Efficacy and Safety of Etrasimod in Patients with Moderately to Severely Active Isolated Proctitis: Results From the Phase 3 ELEVATE UC Clinical Programme

Abstract

Background and Aims

Pivotal trials in ulcerative colitis have historically excluded patients with isolated proctitis. Etrasimod is an oral, once-daily, selective sphingosine 1-phosphate_1,4,5 receptor modulator for the treatment of moderately to severely active ulcerative colitis. This post hoc analysis assessed efficacy and safety of etrasimod 2 mg once daily in patients with isolated proctitis (centrally read) from the phase 3 ELEVATE UC 52 and ELEVATE UC 12 trials.

Methods

Patients, including those with isolated proctitis (<10 cm rectal involvement) who met all other inclusion criteria in ELEVATE UC 52 and ELEVATE UC 12, were randomised 2:1 to receive etrasimod or placebo. Primary, secondary and other identified efficacy endpoints and safety were assessed.

Results

We analysed data from 64 and 723 patients at Week 12 (both trials pooled), and 36 and 397 patients at Week 52 (ELEVATE UC 52 only) with isolated proctitis and more extensive colitis (≥10 cm rectal involvement), respectively. Patients with isolated proctitis receiving etrasimod demonstrated significant improvements versus placebo, including clinical remission rates at Weeks 12 (42.9% vs 13.6%) and 52 (44.4% vs 11.1%), endoscopic improvement (52.4% vs 22.7%) at Week 12 and bowel urgency numerical rating scale score at Week 12 (all p < 0.01). Generally similar trends were observed in patients with more extensive colitis. Safety was consistent across subgroups, with no new findings.

Conclusions

Etrasimod demonstrated significant improvements versus placebo in patients with isolated proctitis, and those with more extensive disease, in most efficacy endpoints at Week 12 and 52.

Clinicaltrials.gov: NCT03945188; NCT03996369

Graphical Abstract

Graphical Abstract.

1. Introduction

Ulcerative colitis (UC) is a chronic inflammatory disease with a wide range of presentations.^1,2 Disease extent can be classified as proctitis, left-sided colitis or extensive colitis (pancolitis), according to the colonic segments involved.^1,2 Isolated proctitis, commonly referred to as ulcerative proctitis, is defined as inflammation limited to the rectum.^1,2 Patients with isolated proctitis often present with a diverse array of symptoms including rectal bleeding (RB), urgency, faecal leakage or incontinence and tenesmus, that result in reduced health-related quality of life.^3,4 The presence of diarrhoea is variable, and between 10−50% of patients have delayed transit and constipation.^5,6

Approximately 30% of patients with UC typically present with isolated proctitis^7,8; despite this, pivotal phase 3 clinical trials of systemic therapies in patients with moderately to severely active UC consistently exclude this subgroup of patients.⁹ This is likely due to several factors. Firstly, a belief has been established that patients with isolated proctitis are more difficult to treat,⁹ and therefore are not appropriate to include in controlled trials as they could introduce increased variability and obscure potential treatment effects. However, it is possible that this belief is the result of referral bias, whereby patients with limited disease who are refractory to medical management are more likely to be referred to tertiary care centres than patients with more extensive disease, who are more likely to be offered surgery without referral. Secondly, the trial outcome measures established for patients with more extensive disease may not be optimal for the assessment of efficacy in those with isolated proctitis; patients with isolated proctitis are understood to have a high symptomatic burden, and are more likely to experience stool frequency and tenesmus.^4,10 This is consistent with the observation that patients with UC with more extensive colitis also have distal inflammation, which is believed to be the driver of these symptoms.¹¹

Patients with isolated proctitis are also often considered to have a milder form of UC compared to those with more extensive disease; however, a third of patients with isolated proctitis are refractory to rectal and oral therapy with 5-aminosalycylates (5-ASA) and/or corticosteroids (CSs)¹² and a population-based study in paediatric patients with isolated proctitis demonstrated that this subgroup is at a similar risk of disease extension, colectomy, and anti-TNFi therapy to patients with more extensive colitis.¹³ Strong evidence regarding the experience of those with isolated proctitis is lacking,^9,10 and the existing data do not provide a strong rationale for excluding patients from controlled trials. New treatment options are therefore needed for patients with refractory isolated proctitis.

Etrasimod is an oral, once-daily, selective sphingosine 1-phosphate (S1P)_1,4,5 receptor modulator for the treatment of moderately to severely active UC. In the phase 3 induction and maintenance study with “treat-through” design, ELEVATE UC 52 (NCT03945188), and induction study ELEVATE UC 12 (NCT03996369), enrolled patients had a modified Mayo score (MMS) of 4–9 with a centrally read endoscopic subscore (ES) ≥2 and rectal bleeding subscore (RBS) ≥1, and a documented history of inadequate response, loss of response, or intolerance to at least one UC therapy. All efficacy endpoints were met in both trials.¹⁴ Notably, for the first time in a pivotal phase 3 trial of a systemic UC treatment, the subpopulation of patients with isolated proctitis could be enrolled in the ELEVATE UC clinical programme, provided the patients met all other eligibility criteria.¹⁴ This post hoc analysis assessed predefined efficacy and safety outcomes in the subgroup of patients with isolated proctitis.

2. Materials and methods

2.1. Patients and study design

Efficacy and safety data from patients who participated in the completed phase 3 ELEVATE UC 52 and ELEVATE UC 12 clinical trials were analysed. Full study design details have been previously published.¹⁴

Eligible patients (aged 16–80 years) with moderately to severely active UC (MMS 4–9; centrally read ES ≥2 and RBS ≥1) were randomised 2:1 to etrasimod 2 mg QD or placebo. Patients with isolated proctitis at baseline (<10 cm rectal involvement) were also enrolled in the clinical trials, provided they met all other eligibility criteria. This subpopulation was capped at 15% of the total patients.¹⁴

Background concomitant treatment with oral 5-ASA and/or oral CSs was permitted provided patients were on a stable dose ≥2 weeks immediately prior to randomisation for 5-ASA and 4 weeks immediately prior to the screening endoscopy for CS use. CS tapering was expected to begin after Week 12 in ELEVATE UC 52.¹⁴ Patients were excluded from the study if they received treatment with topical rectal 5-ASA, short-chain fatty acid enemas or rectal CSs ≤2 weeks prior to and during screening. Any rectal therapy (including 5-ASA and CSs) was prohibited during the study, other than that required for endoscopy preparation.

