Comparative effectiveness of ustekinumab versus infliximab in the management of perianal fistulizing Crohn's disease: a retrospective study in China
Article information
Abstract
Background/Aims
Ustekinumab (UST) and infliximab (IFX) are both effective in the treatment of perianal fistulizing Crohn’s disease (CD), but limited research has focused on comparing the efficacy of UST versus IFX in this field. This study aimed to compare the effectiveness of UST or IFX in treating perianal fistula of CD patients naive to biological agents in a real-world setting.
Methods
A retrospective cohort study included patients with perianal fistulizing CD treated with UST or IFX was conducted to evaluate the rates of luminal and perianal fistula response and remission at 6 months after treatment.
Results
Ninety-seven patients (49 UST and 48 IFX) were enrolled. Compared to IFX, UST exhibited significantly higher rates of treatment success (89.8% vs. 50.0%, P< 0.001) and intestinal clinical response (85.7% vs. 68.8%, P= 0.048), but no significant differences in fistula remission, fistula response, fistula closure, intestinal clinical remission, endoscopic remission and endoscopic response was observed. Furthermore, multivariate analyses demonstrated complexity of fistula was conversely associated with fistula remission between the UST and IFX groups. Finally, the rates of disease relapse and operation in the IFX group were higher as compared to the UST group during follow-up.
Conclusions
UST may serve as a promising alternative to IFX for the treatment of perianal fistulizing CD.
INTRODUCTION
Crohn’s disease (CD) is a chronic inflammatory disease affecting the gastrointestinal tract, characterized by transmural inflammation that can destroy the integrity of the intestinal mucosa favoring the development of abscesses and fistulas [1]. The formation of fistulas might make abnormal tracks between intestines or between the intestine and other organs such as the bladder, vagina or adjacent tissue [2,3]. An abnormal cross-talk between the rectum or anal canal and the external perianal or ischioanal skin indicates a perianal fistula has formed. Perianal fistula is one of the most common and difficult complications in CD, with an estimated incidence up to 28% within the first two decades after diagnosis [4]. Perianal fistulizing CD might be caused by a complex interplay of genetic predisposition, immune dysregulation, gut dysbiosis, and various unknown physiological and mechanical factors [2]. Most notably, patients with perianal CD are inclined to have significantly higher rates of hospitalizations, surgeries, healthcare costs and overall mortality, and a significantly decreased health-related quality of life [5,6]. Furthermore, the presence of perianal fistula might be a predictor of poor long-term prognosis in CD [3]. A complete assessment of fistula characteristics is the basis for optimal management and must include the anatomical description and classification of perianal fistula. Clinically, endoscopy helps to assess the presence of proctitis, and magnetic resonance imaging (MRI) to determine the anatomy of fistula tracts and presence of abscesses [7].
Although medical treatment for CD is mainly focused on the control of bowel inflammation, a range of medical and surgical treatment options are available for different types of perianal fistulas [8-10]. For simple perianal fistulas, antibiotics and immunomodulators are appropriate measures. For complex perianal fistula, the efficacy of biological agents is being evaluated, and anti-tumor necrosis factor (TNF) treatment remains the most accepted therapy with concomitant use of antibiotics or immunomodulators [11]. In addition, stem cell-based therapies have become an attractive approach for patients with CD who develop perianal fistulae [10]. Mesenchymal stem cells and adipose-derived mesenchymal stem cells have been used to treat complex perianal fistulas with promising outcomes from several studies [12-15]. Furthermore, hyperbaric oxygen therapy and exclusive enteral nutrition have also been investigated as adjunct therapies in this field [16-18]. Despite the expansion of the medical armamentarium, a large proportion of the patients require surgical interventions to achieve fistula closure, and intersphincteric fistula tract ligation, mucosal advancement flap surgery or fistula laser closure are the most common surgical approaches [19,20].
