Ecacy and Safety of Thalidomide in Crohn’s Disease Patients with Perianal Fistula and Abscess

Medical therapies of parianal Crohn’s disease (CD) are limited. Thalidomide is an effective medical therapy to alleviate disease activity of CD. However, the effects and safety of thalidomide in the treatment of perianal stula and abscess was not evaluated. This retrospective cohort study was performed at a tertiary referral centre and recruited 73 patients with perianal CD who received thalidomide (50–100 mg) daily for 1 year. Data collected included demographics, medications, and disease behaviour. Clinical assessment of CD was conducted using the Crohn’s Disease Activity Index (CDAI), and perianal lesions were evaluated using the Fistula Drainage Assessment index and Perianal Disease Activity Index (PDAI). At the same time, the occurrence of adverse effects was recorded during treatment. Wilcoxon's signed-rank test and Student’s t-test were used to analyse the data. resonance birth severe human immunodeciency virus; ongoing major uncontrolled major transplanted organs; participation other and All


Introduction
Perianal Crohn's disease (PCD), which plays a major negative role in the quality of life of patients, is de ned as in ammation at or near the anus, including tags, ssures, stulae, abscesses, or stenosis. The reported incidence of PCD varies from 3-80%, depending on the location of the involvement of the gastrointestinal tract [1]. Over 50% of rectal and colonic Crohn's disease (CD) patients are associated with perianal complications [2]. Such complications are a marker of more severe disease activity, and are associated with multiple surgical interventions and frequent relapses. Treatment of parianal complications is very challenging and usually requires a combination of medical and surgical interventions.
With regard to medical management of PCD, the role of antibiotics, immunosuppressive drugs (e.g. thiopurines and oral tacrolimus), and biological agents in the management of PCD have been reported with variable success rates when used as single agents or in combination [2]. It was reported that there was no signi cant bene t of using aminosalicylates for the treatment of luminal CD [3], and there were no speci c trials in PCD. Although corticosteroids are sometimes used in combination with other drugs to treat PCD, their contribution is limited [4,5]. Several studies have reported that antibiotics may improve symptoms and contribute to the healing of stulas [2]. Multiple studies and randomised controlled trials showed that anti-TNF alpha treatments, including in iximab and adalimumab, and other biologics such as vedolizumab, and certolizumab, are superior to placebo in induction and maintenance therapy for perianal stulas in CD [2,[6][7][8][9]. However, the percentage of patients who do not achieve improvement or closure is high partly due to the development of antibodies against these agents that can result in a loss of clinical response. Besides, anti-TNF agents have been associated with opportunistic infections, serum sickness-like reactions, autoimmune disorders, and sepsis. The high cost of anti-TNF agents also limits their clinical application, especially in developing countries.
Thalidomide (a-N-phthalimidoglutarimide) was originally used over 40 years ago as a sedative and removed from the market because of its teratogenic effects. However, it has been used as an inexpensive anti-in ammatory immunosuppressant in many immune diseases more recently [10]. The e cacy of thalidomide in the treatment of patients with PCD has been reported. It acts by inhibiting TNF-a, interleukin-12, interferon-γ, and nuclear factor kappa B activation [11]. In a retrospective study, thalidomide was effective in CD patients with perianal stula [9]. In an open-label study, it was con rmed that thalidomide plays an important role in the closure of the stula and the reduction of drainage stula in CD patients with perianal stula [12].
Although a few small trials have been conducted to explore thalidomide for PCD, the safety and e cacy of thalidomide for PCD remain to be evaluated. Our retrospective study aimed to evaluate the e cacy and safety of thalidomide in the treatment of perianal stulizing CD and CD with perianal abscess.

Ethics statement
The study protocol was approved by the Clinical Research Ethics Committee of The First A liated Hospital of Sun Yat-sen University (Application ID: [2015] 103).

Study design and patients
This was a retrospective observational cohort study of patients with an established diagnosis of CD at the Gastroenterology Department of the First A liated Hospital, Sun Yat-sen University (China). Patients with PCD treated with thalidomide who had complete clinical data between June 2008 and May 2018 were screened. CD diagnosis was based on a combination of clinical, radiologic, endoscopic, and histologic ndings. The inclusion criteria were as follows. (1) After treatment with biologics, azathioprine, and/or methotrexate for more than 3 months, the perianal disease was still active. (2) The available perianal examination included an anal examination, perianal ultrasonography, perianal magnetic resonance imaging, and endoscopic ultrasonography. (3) The patient would provide informed consent in their case notes before treatment. The exclusion criteria were concomitant biologics during thalidomide treatment or within 3 months before a clinical assessment; birth plan; tumours; severe peripheral neuritis; human immunode ciency virus; ongoing major infections; uncontrolled major diseases; transplanted organs; participation in other experimental studies; and patients with incomplete data. Informed consent obtained from all participants, and the informed consent of participants under 16 years old was obtained from the parent and/or legal guardian of the participant. All methods were carried out in accordance with relevant guidelines and regulations.