In this post hoc analysis, Week 12 data were pooled from the ELEVATE UC 52 and ELEVATE UC 12 trials into two subgroups: patients with isolated proctitis and those with more extensive colitis (defined as ≥10 cm rectal involvement). Patients included in the isolated proctitis subgroup were identified from centralised endoscopy reading. At screening for ELEVATE UC 52 and ELEVATE UC 12, extent of disease (<10 cm or ≥10 cm rectal involvement) was evaluated by a local endoscopist and the final assessment was determined by a central reader. If the local score did not match with the first central reader’s score, then the extent of disease was confirmed by a second central reader; in cases of disagreement between the two central readers, extent of disease was confirmed by a central adjudicator. Week 52 data for both subgroups were from ELEVATE UC 52 only.

2.2. Efficacy and safety evaluations

Efficacy outcomes in this post hoc analysis were the predefined efficacy endpoints in the ELEVATE UC 52 (at Week 12 and Week 52) and ELEVATE UC 12 (at Week 12) trials (Table 1). Cessation of RB and achievement of symptomatic remission were additionally assessed at every study visit, and the bowel urgency numeric rating scale (NRS) was also assessed at Weeks 12 and 52 (Table 1).

Table 1.

Endpoint definitions.

Endpoint	Definition
Primary
Clinical remission	SFS = 0 (or = 1 with a ≥1-point decrease from baseline), RBS = 0, and endoscopic subscore ES ≤1a.
Key secondary
Endoscopic improvement	ES ≤1a.
Symptomatic remission	SFS = 0 (or = 1 with a ≥1-point decrease from baseline) and an RBS = 0.
Endoscopic improvement-histologic remission (EIHR; mucosal healing)	ES ≤1a with histological remission as measured by a Geboes Index score <2.0.
Secondary
Clinical response	≥2-point and ≥30% decrease from baseline in MMS, and a ≥1-point decrease from baseline in RBS or an absolute RBS ≤1.
CS-free clinical remission	Clinical remission at Week 52 with no CS use for at least the last 12 study weeks immediately before Week 52.
Sustained clinical remission	Clinical remission at both Weeks 12 and 52.
Other endpoints
Cessation of RB	RBS = 0. The RBS is a non-invasive component of the Mayo score, a measure of disease activity. It is ranked on a scale of 0–3, with higher scores indicating more severe disease.15,16
Improvement in bowel urgency NRS score	Bowel urgency NRS score is a validated measure of urgency severity in patients with UC, with the patient’s perception of bowel urgency severity over 24 hours assessed on an 11-point NRS ranging from 0 (“no urgency”) to 10 (“worst possible urgency”).17,18 A 3-point improvement in the bowel urgency NRS score is considered meaningful,17 and most patients surveyed reported a 1- to 2-point change as the minimal desired improvement.18

^aES ≤ 1 excludes friability.

Treatment-emergent adverse events (TEAEs) are reported for patients with isolated proctitis and those with more extensive colitis. Serious AEs (SAEs) were defined as any events that resulted in death; were life-threatening; required patient hospitalisation or prolongation of existing hospitalisation; resulted in a persistent or significant disability or incapacity; resulted in a congenital anomaly or birth defect; or were deemed medically significant. AEs of special interest (AESI) analysed included pulmonary disorders, infections (severe infections [Common Terminology Criteria for Adverse Events version 5.0, Grade ≥3], opportunistic infections, herpes simplex, and herpes zoster), liver injury, malignancies, macular oedema and cardiovascular events. AESIs are described in the Supplementary Methods.

2.3. Statistical analyses

Analyses were based on treatment assignment to the etrasimod 2 mg QD and placebo groups in the full analysis (with baseline MMS 4−9) and safety analysis sets, which included all randomised patients who received ≥1 dose of study treatment in the ELEVATE UC clinical programme.

Analyses of binary endpoints such as the primary and key secondary efficacy endpoints at each visit were performed using the Cochran–Mantel–Haenszel (CMH) method assuming common proportion difference adjusting for three reported stratification factors in randomisation (naïvety to prior biologic/Janus kinase inhibitor therapy, baseline CS use and baseline disease activity [MMS: 4–6 or 7–9]) and, for pooled study data, additionally adjusting for study stratification. Patients with missing responses were considered nonresponders. Adjusted treatment difference (etrasimod 2 mg QD vs placebo) and 95% confidence intervals (CIs) in response rate were generated from the CMH method.

Analyses of repeated change from baseline values in bowel urgency NRS were performed using a mixed-effect model with repeated measures (MMRM) with a covariate for baseline score, the same three reported stratification factors as above, treatment, visit, and treatment-by-visit interactions. An unstructured variance-covariance matrix was used. Analyses of change from baseline values at a single visit were performed using Analysis of Covariance (ANCOVA) with a covariate for baseline score, the same three reported stratification factors and treatment. Missing values are not explicitly imputed in these models. For pooled study data, MMRM and ANCOVA were additionally adjusted for study stratification. Least-squares mean (LSM) change from baseline treatment difference (etrasimod 2 mg QD vs placebo) and 95% CIs by visit were generated from these models. All analyses were performed separately by isolated proctitis and more extensive colitis subgroups. Two-sided nominal p values were reported without adjustment for multiple comparisons.

To assess potential differences in treatment effects (etrasimod vs placebo) between patients with isolated proctitis and patients with more extensive colitis, additional analyses were performed on the binary efficacy endpoints; these include clinical remission, endoscopic improvement, symptomatic remission, endoscopic improvement-histological remission (EIHR), and clinical response at Week 12 (both trials) and Week 52 (ELEVATE UC 52 only), and CS-free clinical remission and sustained clinical remission at Week 52 (ELEVATE UC 52 only). Three different testing approaches were used to evaluate whether the observed differences in treatment effect were supported by data across different modelling assumptions. The first method evaluated the absolute difference in treatment effects estimated by the CMH-adjusted method, adjusting to the stratification factors. The second method evaluated the absolute difference in treatment effects using the crude (unadjusted) estimates of proportion difference without adjusting to the stratification factors. Both the first and second methods assumed asymptotic normal distribution of the absolute difference. The third method employed logistic regression, which modelled each endpoint on the fixed effects of treatment, isolated proctitis subgroup, and treatment-by-isolated-proctitis-subgroup interaction, adjusting to the stratification factors; the interaction effect was assessed using the Wald statistics of the logistic regression. All three methods evaluated if there was statistical evidence that the absolute difference in treatment effects of the two subgroups was not zero. Two-sided p values were generated without adjustment for multiple testing due to the exploratory nature of these analyses.