Currently, the types of biologics for CD in China include anti- TNF agents, anti-interleukin 12/23 agents, and anti-α4β7 integrin agents. Anti-TNF therapies have demonstrated efficacy in several clinical trials, and infliximab (IFX) was recommended in the European Crohn’s and Colitis Organisation guidelines as a first-line medical therapy following adequate drainage of complex perianal fistulas [21]. Accumulating evidence has found that higher IFX trough levels are associated with fistula healing and closure, but up to 60% of patients treated with maintenance IFX lose response within 1 year due to subtherapeutic anti-TNF trough levels [22]. Data with other biological therapies for complex perianal fistula is scarce, and the evidence is limited to retrospective studies and post hoc analyses of clinical trials for luminal CD [23-25]. Ustekinumab (UST) presents a promising medication for the treatment of perianal fistulizing disease with an excellent safety profile in a recent post hoc analysis from a pooled study of CERTIFI, UNITI-1 and UNITI-2, but no definite conclusions can be drawn from studies not designed primarily to assess fistula response/remission [8,26]. Further prospective studies with a larger patient population and long-term follow-up are needed to further establish the efficacy of UST in the treatment of perianal fistula of CD.
The aim of the current study was to compare the efficacy and safety of UST and IFX in the treatment of patients with perianal fistulizing CD, and identify potential predictors of treatment success.
METHODS
1. Study Design and Population
This is a retrospective monocentric cohort study of enrolled patients with perianal fistula at the First Affiliated Hospital of Sun Yat-sen University (Guangzhou, China) from June 2014 to June 2024. The current study was approved by the Institutional Ethics Committee of Sun Yat-sen University (No. [2024]699) and adhered to the principles of the Declaration of Helsinki. The requirement for informed consent was waived by the Institutional Ethics Committee.
To avoid selection bias, all patients who met inclusion criteria and did not meet any exclusion criteria were included. The inclusion criteria were included: (1) patients were diagnosed with CD based on clinical, endoscopic, radiological and histopathological criteria; (2) active perianal fistula was confirmed by clinical symptoms and MRI at baseline; (3) the age of included patients was older than 18 years of age; or (4) clinical, endoscopic and radiological results were assessed at 6 months after UST or IFX therapy initiation. Patients were excluded if they (1) were diagnosed without perianal fistulae, (2) received any other biologic use prior to UST or IFX treatment, (3) were lost to follow-up or had insufficient follow-up time, and (4) had incomplete medical records including missing medication exposures or surgical history.
UST induction therapy consisted in an intravenous dose according to patient weight (260 mg for those weighing <55 kg, 520 mg for those weighing >85 kg, and 390 mg for those weighing between 55 and 85 kg) followed by subcutaneous administration of 90 mg every 8 or 12 weeks. Patients received standard IFX induction dosing (5 mg/kg intravenously at weeks 0, 2, and 6) followed by maintenance therapy. Patients with dose escalation (including dose escalation, and/or dose interval shortening) were also included. If needed, perianal surgery was performed before starting the UST or IFX therapy.
2. Data Collection
Baseline data related to demographics and CD characteristics (both luminal and perianal) were derived from medical records, including age, sex, smoking status, disease course, Montreal classification, extraintestinal manifestations, number of fistula openings, fistula-related procedures and surgeries, and medical history. Data regarding patient-reported symptoms and clinician physical examination were extracted from the standardized template of inflammatory bowel disease visits. Laboratory values (C-reactive protein [CRP], erythrocyte sedimentation rate, albumin) were also recorded. Furthermore, colonoscopy reports and images (MRI) were reviewed prior to induction and at intervals of 6 months after induction to assess luminal and anal fistula lesions. The missing data were replaced by the data from the prior month to the following month for processing.
