Patients With Crohn’s Disease and Permanent Ileostomy Are Universally Excluded From Clinical Trials: A Systematic Review

Abstract

INTRODUCTION:

We performed a systematic review to investigate whether patients with Crohn’s disease (CD) and permanent ileostomy (PI) have been included in clinical trials evaluating biologics and small molecules.

METHODS:

MEDLINE, Embase and Cochrane library (CENTRAL) data bases were searched from inception to May 16, 2022 for placebo controlled induction and/or maintenance randomized controlled trials assessing biologics and oral small molecules in adult patients with active CD.

RESULTS:

Of the 81 induction and maintenance trials assessing biologics and oral small molecules in CD, none permitted the enrollment of patients with PI. Patients with CD and PI have been universally excluded from pharmaceutical trials of biologics and small molecules to date.

DISCUSSION:

There is an urgent need to identify barriers to enrollment and develop eligibility and outcome measures enabling the inclusion of patients with CD and PI into clinical trials.

INTRODUCTION

Approximately 7 million people live with inflammatory bowel disease worldwide and between 1990 and 2017, there has been an 85% increase in the prevalence of inflammatory bowel disease globally causing significant burden on healthcare systems (1). Crohn’s disease (CD) can be progressive, leading to stricturing and penetrating complications requiring surgery in up to 40% of patients within 10 years of diagnosis, with one-third requiring a second resection within 10 years of the first resection, thus increasing the risk of permanent ileostomy (PI) (2). PI is typically performed in medically refractory ileo-colonic/colonic CD after subtotal colectomy or total proctocolectomy. The annual risk of any stoma formation in patients with CD undergoing surgery is 1.8 stomas per 100 person years (3) and, with a 5-year cumulative risk of stoma surgery estimated at 2.5% from a population-based study (4). PI can be associated with negative psychological outcomes including an increase in antidepressant use, anxiety, and a negative impact on body image and intimacy (5). Approximately one-third of patients with PI after total proctocolectomy experience clinical recurrence in the small bowel which may need subsequent medical therapy, and 16% experience significant recurrence requiring further surgery (6). A meta-analysis of 18 cohort studies showed cumulative clinical recurrence rates of 23.5% and 40% at 5 and 10 years, respectively (7). Although rates of surgery have declined for CD in the past 2 decades (2), the rate of elective stoma surgery has remained stable (3). There is very limited evidence on the efficacy of biologics and oral small molecules in patients with PI. Therefore, we performed a systematic review to investigate whether patients with CD and PI have been included in CD clinical trials evaluating biologics and small molecules.

METHODS

MEDLINE, Embase, and the Cochrane library (CENTRAL) databases were searched from inception to May 16, 2022. Eligible studies were randomized controlled trials fulfilling the following criteria: (i) a placebo-controlled induction and/or maintenance trial of biologic agents and oral small molecules in adults with active CD, (ii) use of the Crohn’s Disease Activity Index or Harvey-Bradshaw Index for enrollment and/or assessment of response/remission, and (iii) minimum duration of at least 2 weeks for induction and 4 months for maintenance of remission trials. Trials where only the abstract was available without the eligibility criteria were excluded. In addition, trials of probiotics, antibiotics, corticosteroids, 5-aminosalicylates, immunomodulators, complementary therapies, or devices were excluded as were trials of hospitalized patients with severe CD.

RESULTS

After excluding duplicates, 11,693 records were selected for screening (Figure 1). In total, 144 records were selected for full-text assessment and 81 randomized controlled trials (induction: 58 and maintenance: 23) with 22,000 participants were eligible for inclusion (Table 1).

Figure 1.

PRISMA diagram showing study selection.

Table 1.