Numbers and percentages of patients with TEAEs, SAEs and AESI were evaluated.

Both studies were registered with ClinicalTrials.gov, were conducted in compliance with the Declaration of Helsinki and were approved by the Institutional Review Boards at each investigational centre participating in the studies. All patients provided written informed consent.

3. Results

3.1. Patients

This analysis included 64/787 (8.1%; pooled) and 36/433 (8.3%; ELEVATE UC 52) patients with isolated proctitis and 723/787 (91.9%; pooled) and 397/433 (91.7%; ELEVATE UC 52) patients with more extensive colitis enrolled in the ELEVATE UC clinical programme for endpoints up to Week 12 and from Weeks 16–52, respectively. Baseline demographics and clinical characteristics of patients with isolated proctitis and those with more extensive colitis showed that mean MMS was slightly lower in the isolated proctitis subgroup and lower proportions of patients with isolated proctitis had a baseline ES = 3 (Table 2). Baseline mean and median FCP and hsCRP were lower in the isolated proctitis subgroup than in those with more extensive colitis. More patients with isolated proctitis were also biologic/Janus kinase inhibitor naïve and did not have CS use at baseline, compared with those with more extensive colitis. In patients with isolated proctitis, higher proportions of those receiving placebo were biologic/Janus kinase inhibitor naïve versus those receiving etrasimod 2 mg QD (90.9% vs 73.8%).

Table 2:

Baseline demographics and clinical characteristics of patients with isolated proctitis and those with more extensive colitis in the etrasimod ELEVATE UC clinical programme.

	Patients with isolated proctitis		Patients with more extensive colitis
	ELEVATE UC 52 + ELEVATE UC 12		ELEVATE UC 52 + ELEVATE UC 12
	Placebo QD (N = 22)	Etrasimod 2 mg QD (N = 42)	Placebo QD (N = 238)	Etrasimod 2 mg QD (N = 485)
Age, years, mean (SD)	41.4 (14.0)	38.6 (11.0)	39.4 (13.7)	41.0 (14.0)
Sex, female, n (%)	10 (45.5)	22 (52.4)	89 (37.4)	218 (44.9)
BMI, kg/m2, mean (SD)	25.6 (4.3)	25.2 (6.0)	25.2 (5.0)	24.9 (5.2)
Duration of UC, years, mean (SD)	6.9 (6.6)	7.4 (8.1)	6.7 (6.4)	7.4 (7.3)
Baseline MMS, mean (SD)	5.8 (1.3)	6.0 (1.2)	6.7 (1.1)	6.7 (1.2)
Baseline ES = 3, n (%)	6 (27.3)	13 (31.0)	142 (59.7)	279 (57.5)
Baseline CS use, n (%)	3 (13.6)	8 (19.0)	73 (30.7)	150 (30.9)
Biologic/Janus kinase inhibitor naïve, n (%)	20 (90.9)	31 (73.8)	156 (65.5)	333 (68.7)
Baseline urgency NRS score, mean (SD)	5.1 (2.7)	5.7 (2.6)	6.8 (2.5)	6.7 (2.5)
Baseline faecal calprotectin, mg/kg
Median	57.1	378.1	1135.1	1121.8
Mean (SD)	733.3 (1181.5)	823.7 (1392.6)	2526.6 (5053.3)	2597.1 (5057.0)
Baseline hsCRP, mg/L
Median	1.1	1.2	3.3	4.3
Mean (SD)	1.9 (3.2)	3.0 (4.7)	10.4 (17.7)	9.1 (14.7)

BMI, body mass index; hsCRP, high sensitivity C-reactive protein; QD, once daily.

3.2. Efficacy

Significantly higher proportions (p < 0.05) of patients with isolated proctitis receiving etrasimod versus placebo achieved clinical remission, endoscopic improvement, symptomatic remission, EIHR, and clinical response at Week 12, and clinical and symptomatic remission at Week 52 (Figure 1A and Figure 2). For the subgroup with more extensive colitis, treatment with etrasimod was superior to placebo on all primary and key secondary endpoints at Week 12 and 52 (Figure 1B and Figure S1). Furthermore, significantly greater proportions of patients receiving etrasimod versus placebo achieved CS-free clinical remission and sustained clinical remission at Week 52 in both isolated proctitis and more extensive colitis subgroups (p < 0.001 for all; Figure 3 and Figure S2).

Figure 1.

Clinical remission in (A) patients with isolated proctitis and (B) those with more extensive colitis. Nominal p values are reported without adjustment for multiple comparisons. Treatment comparisons and 2-sided p values were obtained using the Cochran–Mantel–Haenszel method assuming common proportion difference within each subgroup, adjusting for naïvety to prior biologic/Janus kinase inhibitor therapy, baseline corticosteroid use and baseline disease activity (MMS 4–6 or 7–9) and, for pooled study data (Week 12), additionally adjusting for study stratification. Patients with missing responses were considered nonresponders. Δ, adjusted percentage difference for etrasimod minus placebo; QD, once daily.

Figure 2.

(A) Endoscopic improvement, (B) symptomatic remission, (C) EIHR and (D) clinical response in patients with isolated proctitis. Nominal p values are reported without adjustment for multiple comparisons. Treatment comparisons and 2-sided p values were obtained using the Cochran–Mantel–Haenszel method assuming common proportion difference within each subgroup, adjusting for naïvety to prior biologic/Janus kinase inhibitor therapy, baseline corticosteroid use and baseline disease activity (MMS 4–6 or 7–9) and, for pooled study data (Week 12), additionally adjusting for study stratification. Patients with missing responses were considered nonresponders. Δ, adjusted percentage difference for etrasimod minus placebo; QD, once daily.

Figure 3.

(A) CS-free clinical remission and (B) sustained clinical remission in patients with isolated proctitis. Nominal p values are reported without adjustment for multiple comparisons. Treatment comparisons and 2-sided p values were obtained using the Cochran–Mantel–Haenszel method assuming common proportion difference within each subgroup, adjusting for naïvety to prior biologic/Janus kinase inhibitor therapy, baseline corticosteroid use and baseline disease activity (MMS 4–6 or 7–9). Patients with missing responses were considered nonresponders. Δ, adjusted percentage difference for etrasimod minus placebo; QD, once daily.