3. Definitions
Fistula type based on Park’s classification and American Gastroenterological Association (AGA) classification was determined using MRI.2 According to the connection between fistulae and external sphincter muscle, 5 groups of fistulae are recognized: (1) superficial, without crossing any sphincter or muscular structure; (2) intersphincteric, between the internal sphincter and external sphincter, coursing through the intersphincteric space to the perianal skin; (3) transsphincteric, crossing the external sphincter; (4) supra-sphincteric, penetrating the intersphincterie space and spreading over the top of the puborectalis and crossing the levator muscle before reaching the skin; (5) extra-sphincteric, outside the external sphincter and penetrating the levator muscle. Perianal fistulas can also be categorized as simple or complex. A simple fistula is low (superficial, intersphincteric or transsphincteric) with a single external opening and no abscesses, rectovaginal involvement, or anorectal stricture. All other fistulas are classified as complex fistulas.
4. Study Outcomes
The primary endpoint was the proportion of participants who achieved fistula remission, defined as closure of all draining fistulas present at baseline and no fluid collections of >1 cm in at least 2 dimensions on pelvic MRI. Secondary endpoint was fistula response, defined as a reduction from baseline of 50% or more in the number of draining tracts after UST or IFX therapy and no fluid collections of >1 cm in at least 2 dimensions on pelvic MRI. Crohn’s Disease Activity Index (CDAI) was evaluated, and intestinal clinical remission was defined as a CDAI <150, with intestinal clinical response defined as a >70 reduction in CDAI and/or CDAI <150. Rutgeerts score and Simple Endoscopic Score for Crohn’s Disease (SES-CD) were used to evaluate endoscopic response or remission [27,28]. Endoscopic response was defined as a reduction of one grade from baseline in Rutgeerts score or a reduction of >50% in SES-CD, and endoscopic remission was defined as a Rutgeerts score ≤ i1 or SES-CD ≤2. Treatment success was defined as: (1) clinical success assessed at 6 months of treatment based on the physicians’ judgment, and (2) no need for specialized therapy (antibiotics and/or surgery) for perianal lesions [29]. CRP normalization was defined as a CRP level of <5 mg/L according to the center’s inspection standards. Clinical, endoscopic, and radiological evaluations were collected at 6 months after biological agent initialization. Clinical relapse of perianal fistulizing CD was defined as the appearance of drainage through fistulas that were previously in remission and/or increasing luminal activity (both a CDAI ≥ 150 and an increase in the CDAI ≥ 70) [30]. The onset of concomitant therapy or adjustment of adjuvant therapy to maintain perianal fistulizing CD remission or surgical treatment for perianal lesions was also considered relapse. Biologic persistence was defined as the absence of clinical relapse (including intestinal and perianal fistula activity, treatment escalation, and surgery) and unplanned surgical treatment.
5. Statistical Analysis
Continuous data were expressed as mean ±standard deviation or median and interquartile range (IQR). Categorical data were presented as frequencies and percentages. Student t test or Wilcoxon test were used to compare the indexes before and after treatment. Univariate and multivariate logistic regression analyses were performed to determine the factors that predicted treatment success, expressed as odds ratios (OR) and 95% confidence intervals (CI). Prior selection variables included age at diagnosis of CD, sex, smoking status, duration of CD before initiation of biologic therapy, perianal abscess, complexity of fistula, history of intestinal surgery caused by CD, history of anal fistula surgery, history of medication, and concomitant therapy with steroid or immunosuppressants. The relapse-free and surgery-free survival of patients with perianal fistulizing CD were calculated by the Kaplan-Meier method. All statistical analyses were performed using SPSS for Windows, ver. 25.0 (IBM Corp., Armonk, NY, USA) and R version 4.2.1 (R Foundation for Statistical Computing, Vienna, Austria). A two-sided P-value of <0.05 was used as the level of statistical significance.