Characteristics of included induction and maintenance of Crohn’s disease trials

Characteristic	Induction trials	Maintenance trials	Total
No. of trials (n)	58	23	81
No. of participants (n)	15,612	6,388	22,000
Trials including participants with ostomy, n (%)
Ostomy excluded	39 (67.2)	16 (69.6)	55 (68.0)
Ostomy not stated	19 (32.8)	7 (30.4)	26 (32.1)
Class of drug, n (%)
Biologics	51 (87.9)	19 (82.6)	70 (86.4)
Small molecules	7 (12.1)	4 (17.4)	11 (13.6)
Trial phase, n (%)
I	2 (3.4)	0 (0.0)	2 (2.5)
II	32 (55.2)	9 (39.1)	41 (50.6)
III	24 (41.4)	14 (60.9)	38 (47.0)
Age, y
Mean (SD)	36.8 (11.9)	36.8 (12.3)	—
Range	27.0–41.6	31.2–39.5
Sex of participants,a n(%)
Male	7,292 (46.7)	2,136 (45.1)	—
Female	8,320 (53.3)	2,603 (54.9)

^aData for the sex of the participants were available for only 16 maintenance trials.

Among induction studies, the majority were phase 2 studies (n = 32, 55.2%), with 87.9% (n = 51) evaluating biologics and the remaining studies evaluating oral small molecules. Most of the participants were female (53.3%), with a mean age of 36.8 ± 11.9 years. Thirty-nine induction studies (67.2%) excluded patients with an ostomy, and the remaining trials (n = 19, 32.8%) did not comment.

Among maintenance trials, 82.6% (n = 19) assessed biologics and 17.4% (n = 4) oral small molecules. Most of the participants were female (54.9%), with a mean age of 36.8 ± 12.3 years. Sixteen studies (69.6%) excluded patients with an ostomy, and the remaining 7 (30.4%) studies did not comment. All studies that did not comment on the exclusion of patients with stoma did not allow the enrollment of these patients as part of the inclusion criteria. For trials reporting on patients with a prior bowel resection (21 induction and 13 maintenance trials), the numbers ranged from 2.4% to 51%.

DISCUSSION

Postoperative recurrence after ileocolonic resection with primary anastomosis occurs in almost all patients, with endoscopic lesions preceding clinical recurrence (8), for which antitumor necrosis factor therapy is effective in the prevention of endoscopic recurrence (9,10). However, we cannot infer that evidence for treating postoperative CD after ileocolonic resection and primary anastomosis can be used to guide the management of patients with PI. We found that no clinical trial of a biologic or oral small molecule published to date included patients with a PI. Optimal management of CD requires timely rational decision making supported by high-quality evidence. Unfortunately, the quality of evidence guiding the management of CD patients with PI is poor; therefore, there is a need to include CD patients with a PI in pharmacological trials. However, several barriers precluding the recruitment of these patients in CD trials need to be addressed. First, PI is often considered a last resort in refractory patients who have exhausted available medical therapies, making them less likely to be candidates for trial recruitment. Second, the commonly used clinical outcome measures such as the Crohn’s Disease Activity Index, Harvey-Bradshaw Index, and patient-reported outcomes based on stool frequency have not been developed or validated in patients with PI. In addition, the operating characteristics of endoscopic instruments, such as CD endoscopic index of severity and simple endoscopic score for CD, have not been evaluated in patients with CD and PI, posing challenges in determining endoscopic outcomes for these patients in clinical trials. The partially validated Rutgeerts score to predict postoperative clinical recurrence after ileocolonic resection and primary anastomosis (11) has not been systematically evaluated in patients with PI. Another barrier for enrollment may be difficulty in achieving sufficient sample size, given this is a subtype of CD. To overcome these challenges, it would be important to evaluate the operating properties of existing disease activity instruments for CD in patients with PI, which would inform the need to develop novel indices. In addition, regulatory authorities could play a key role in prioritizing research in this area by providing support for accelerated drug development pathways, for example through orphan designation and/or enabling more efficient clinical trial designs. Data gathered from real world evidence and pharmacoepidemiologic studies could serve as valuable information for demonstrating the feasibility and safety of enrolling such a population.

In conclusion, we found that patients with CD and PI have been universally excluded from all pharmaceutical trials of biologics and small molecules to date. This group represents a high-risk patient population that have universally already failed advanced therapy before requiring ileostomy formation and may require medical therapy to prevent and treat recurrence. The quality of existing evidence to guide the management of CD patients with PI is poor, and there is an urgent need to identify barriers to enrollment and develop eligibility and outcome measures that enable the inclusion of patients with CD and PI into clinical trials.