We further performed additional analyses to assess for a potential difference in treatment effects (etrasimod vs placebo) between patients with isolated proctitis and those with more extensive colitis (i.e. treatment-by-isolated-proctitis-subgroup interaction) for the primary and key secondary endpoints. The CMH-adjusted method showed that absolute differences in treatment effect between patients with isolated proctitis and those with more extensive colitis may exist for clinical remission at Week 12, symptomatic remission at Weeks 12 and 52, and EIHR at Week 12 (Table 3). While clinical remission at Week 52, endoscopic improvement at Week 12, CS-free clinical remission at Week 52, and sustained clinical remission at Week 52 had p values trending to lower values, absolute difference in treatment effects were not suggested based on the evidence. Additionally, none of the p values testing the absolute difference in treatment effects between patients with isolated proctitis and those with more extensive colitis using the crude method or logistic regression method were <0.1. Therefore, the trial data does not demonstrate a difference in treatment effect between isolated proctitis and more extensive subgroups.

Table 3.

Absolute differences in treatment effects between isolated proctitis and more extensive colitis using three statistical approaches.

Endpoints Absolute difference in treatment differences (SE) [p]	CMH-adjusted methoda	Crude methodb	Logistic regression methodc
Clinical remission
Week 12	30.8 (10.7) [0.0039*]	14.4 (11.0) [0.1893]	N/A (N/A) [0.3634]
Week 52	20.4 (11.5) [0.0758]	9.4 (14.7) [0.5191]	N/A (N/A) [0.8681]
Endoscopic improvement
Week 12	24.7 (13.1) [0.0583]	13.1 (12.2) [0.2858]	N/A (N/A) [0.4156]
Week 52	3.5 (19.7) [0.8571]	−7.4 (18.9) [0.6943]	N/A (N/A) [0.4859]
Symptomatic remission
Week 12	23.6 (11.6) [0.0425*]	8.7 (12.3) [0.4816]	N/A (N/A) [0.3815]
Week 52	30.8 (12.2) [0.0117*]	21.7 (14.9) [0.1457]	N/A (N/A) [0.2649]
EIHR
Week 12	28.5 (13.2) [0.0302*]	12.4 (10.8) [0.2478]	N/A (N/A) [0.7458]
Week 52	−20.7 (19.8) [0.2961]	−21.9 (18.4) [0.2331]	N/A (N/A) [0.1037]
Clinical response
Week 12	11.4 (14.4) [0.4270]	6.3 (13.2) [0.6332]	N/A (N/A) [0.4500]
Week 52	7.1 (18.8) [0.7053]	−2.9 (19.0) [0.8775]	N/A (N/A) [0.9214]
CS-free clinical remission
Week 52	19.6 (11.5) [0.0871]	8.7 (14.6) [0.5519]	N/A (N/A) [0.9523]
Sustained clinical remission
Week 52	20.2 (10.8) [0.0617]	3.2 (13.9) [0.8193]	N/A (N/A) [0.4954]

For Week 52, only data from ELEVATE UC 52 were used. For Week 12, data were pooled from both ELEVATE UC 52 and UC 12.

^aIn CMH-adjusted method, adjusted treatment difference was generated adjusting to study stratification (Week 12 pooled data only), baseline CS use, prior biologic/Janus kinase inhibitor use, baseline disease activity (MMS: 4–6 or 7–9); interaction was assessed based on the difference between isolated proctitis (yes vs no), and 2-sided p value was given assuming normal approximation.

^bIn crude method, treatment difference was the simple difference in response proportion without adjusting to baseline factors; interaction was assessed based on the difference between isolated proctitis (yes vs no), and 2-sided p value was given assuming normal approximation.

^cIn logistic regression method, response was modeled on treatment, isolated proctitis (yes vs no), treatment-by-isolated-proctitis interaction, adjusting to study stratification (Week 12 pooled data only), baseline CS use, prior biologic/Janus kinase inhibitor use, baseline disease activity (MMS: 4–6 or 7–9); interaction was assessed based on the interaction term, and 2-sided p value was given based on Wald test. Difference in treatment effects is not presented as the logistic regression model did not report an arithmetic difference in proportions.

^* p < 0.05.

CMH, Cochran–Mantel–Haenszel; N/A, not applicable.

Among patients with isolated proctitis, significantly higher proportions of patients receiving etrasimod had RB cessation (RBS = 0) versus placebo from Week 2 (percentage difference Δ = 26.0% [95% CI 2.5, 49.5]; p < 0.05), with sustained improvements over time (statistically significant difference in patients with isolated proctitis at Weeks 8, 12, 16, 32, and 40, all p < 0.05; Figure 4A). Similarly, in the isolated proctitis group, significantly more patients receiving etrasimod versus placebo achieved symptomatic remission starting at Week 4 (Δ = 23.9 [95% CI 1.6, 46.2]; p < 0.05; Figure 4B), which was maintained at most study visits through to Week 52. In patients with more extensive colitis, significant differences between placebo and etrasimod were first seen at Week 4 for both RB cessation (Δ = 18.4% [95% CI 11.5, 25.3]; p < 0.0001) and symptomatic remission (Δ = 13.5% [95% CI 7.7, 19.3]; p < 0.0001), and maintained at all subsequent study visits (Figure S3).

Figure 4.

Proportion of patients with isolated proctitis with (A) cessation of rectal bleeding (RBS = 0) over time and (B) achievement of symptomatic remission over time. *p < 0.05 for etrasimod versus placebo. Missing responses were considered as nonresponse. N, total number of patients; n, number of patients with evaluable data at a visit; QD, once daily.

In both patients with isolated proctitis and those with more extensive colitis, a statistically significant difference between etrasimod and placebo was seen for improvements in bowel urgency NRS score at Week 12 (patients with isolated proctitis: Δ = −2.6 [95% CI −4.5, −0.7], p = 0.007; patients with more extensive colitis: Δ = −1.3 [95% CI −1.8, −0.8], p < 0.0001) but not at Week 52 (patients with isolated proctitis: Δ = 0.2 [95% CI −3.2, 3.6], p = 0.909; patients with more extensive colitis: Δ = −0.5 [95% CI −1.4, 0.5], p = 0.313; Figure 5 and Figure S4).