RESULTS
1. Patient Characteristics
A total of 97 patients with perianal fistula CD, who met the inclusion criteria, were included in this study. Among them, 49 patients received UST treatment (UST group) and 48 patients received IFX treatment (IFX group) as first-line biologic after diagnosis, respectively (Fig. 1). The demographic and clinical characteristics between these 2 groups are shown in Table 1, and there were no significant differences in demographic and clinical variables between the 2 groups. The mean age at biologics initiation in UST-treated patients was 29.8 years and 31.5 years in IFX-treated patients (P=0.456). Most of the included patients were current non-smokers (96.9%), and no significant difference was observed between both cohorts. The disease behavior of most patients was stricturing or penetrating (65.3% in UST group and 54.2% in IFX group) and most fistulas were defined as complex (69.4% in UST group and 68.8% in IFX group) according to the AGA classification. Compared with the IFX group, the proportion of patients with a previous history of intestinal surgery in the UST group was higher (10.2% vs. 27.1%, P=0.033), but previous history of perianal surgery showed no statistical difference (28.6% vs. 35.4%, P=0.470). In terms of medication history, the rate of concomitant therapy with immunomodulators (thiopurines, methotrexate, or thalidomide) was significantly lower in UST group (2.0% vs. 33.3%, P<0.001) as compared to the IFX group. In addition, no statistical differences were observed in fistula type, anatomic complexity of fistula, fistula openings, fistula tract and serological biomarkers between these 2 groups.
Study flowchart. CD, Crohn’s disease; UST, ustekinumab; IFX, infliximab.
Demographics and Clinical Characteristic of Patients with Crohn’s Disease
2. Efficacy Outcomes between UST and IFX Groups
At 6 months, the proportion of participants treated with UST achieving the primary endpoint (fistula remission) was 16.3% (8/49), and the proportion was 22.9% (11/48) in the IFX group (P=0.419) (Table 2). In addition, rates of intestinal clinical response (85.7% vs. 68.8%, P=0.048) and treatment success (89.8% vs. 50.0%, P< 0.001) in the UST group were significantly higher than those in the IFX group. However, there were no significant differences in fistula response (P=0.920), intestinal clinical remission (P= 0.624), endoscopic remission (P= 0.890) and response (P=0.495), and CRP normalization (P=0.313) between the 2 groups.
Treatment Outcomes between UST and IFX Groups
3. Biologic Persistence between UST and IFX Groups
During a median follow-up of 10.1 months (IQR, 6.3–14.0 months) in UST group, 3 out of 49 patients (6.1%) required replacement of another biologics, 1 out of 49 patient (2.0%) was given the additional immunomodulators, and 4 out of 49 patients (8.2%) underwent optimized treatment (including dose escalation, re-induction and/or dose interval shortening) (Table 3). In addition, 10 out of 49 patients (20.4%) had a clinical relapse of perianal fistulas and 4 out of 49 patients (8.2%) received surgery after UST treatment initiation. Of the 48 patients from the IFX group, after a median follow-up time of 31.0 months (IQR, 16.3–41.3 months). Sixteen of 48 patients (33.3%) received treatment escalation of another biologics, 8 out of 48 patient (16.7%) required the addition of immunomodulators or seton use in combination and 3 out of 48 patients (6.3%) had optimized treatment. Besides, 28 out of 48 patients (58.3%) underwent disease relapse and 6 out of 48 patients (12.5%) received surgery after IFX therapy initiation. Furthermore, biologic persistence (disease relapse-free survival [log-rank P=0.153] and surgery-free survival [log-rank P=0.779]) showed no significant difference in the UST group compared to the IFX group (Fig. 2). It is noteworthy that the cumulative survival rate of disease relapse-free in the UST group was higher than that in the IFX group at 12 months after treatment, but this superiority was reversed over time.
Follow-up Outcomes between UST and IFX Groups
Kaplan-Meier curve for clinical outcomes during the follow-up period according to the time of initiation of UST or IFX therapy. (A) Survival of disease relapse-free in UST and IFX groups. (B) Survival of operation-free in UST and IFX groups. UST, ustekinumab; IFX, infliximab.
