 1995; Higgins et al. 1998; Stefanich et al. 2011]. Because only 1–2% of circulating lymphocytes express α_Eβ₇ [Parker et al. 1992; Stefanich et al. 2011], whereas it is present in over 90% of intraepithelial lymphocytes, as well as in intestinal dendritic cells [Parker et al. 1992; Cepek et al. 1993], etrolizumab can play a peculiar key role in blocking immunological pathways that trigger and maintain chronic inflammation directly at the mucosal level, with no significant systemic effects, as confirmed by Stefanich and colleagues in cynomolgus monkeys [Stefanich et al. 2011].

Figure 1.

Mechanism of action of etrolizumab. Like the other antiadhesion molecules, etrolizumab blocks α₄β₇ integrin at the endothelial level, but it has a peculiar action as a blocker of α_E integrin interaction with epithelial cells, thus also controlling inflammation at the mucosal level.

Data from clinical trials seem promising in UC [Rutgeerts et al. 2013; Vermeire et al. 2014]. At week 10, etrolizumab was more effective than placebo in inducing clinical remission and endoscopic healing. Moreover, data from the phase II trial showed a decreasing trend in the proportion of aE+ cells in the intestinal crypt epithelium in patients receiving etrolizumab compared with those receiving placebo, which was higher in patients who achieved clinical remission at week 10, and concomitant to a significantly increased expression of E-cadherin. At the same time, no apparent decrease in aE+ cells in the lamina propria was observed [Vermeire et al. 2014], suggesting a highly selective action at the mucosal level, which sounds crucial in UC. These data need further confirmation from the ongoing studies.

Targeting therapy with etrolizumab

In the last decade, the only biological agents available for patients with UC, who are refractory to conventional therapies, have been anti-TNFs. Thus, antiadhesion molecules represent a revolution in the management of UC. Like in other immune-mediated diseases, patients eligible for monoclonal antibodies will now benefit from different molecules with different mechanisms of action. It is therefore crucial to identify the right patient for the right therapy. Tailoring therapies based on clear predictors of response can dramatically reduce the risk of giving patients ineffective therapies, reducing the risk of AEs and also reducing unnecessary costs.

Tew and colleagues performed a retrospective analysis of 110 patients with UC treated with etrolizumab in a phase II randomized controlled trial plus 21 additional patients with UC or without any IBD enrolled in a prospective observational cohort study as a control group [Tew et al. 2015]. They found that patients who responded to etrolizumab had higher expression of genes associated with colon tissues collected at baseline from patients who had a clinical response to etrolizumab and expressed higher levels of T-cell-associated genes than patients who did not respond (p < 0.05). In particular, they found that colonic CD4 integrin aE⁺ lymphocytes expressed higher levels of genes coding for granzyme A (GZMA) than CD4 aE^– cells (p < 0.0001) and higher levels of GZMA messenger RNA in UC than in controls (p < 0.05). A higher proportion of patients with high levels of GZMA mRNA (41%) or α_E integrin (ITGAE) mRNA (38%) than those with low levels of GZMA (6%) or ITGAE mRNA (13%) achieved clinical remission (p < 0.05) and mucosal healing (41% GZMA versus 19% GZMA^low and 44% ITGAE^high versus 19% ITGAE^low) among patients with UC treated with 100 mg of etrolizumab. Compared with ITGAE^low and GZMA^low patients, patients with ITGAE^high and GZMA^high had higher baseline numbers of epithelial crypt-associated integrin aE^high cells (p < 0.01 for both), but a smaller number of crypt-associated integrin aE⁺ cells after receiving etrolizumab (p <0.05 for both). After 10 weeks of etrolizumab treatment, patients expressing high levels of GZMA at baseline showed a 40–80% decrease in expression of genes associated with T-cell activation and genes encoding inflammatory cytokines (p < 0.05) since baseline. These data suggest that a preliminary analysis of gene expression for granzyme A and ITGAE may identify patients who are more likely to respond to etrolizumab by avoiding ineffective treatment in patients with a low chance of success.

The choice of etrolizumab as a first- or second-line treatment in patients with UC needing biological treatment requires more solid evidence from further clinical trials. Vermeire and colleagues found higher rates of response in patients receiving 100 mg etrolizumab who were treated with concomitant corticosteroids and in those who were naïve to anti-TNFs. However, the number of patients included in this subgroup analysis is too low to give any final recommendation for using etrolizumab in UC. Currently, no direct comparison between biologics is available. Indirect comparison from randomized clinical trials may suggest anti-integrins can be effective as a first-line biological therapy in UC [Danese et al. 2014], as well as anti-TNFs, but with a potential better safety profile [Danese et al. 2014]. Two studies comparing etrolizumab with adalimumab [ClinicalTrials.gov identifier: NCT02171429] and infliximab [ClinicalTrials.gov identifier: NCT02136069] in patients with UC naïve to biologics are currently ongoing.

Conclusion

In conclusion, etrolizumab could be a valid therapeutic option in IBD. The mechanism of action warrants gut selectivity and additional mucosal selectivity in UC, together with the good safety profile of anti-integrins. Further clinical studies are expected to confirm the currently available data on efficacy and safety, and will hopefully provide stronger evidence about the right patient who could benefit from etrolizumab.