Figure 5.

Improvement in bowel urgency NRS in patients with isolated proctitis. Data are reported as observed. ^aEstimates at Week 12 are from ANCOVA for change from baseline with a covariate for baseline score, and stratification factors for naive to biologic/Janus kinase inhibitor therapy at study entry, baseline CS use, baseline disease activity (MMS: 4–6 or 7–9), study stratification and treatment. Estimates at Week 52 are from a mixed model for repeated measures for change from baseline with a covariate for baseline score, same three stratification factors, treatment, visit, and treatment by visit interaction, including data from ELEVATE UC 52 only. Δ, least-squares mean difference; ANCOVA, analysis of covariance; LSM, least-squares mean; QD, once daily.

3.3. Safety

Among all patients with isolated proctitis, the most common TEAEs to occur in ≥4% of patients receiving etrasimod included headache, arthralgia, and increased blood creatine phosphokinase (Table 4). Among patients with more extensive colitis, the most common TEAEs to occur in ≥4% of patients receiving etrasimod were headache, worsening of ulcerative colitis or ulcerative colitis flare and pyrexia (Table 4). SAEs were reported by two (4.8%) and zero patients with isolated proctitis receiving etrasimod and placebo, respectively; among patients with more extensive colitis, SAEs were reported by 24 (5.0%) and 11 (4.6%) patients receiving etrasimod and placebo, respectively.

Table 4.

TEAEs and AESIs in patients with isolated proctitis and those with more extensive colitis.

	Patients with isolated proctitis		Patients with more extensive colitis
	ELEVATE UC 52 + ELEVATE UC 12		ELEVATE UC 52 + ELEVATE UC 12
AEs, n (%) [IR, per 100 PY]	Placebo (N = 22)	Etrasimod 2 mg QD (N = 42)	Placebo (N = 238)	Etrasimod 2 mg QD (N = 485)
Most common TEAEs (≥4% of patients)a
Headache	0	4 (9.5) [17.4]	9 (3.8) [8.9]	31 (6.4) [12.9]
Arthralgia	0	2 (4.8) [8.4]	6 (2.5) [5.9]	15 (3.1) [6.0]
Blood creatine phosphokinase increased	0	2 (4.8) [8.2]	2 (0.8) [1.9]	6 (1.2) [2.4]
Dizziness	0	2 (4.8) [8.6]	1 (0.4) [1.0]	16 (3.3) [6.5]
Pyrexia	0	1 (2.4) [4.0]	9 (3.8) [8.8]	21 (4.3) [8.5]
Ulcerative colitisb	0	0	12 (5.0) [11.8]	29 (6.0) [11.6]
SAEs	0	2 (4.8) [8.3]	11 (4.6) [10.9]	24 (5.0) [9.7]
AESIc
Pulmonary disordersd	0	0	1 (0.4) [1.0]	1 (0.2) [0.4]
Infections	1 (4.6) [11.9]	0	8 (3.4) [7.9]	10 (2.1) [4.0]
Severe infectionse	0	0	5 (2.1) [4.9]	3 (0.6) [1.2]
Herpes zoster	0	0	2 (0.8) [1.9]	2 (0.4) [0.8]
Herpes simplex	0	0	0	1 (0.2) [0.4]
Opportunistic infections (narrow)f,g	1 (4.6) [11.9]f	0	0	1 (0.2) [0.4]
Liver injury	0	0	2 (0.8) [1.9]	7 (1.4) [2.8]
Malignancies	0	0	0	0
Macular oedemai	0	0	0	1 (0.2) [0.4]
Cardiovascular events	0	1 (2.4) [4.0]h	2 (0.8) [2.0]	19 (3.9) [7.6]

For AEs with 0 pts with events, % and IR are also 0, so not displayed. Opportunistic infections is based on the Standardized Medical Dictionary for Regulatory Activities.

^aThe most common TEAEs included above are the most frequently occurring TEAEs among etrasimod-treated patients only, per the isolated proctitis subgroup. Common TEAEs were defined as those reported in >1% of patients (etrasimod group) and with higher IR (proportion) in the etrasimod group than in the placebo group, in the overall population.

^b‘Colitis ulcerative’ includes UC worsening/UC flares.

^cAEs are AESI and met the review criteria.

^dPulmonary disorders were identified based upon appropriate reported event terms/results of lung function tests that identified potential airflow obstruction (FEV1, FVC) and/or decreased gas exchange (DLCO). Airflow obstruction: A decline/decrease in pulmonary function tests (FEV1 or FVC) by ≥20% compared to baseline or ≤50% of predicted at the last documented on-drug test; decreased gas exchange: DLCO corrected for haemoglobin decreased by ≥20% compared with baseline at the last documented on-drug test.

^eSevere infections were defined as those that met the Common Terminology Criteria for Adverse Events version 5.0, Grade ≥3.

^fOpportunistic infections is based on the Standardized Medical Dictionary for Regulatory Activities query (version 24.1), narrow scope.

^gOpportunistic infection was tuberculosis in a patient receiving placebo; the patient discontinued from the study; the event resolved with treatment.

^hCardiovascular event was arterial hypertension unlikely related to study treatment; the patient recovered without change to study treatment.

ⁱTwo additional cases of macular oedema were reported by study investigators; however, these did not meet the review criteria for AESI.

AE, adverse event; AESI, AEs of special interest; DLCO, diffusion capacity of the lungs for carbon monoxide; FEV1, forced expiratory volume; FVC, forced vital capacity; QD, once daily; SAE, serious AE; TEAE, treatment-emergent adverse event.

Proportions of patients with infections were similar between treatment groups in patients with isolated proctitis and those with more extensive colitis (Table 4). No patients with isolated proctitis experienced herpes zoster or herpes simplex infections. A 21-year-old, male patient with isolated proctitis receiving placebo in ELEVATE UC 52 experienced a non-serious opportunistic infection event of tuberculosis (TB) on Day 32 and treatment was discontinued on Day 40; this patient lived in a country with a high prevalence of TB. No patients with isolated proctitis experienced AESIs of liver injury, pulmonary disorder, malignancy or macular oedema. A 60-year-old, female patient with isolated proctitis receiving etrasimod 2 mg QD had experienced an AESI of arterial hypertension. The event lasted 1 day, resolved after one dose of antihypertensive medication and was determined unlikely to be related to the study treatment; therefore, no change was made to the treatment.