4. Predictors Associated with Fistula Remission between UST and IFX Groups
Baseline factors associated with fistula remission in UST and IFX groups were further identified. According to the multivariate analysis, factors associated with 6-month fistula remission included complex perianal fistula (OR, 7.319; 95% CI, 1.177–45.515; P=0.033), older than age 35 (OR, 0.139; 95% CI, 0.022–0.904; P=0.039) (Table 4). However, univariate analysis only demonstrated a significant difference in complex perianal fistulas (OR, 5.167; 95% CI, 1.04–25.565; P=0.044) in UST group, indicating complex perianal fistulas affecting the 6-month fistula remission. In the IFX group, complex perianal fistula was conversely associated with 6-month fistula remission in univariate analysis (OR, 0.158; 95% CI, 0.037–0.676; P=0.013) and in the multivariable analysis (OR, 0.218; 95% CI, 0.048–0.989; P=0.048). No other factors, including sex, smoking status, disease behavior, concomitant medication and previous history of perianal surgery predicted 6-month fistula remission in patients with CD treated with UST or IFX.
Univariate and Multivariate Analyses of Factors Associated with Fistula Remission between UST and IFX Groups
5. Safety between UST and IFX Groups
For the enrolled patients during follow-up, 10.2% of patients (n =5) treated with UST and 10.4% of patients (n =5) treated with IFX were reported adverse events, respectively (Table 5). Infection was the most common adverse event in both groups, but there was no statistical difference. Additionally, UST-related adverse events also included 1 neurologic event, and IFX-related adverse events included 1 neurologic event, 1 dermatologic event, and 1 allergy.
Adverse Events between UST and IFX Groups
DISCUSSION
There is limited data on the role of newer biological agents such as UST in perianal fistulizing CD, especially the efficacy superiority of anti-TNF agents versus novel biological agents remained unknown. Our study is currently the first real-world study to compare the efficacy of UST versus IFX as a first-line biologic for the treatment of perianal fistulizing CD. Although UST and IFX groups presented equal efficacy outcomes in fistula remission and fistula response, UST had higher rates of improvement in intestinal clinical remission and treatment success compared to IFX noted on clinician assessment and physical examination. In addition, we identified complex perianal fistulas in patients with CD might be positively associated with higher rates of 6-month fistula remission in the UST group, but opposite correlation existed in the IFX group.
UST, an antagonist of the p40 subunit of interleukin-12 and interleukin-23 in the inflammatory pathway, had been approved for the treatment of moderate to severe CD in 2016 [31,32]. Current published trials showed a statistically significant difference over placebo in induction and maintenance of remission for luminal CD with an excellent safety profile, but the efficacy of UST in perianal fistula healing is still lacking [33]. A real-world experience and systematic review with meta-analysis conducted in 2021 found that 48.1% of patients (13/27) achieved fistula response with none achieving fistula remission in their cohort at 6 months after UST treatment, which was in concordance with the results from our study that 49% of patients achieved fistula response [34]. Remission rates for perianal fistula CD are well known to be exceedingly low, and this is further supported by our data that approximately 16% of patients achieved fistula remission after initiating UST therapy. Interestingly, there was a drastic increase in the rate of fistula remission in both UST and IFX groups when comparing the 6-month remission rate (9.5%) reported by other studies [34,35]. The most likely explanation for this difference is that our study was specifically using first-line biologics after diagnosis of perianal fistula CD (second- line agents are known to be less efficacious).
There is increasing evidence supporting that certain factors (particularly higher drug levels and combination therapy) are associated with fistula healing in CD [36-38]. In our study, the presence of complex perianal fistulas predicted 6-month fistula remission in the UST group, but complex perianal fistulas might be a risk factor for those treated with IFX. This result might be influenced by the sample size or the shorter disease duration in the UST group, indicating that these patients might have received treatment at a more recent time, and their willingness to seek medical care and living environment have improved compared to before. Furthermore, presence or absence of previous perianal surgery was not statistically significant in predicting fistula response or remission in both UST and IFX groups. Other clinical features also failed to predict fistula remission at 6 months after UST or IFX initiation in the current study. These findings may suggest that patients with complex perianal fistulas were specifically using UST as first-line biologics after diagnosis of CD, and they might have higher fistula response and remission rates compared to those received who IFX therapy.