4. Discussion

Patients with isolated proctitis are generally excluded from clinical trials, resulting in a paucity of good quality evidence for treatment in this patient population.⁹ Therefore, to our knowledge, ELEVATE UC 52 and ELEVATE UC 12 are among the first phase 3 studies to evaluate the efficacy and safety of a systemic UC treatment in patients with isolated proctitis, thus providing more information for physicians treating these patients.

This post hoc analysis of the etrasimod ELEVATE UC clinical programme demonstrated that etrasimod 2 mg QD was efficacious and well tolerated in patients with isolated proctitis. Most efficacy endpoints were met by higher proportions of patients receiving etrasimod versus placebo in the isolated proctitis subgroup as well as in those with more extensive colitis. Patients with isolated proctitis receiving etrasimod also demonstrated consistent symptomatic improvements versus placebo, with higher proportions achieving RB cessation and symptomatic remission as early as Week 2 and 4, respectively, and maintained to Week 52, and significant improvements in bowel urgency at Week 12, which was the earliest time point evaluated. Safety was consistent between patients receiving etrasimod and placebo, in patients with isolated proctitis and in those with more extensive disease, and in line with the overall ELEVATE UC clinical programme population.¹⁴

Furthermore, to examine if a difference in treatment effect exists between the proctitis and more extensive colitis subgroups, we performed exploratory analyses to assess the existence of an interaction between proctitis subgroup and treatment effect. Using the CMH-adjusted method, interactions existed for clinical remission at Week 12, symptomatic remission at Weeks 12 and 52, and EIHR at Week 12. However, no interactions using the crude method or logistic regression method met the statistical threshold; this suggests that the observed absolute difference in treatment effects in selected endpoints and time points could be dependent on the stratification in the CMH-adjusted method and on the small sample size, rather than true differences in the data. The lack of consistent statistical evidence across endpoints and time points may also indicate that the absolute difference can be dependent on some components of the endpoints. The loss of efficiency in the CHM-adjusted method is well-known when it is applied to smaller sample sizes and a larger number of stratification factors (prior biologic/Janus kinase inhibitor therapy at study entry, baseline CS use, and baseline disease activity [MMS 4−6 or 7−9]), resulting in relatively unstable estimation. As such, we can say that the treatment effect of etrasimod vs placebo is similar between patients with isolated proctitis and those with more extensive colitis based on the trial data. Placebo response rates were also similar for patients with isolated proctitis versus those with more extensive disease. Based on these findings, the notion that isolated proctitis is more difficult to treat is not supported by our study.

Rectal topical formulations (suppositories and enemas) typical of 5-ASA, but occasionally CSs, are administered as first-line treatment for isolated proctitis, with the aim to relieve patients’ symptoms by directly targeting the site of inflammation and thus, preferential healing of the distal large bowel.^1,19 Oral 5-ASAs are less effective than topical 5-ASAs in treating proctitis, probably due to low concentrations of the drug reaching the rectum; however, they are associated with better compliance.^20,21 Similarly, rectally administered corticosteroids, particularly rapidly metabolising corticosteroids with low systemic bioavailability such as budesonide enemas, are also effective in inducing remission in isolated proctitis.^22,23 In patients that fail on combination therapy of topical and oral 5-ASA along with topical corticosteroids, treatment is further escalated to oral corticosteroids.^24,25

When further treatment intensification is necessary in those refractory to 5-ASA/steroid therapy or steroid dependant, other advanced systemic therapies (including immunomodulators and/or TNFi therapy) have been recommended in guidelines and by expert panels for use in patients with isolated proctitis,^25,26 despite a lack of randomised clinical trials, and thus limited or poor quality evidence, and largely by extrapolation from data on more extensive colitis. A small retrospective cohort study of patients receiving TNFi therapy did find that it was effective in inducing clinical remission and mucosal healing in patients with refractory proctitis,²⁷ highlighting the unmet need for the development of novel therapies to manage the treatment of these patients. We now show efficacy of etrasimod, an advanced therapy, in patients with isolated proctitis who have failed at least one UC therapy, thus providing evidence to support its use in this underserved subgroup of patients with UC.

The current analysis demonstrated that treatment with etrasimod 2 mg QD was more efficacious than placebo in patients with isolated proctitis, with significantly greater proportions of patients achieving all efficacy endpoints at Week 12 and most efficacy endpoints at Week 52. A smaller sample size as well as placebo response rates may have contributed to a lack of statistically significant treatment difference for some endpoints at the later time point. A meta-analysis of randomised controlled trials found that patients with isolated proctitis had placebo endoscopic response and remission rates of 32% and 18% after induction.⁹ It should be noted however, that some of the studies included in the meta-analysis were performed before the year 2005, prior to the introduction of central reading which is associated with reduced placebo endoscopic response and remission rates.^9,28 Similarly, in the present study placebo response rates for the endpoints of endoscopic improvement and EIHR were 22.7% and 13.6%, respectively, at Week 12. Placebo response rates were higher at Week 52 for both outcomes (33.3% for both). Further studies are required to determine long-term placebo response in this subgroup.

In addition to clinical and endoscopic improvement as a goal of long-term therapy, short-term symptomatic improvement is key to the management of isolated proctitis,⁴ particularly for urgency and tenesmus which are symptoms traditionally viewed as problematic in patients with isolated proctitis.^1,5 Although improvements in urgency NRS have previously been shown in an overall population of patients with UC treated with mirikizumab,²⁹ we demonstrated that patients with isolated proctitis (the subgroup of patients most likely to suffer from urgency) treated with etrasimod had significant improvements in bowel urgency NRS at Week 12 compared with those receiving placebo; however, this significant change was not seen in Week 52 data, probably due to the “as observed” nature of the data (with no imputation for missing data), whereby patients without improvement in their symptoms in both the placebo and etrasimod arms would have discontinued the study after Week 12 (at which time they could enter the open-label extension) and prior to Week 52, resulting in a low sample size, composed mostly of responders, for analysis. Poor symptom control is associated with reduced quality of life, and inclusion of patient-centred measures in clinical trials has been strongly advocated.^4,30 A randomised controlled trial of patients with mild to moderately active UC found that a shorthand version of the total Mayo score comprising only the RBS and SFS was as effective as the full score in identifying patient-perceived clinical response.¹⁵ In this analysis, higher proportions of patients with isolated proctitis had complete cessation of RB (RBS = 0) by Week 2 and achieved symptomatic remission (defined as SFS = 0 [or = 1 with a ≥1-point decrease from baseline] and an RBS = 0) by Week 4 versus placebo, which was carried through to Week 52. Although these results are encouraging, additional studies of systemic therapies in patients with isolated proctitis, with a focus on patient-centred outcomes, are required to further substantiate these findings.

As expected, there were no new safety signals identified when patients with isolated proctitis were compared with those with more extensive colitis, as well as the overall ELEVATE UC clinical programme population.¹

This study is limited by the fact that patients included in this analysis were enrolled based upon the inclusion and exclusion criteria for the ELEVATE UC clinical programme and therefore may not be representative of the overall proctitis population. This was a post hoc analysis with a relatively small sample size. Furthermore, most efficacy data were only collected at Weeks 12 and 52; therefore, the time-dependent effects on efficacy at other time points cannot be assessed. Finally, the Mayo score was not designed to evaluate patients with isolated proctitis and may not be the most appropriate scoring system for this population.

In conclusion, the efficacy and safety of etrasimod 2 mg QD in patients with isolated proctitis (also referred to as ulcerative proctitis) were consistent with those in the overall population of the ELEVATE UC clinical programme.¹⁴ Patients receiving etrasimod demonstrated greater symptomatic improvements over time compared with those receiving placebo. This analysis expands upon the known efficacy of etrasimod in UC into a subpopulation of patients who have been systematically excluded from drug development trials. Further prospective studies are required in larger cohorts of patients with isolated proctitis to better establish the effects of systemic therapy in this subgroup of patients.

Supplementary Data

Supplementary data are available at ECCO-JCC online.

Glossary

Abbreviations:

Δ

adjusted percentage difference for etrasimod minus placebo

5-ASA

5-aminosalicylates

AE

adverse event

AESI

adverse events of special interest

ANCOVA

Analysis of Covariance

CI

confidence interval

CMH

Cochran–Mantel–Haenszel

CS

corticosteroid

DLCO

diffusion capacity of the lungs for carbon monoxide

EIHR

endoscopic improvement-histological remission

ES

endoscopic subscore

FEV1

forced expiratory volume

FVC

forced vital capacity

IR

incidence rate

LSM

least-squares mean

MMRM

mixed-effect model with repeated measures

MMS

modified Mayo score

NRS

numeric rating scale

PY

patient-year

QD

once daily

RB

rectal bleeding

RBS

rectal bleeding subscore

S1P

selective sphingosine 1-phosphate

SAE

serious adverse event

SD

standard deviation

SE

standard error

SFS

stool frequency subscore

TB

tuberculosis

TEAE

treatment-emergent adverse event

UC

ulcerative colitis

Conflicts of Interest

LPB reports fees from AbbVie, Abivax, Adacyte, Alimentiv, Alma Bio Therapeutics, Amgen, Applied Molecular Transport, Arena, Biogen, Bristol-Myers Squibb, Celltrion, CONNECT Biopharm, Cytoki Pharma, Enthera, Ferring, Fresenius Kabi, Galapagos, Genentech, Gilead Sciences, Gossamer Bio, GlaxoSmithKline, HAC-Pharma, IAG Image Analysis, Index Pharmaceuticals, Inotrem, Janssen, Lilly, Medac, Mopac, Morphic Therapeutics, MSD, Nordic Pharma, Norgine, Novartis, OM Pharma, ONO Pharma, OSE Immunotherapeutics, Pandion Therapeutics, Par’Immune, Pfizer, Prometheus, Protagonist, Roche, Roivant, Samsung, Sandoz, Sanofi, Takeda, Theravance, Thermo Fisher, Tigenix, Tillots, Vectivbio, Ventyx, Viatris, Vifor and Ysopia.

MCD reports consulting fees from AbbVie, Abivax, Arena Pharmaceuticals, AstraZeneca, Bristol-Myers Squibb, Celgene, Eli Lilly, Galapagos, Genentech, Gilead Sciences, Janssen Pharmaceuticals, Pfizer Inc, Prometheus Biosciences, Prometheus Laboratories, Takeda and UCB; grants and research support from Janssen; shareholder/royalties from Trellus Health; and directorship/ownership interest in Treullus Health.

BES reports consulting fees from AbbVie, Alimentiv, Amgen, Arena Pharmaceuticals, Artugen Therapeutics, AstraZeneca, Boehringer Ingelheim, Boston Pharmaceuticals, Calibr, Celgene, Celltrion, ClostraBio, Enthera, Equillium, Evommune, Fresenius Kabi, Galapagos, Genentech (Roche), Gilead Sciences, GlaxoSmithKline, Gossamer Bio, Index Pharmaceuticals, Innovation Pharmaceuticals, Inotrem, Kaleido, Kallyope, Merck, Morphic Therapeutics, MRM Health, Progenity, Prometheus Biosciences, Prometheus Laboratories, Protagonist Therapeutics, Q32 Bio, Sun Pharma, Surrozen, Target RWE, Teva, TLL Pharmaceutical, Ventyx Biosciences; speaker fees from Abivax and Lilly; research grants and consulting fees from Bristol-Myers Squibb, Janssen, Pfizer, Takeda, and Theravance Biopharma; and holds stock in Ventyx Biopharma.

JP reports personal fees from AbbVie, Arena, Athos, Atomwise, Boehringer Ingelheim, Celgene, Celsius, Celltrion, Ferring, Galapagos, Genentech/Roche, GlaxoSmithKline, Immunic, Janssen, Mirum, Morphic Therapeutics, Pandion, Pfizer Inc, Progenity, Prometheus, Revolo, Sanofi, Takeda, Theravance Biopharma and Wassermann; and grant support from AbbVie and Pfizer Inc.

SS reports lecture/speaker fees from AbbVie, Arena, Biogen, Bristol-Myers Squibb, Celgene, Celltrion, Falk, Fresenius, Janssen, MSD, Pfizer Inc and Takeda; and consultancy fees from AbbVie, Arena, Biogen, Bristol-Myers Squibb, Celgene, Celltrion, Falk, Fresenius, Gilead Sciences, IMAB, Janssen, MSD, Mylan, Pfizer Inc, Protagonist, Provention Bio, Takeda and Theravance.