We used biologic persistence over time, characterized by Kaplan-Meier methods, as a surrogate marker for sustained efficacy outcomes, with the assumption that patients typically remain on therapy that is efficacious without significant adverse events. During extended follow-up, most patients who initially received UST or IFX continued therapy without serious adverse effects. The higher recurrence rate in the IFX group within 1 year suggests that UST may have better short-term treatment persistence compared to IFX. In addition, the rate of treatment escalation in IFX group is much higher than that in UST group, further indicating that UST induction and maintenance therapy are associated with higher treatment persistence compared to IFX. While clinicians may choose IFX or UST for those with severe luminal disease, we found that UST had higher rates compared to IFX in perianal fistula healing and persistence in CD, which suggests our study may underestimate the impact of UST for patients with perianal disease, especially among those with complex perianal fistulas. However, the follow-up period for the UST group was relatively short, which might have failed to capture the potential long-term recurrence risk. In contrast, the long-term follow-up period for the IFX group might more accurately reflect the long-term efficacy of the treatment, but it might also increase the cumulative risk of recurrence due to the extended time. Future studies should try to balance the follow-up time of the 2 groups as much as possible to ensure the reliability and comparability of the results. Therefore, additional prospective studies are needed to compare the efficacy between anti-TNF agents and UST for perianal fistula healing over time using more objective measures and therapeutic drug monitoring protocols. Therefore, our study supports that UST may be more effective than IFX on perianal fistula healing over time, suggesting that UST may be considered first line among providers over IFX for treatment of perianal fistula CD.
Our cohort study has several strengths. This is the first study to compare the efficacy of UST to IFX as the first-line biologic used in the treatment of perianal fistulizing CD. Besides, patients were excluded from the study if they had received prior biologics following their initial fistula diagnosis, and all included patients were bio-naïve. Furthermore, treatment outcomes observed in this study were based on clinical, endoscopic and radiologic reports, which contributed to determine the efficacy of UST for the treatment of perianal fistulae. There are several limitations to this study. Firstly, it was a single-center study with a relatively small sample size, and the evidence from this retrospective study should be further validated in a larger sample size at multiple inflammatory bowel disease centers. In addition, we only evaluated the short-term efficacy of UST and IFX in the treatment of perianal fistula CD. The long-term outcomes of UST in the treatment of perianal fistula also need to be addressed. Furthermore, we regretfully discovered that there were no clear medical records of patients treated at our center regarding the use of setons. Therefore, we cannot precisely confirm whether each patient received seton treatment as part of their treatment plan. Finally, more potential biomarkers for UST efficacy prediction should be explored to ensure precise and individualized treatment in perianal fistula CD.
In conclusion, our results are encouraging and suggest UST may be a promising option to anti-TNFs for perianal fistulizing disease and with an excellent safety profile, but remission rates remain low. Further prospective studies with a larger patient population and long-term follow-up are needed to further establish the efficacy of UST as a first-line treatment for perianal fistula CD.
Notes
Funding Source
This work was supported by grants from National Natural Science Foundation of China (82100576).
Conflict of Interest
Linxin Liu is affiliated with Boji Medical Biotechnological. No other potential conflicts of interest relevant to this article were reported.
Data Availability Statement
Data will be made available upon reasonable request to corresponding author.
Author Contributions
Conceptualization: Zhuang X, Zeng Z. Methodology: Zhuang X, Zeng Z, Chen M (1st author), Chen Z, Lin J, Chen B, He Y, Chen M (9th author). Software: Chen M (1st), Chen Z, Lin J. Validation: Zhuang X, Chen M (1st). Investigation: Chen M (1st). Formal analysis: Chen M (1st). Data curation: all authors. Visualization: Liu L, Tu T, Li X. Writing-original draft: Zhuang X, Chen M (1st), Lin J. Writing-review & editing: all authors. Approval of final manuscript: all authors.