WR reports speaker fees from AbbVie, Celltrion, Falk Pharma GmbH, Ferring, Galapagos Medice, Janssen, MSD, Pfizer, Pharmacosmos, Roche, Shire, Takeda and Therakos; consultancy fees from AbbVie, Amgen, AOP Orphan, Arena Pharmaceuticals, Astellas, AstraZeneca, Bioclinica, Boehringer Ingelheim, Bristol-Myers Squibb, Calyx, Celgene, Celltrion, Eli Lilly, Falk Pharma GmbH, Ferring, Galapagos, Gatehouse Bio Inc., Genentech, Gilead Sciences, Grünenthal, ICON, Index Pharma, Inova, Janssen, Landos Biopharma, Medahead, MedImmune, Microbiotica, Mitsubishi Tanabe Pharma Corporation, MSD, Novartis, OMass, Otsuka, Parexel, Periconsulting, Pharmacosmos, Pfizer, Protagonist, Provention, Quell Therapeutics, Sandoz, Seres Therapeutics, Setpointmedical, Sigmoid, Sublimity, Takeda, Teva Pharma, Therakos, Theravance and Zealand; is an advisory board member for AbbVie, Amgen, AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Celltrion, Galapagos, Janssen, Mitsubishi Tanabe Pharma Corporation, MSD, Pharmacosmos, Pfizer, Sandoz and Takeda; and receives research funding from AbbVie, Janssen, MSD, Sandoz, Sanofi and Takeda.

BGF is a senior scientific director for Alimentiv Inc, which provides central reading services. He is not a company employee and has no equity stake in the organisation which is owned by a medical trust. He also reports speaker fees for AbbVie, Janssen, and Takeda; is a consultant/advisory board member for AbbVie, AbolerIS, AgomAB Therapeutics, Allianthera, Amgen, AnaptysBio, Applied Molecular Transport Inc, Arena Pharma, Avoro Capital Advisors, Atomwise, BioJamp, Biora Therapeutics, Boehringer Ingelheim, Boxer, Celsius Therapeutics, Celgene/Bristol-Myers Squibb, Connect BioPharma, Cytoki, Disc Medicine, Duality, EcoR1, Eli Lilly, Equillium, Ermium, First Wave, First Word Group, Galapagos, Galen Atlantica, Genentech/Roche, Gilead, Gossamer Pharma, GlaxoSmithKline, Hinge Bio, Hot Spot Therapeutics, Index Pharma, Imhotex, Immunic Therapeutics, JAKAcademy, Janssen, Japan Tobacco Inc., Kaleido Biosciences, Landos Biopharma, Leadiant and L.E.K, Lenczner Slaght, LifeSci Capital, Lument AB, Millennium, MiroBio, Morgan Lewis, Morphic Therapeutics, Mylan, OM Pharma, Origo BioPharma, Orphagen, Pandion Therapeutics, Pendopharm, Pfizer, Play to Know AG, Progenity, Prometheus Therapeutics and Diagnostics, Protagonist, PTM Therapeutics, Q32 Bio, Rebiotix, REDX, Roche, Sandoz, Sanofi, Seres Therapeutics, Silverback Therapeutics, Surrozen Inc., Takeda, Teva, Thelium, Tigenix, Tillotts, Ventyx Biosciences, VHSquared Ltd., Viatris, Ysios, Ysopia and Zealand Pharma; and is a shareholder of Gossamer Pharma.

SD reports lecture fees from AbbVie, Amgen, Ferring Pharmaceuticals Inc., Gilead Sciences, Janssen, Mylan, Pfizer and Takeda; consultancy fees from AbbVie, Allergan, Amgen, AstraZeneca, Biogen, Boehringer Ingelheim, Celgene, Celltrion, Ferring, Gilead Sciences, Hospira, Janssen, Johnson and Johnson, MSD, Mundipharma, Pfizer Inc, Roche, Sandoz, Takeda, TiGenix, UCB and Vifor; and directorship/ownership for Gastroenterology and Endoscopy.

AJY reports consultancy fees from Arena, Bristol-Myers Squibb, Pfizer and Takeda; and lecture fees from Bristol-Myers Squibb.

GRDH is an advisor for AbbVie, Alimentiv, AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Celltrion, Cosmo, Eli Lilly, Galapagos, GlaxoSmithKline, Johnson and Johnson, Pfizer Inc, Polpharma, Prometheus Biosciences, Takeda, Tillotts and Ventyx; and reports speaker fees for AbbVie, Boehringer Ingelheim, Celltrion, Eli Lilly, Johnson and Johnson, Pfizer Inc, Takeda and Tillotts.

MG is an employee and shareholder of Pfizer AG.

KW is an employee and shareholder of Pfizer Canada Inc.

MK, JW, IM, and LB are employees and shareholders of Pfizer Inc.

AM is an employee of Pfizer Ltd, and a shareholder of Pfizer.

SV reports lecture fees from AbbVie, Dr. Falk Pharma, Ferring, Hospira, MSD, Takeda and Tillotts; consultancy fees from AbbVie, AbolerIS Pharma, Alimentiv, Arena, AstraZeneca, Avaxia, BMS, Boehringer Ingelheim, Celgene, CVasThera, Dr Falk Pharma, Eli Lilly, Ferring, Galapagos, Genentech/Roche, Gilead, Hospira, Imidomics, Janssen, Johnson and Johnson, Materia Prima, MiroBio, Morphic, MrMHealth, MSD, Mundipharma, Pfizer Inc, Prodigest, Progenity, Prometheus, Robarts Clinical Trials, Second Genome, Shire, Surrozen, Takeda, Theravance Biopharma, Tillots Pharma AG and Zealand Pharma; and grant/research support from AbbVie, Galapagos, MSD, Pfizer Inc and Takeda.

Funding

This work was funded by Pfizer Inc.

Author Contributions

MG, KW, MK, KL, AM, and JW were involved in the conception and design of the study; JW performed the statistical analyses; WR was a central reader for the trials. All authors contributed to data interpretation. All authors contributed to the development of the manuscript and all authors approved the final version. All authors agree to be accountable for all aspects of the work.

Data Sharing Statement

Upon request, and subject to review, Pfizer will provide the data that support the findings of this study. Subject to certain criteria, conditions, and exceptions, Pfizer may also provide access to the related individual de-identified participant data. See https://www.pfizer.com/science/clinical-trials/trial-data-and-results for more information.