Treatment
Thalidomide (Changzhou Pharmaceutical Factory, China) was started at a dose of 50 mg per night orally, monthly evaluation of clinical response and laboratory assessments were performed, and for patients whose responded insu ciently without adverse events, the dose of thalidomide was increased to 100 mg per night. The dose of thalidomide was reduced gradually, discontinued, or withheld until the adverse effect was resolved according to the nature and severity of any adverse reactions. The medication could then be reintroduced at a lower dose at the discretion of the doctor, and azathioprine, 6-mercaptopurine, or methotrexate could be maintained during thalidomide treatment. The dose modi cations during the treatment period, duration of treatment, and reasons for withdrawal were recorded in the case notes.Some patients received corticosteroids before thalidomide was started. The e cacy of thalidomide was de ned as ine cient for patients with PCD if the patient's symptoms occurred repeatedly during corticosteroid reduction, withdrawal, or increase. Antibiotics were used as appropriate when the perianal abscess was too small to allow incision drainage or inadequate drainage and when perianal stula was complicated with infection.

Clinical evaluation
Patients underwent clinical and laboratory evaluations at baseline and every 4-6 weeks thereafter until 24 months, for the assessment of therapeutic effect and adverse events. Laboratory assessments included blood routine examination, platelet counts (PLT), albumin and C-reactive protein (CRP) levels, erythrocyte sedimentation rate (ESR), total protein level, and electrolytes. Clinical activity was measured using the Crohn Disease Activity Index (CDAI). A CDAI score < 150 was de ned as clinical remission, and clinical recurrence was further divided into mildly active (CDAI scores 150-220), moderately active (CDAI scores 221-450), and severely active (CDAI scores > 450). In addition, the Perianal Disease Activity Index (PDAI) and stula drainage assessment were used to evaluate the clinical activity of perianal disease and its curative effect, and their reliability has been widely accepted and utilised as a means of quantifying perianal disease severity [13,14]. PDAI ranged from 0 to 20, with higher scores indicating more severe disease [13]. In this study, clinical response (i.e. a reduction of 50% or more from baseline in the number of draining stulas observed) and remission (i.e. the stula was not drained despite gentle nger compression) were de ned after two consecutive study visits [15]. In addition, the disappearance of the perianal abscess on imaging was considered to indicate treatment e cacy. Recurrence of the perianal abscess, i.e. the perianal abscess diffused or shrank without disappearing after treatment, indicated treatment ine cacy.

Statistical analysis
Statistical analysis was performed using the dedicated statistical software SPSS version 20.0 (IBM Corp.). Quantitative parameters are expressed as mean ± standard deviation, and paired comparisons before and after the treatment were tested by Wilcoxon's signed-rank test. Continuous data of normal distribution were analysed using Student's t-test. Categorical data were analysed using the chi-square test. All signi cance tests were two-sided, and P < 0.05 was considered statistically signi cant.

Demographic and clinical characteristics
Ninety-three patients with PCD were evaluated between June 2008 and May 2018. Fifteen patients were lost to follow-up, and 4 patients had to be withdrawn because of the occurrence of serious adverse effects. Finally, 73 patients, including 39 patients with perianal stula and 34 patients with perianal abscess, were followed up for one year (Fig. 1). The demographic and clinical characteristics of the patients are shown in Table 1. Among 39 patients with perianal stula, 29 (74.36%) were combined with exudation when thalidomide treatment was started. Among 34 patients with perianal abscess before thalidomide treatment, 32 patients (94.12%) presented with symptoms of perianal local redness, swelling, and pain. The number of patients in clinical remission was greater than that before treatment (Table 2), CRP, PLT, ESR, and CDAI values were signi cantly lower after treatment than before treatment as well (all P < 0.01) ( Table 3).    Table 5. Furthermore, corticosteroids could be reduced to withdrawal in 6 patients (8.22%) who required additional corticosteroids.
Azathioprine was the most frequently used drug in combination with other medications to treat patients with perianal stula and abscess in this study (Table 5). Twenty-ve of 73 patients (34.24%) were treated with additional thalidomide when azathioprine monotherapy failed. The proportion of those who responded to treatment increased to 84.0% after treatment with only azathioprine combined with thalidomide (P < 0.01). We also compared the e cacy of thalidomide monotherapy and thalidomide combined with azathioprine in these patients. The rates of responsive patients were similar between the thalidomide monotherapy group and thalidomide combined with azathioprine group (72.73% [8/11] and 84% [21/25], respectively; P = 0.65).

Adverse Events
The adverse events that occurred during the period of treatment are shown in

Discussion
In this study, thalidomide was found to be bene cial in two-thirds of all patients with perianal stula and abscess. CRP, PLT, ESR, and CDAI values of patients were signi cantly lower after treatment than before treatment. Thalidomide is an effective medical therapy to alleviate the disease activity of CD, and it could improve the clinical symptoms of patients with perianal lesions.
Although previous experience with thalidomide therapy was relatively limited in patients with PCD, the results in this study were generally consistent with the conclusions of previous studies. As early as 1999, two open-label pilot studies of thalidomide evaluated the use of thalidomide for treating stulizing CD. The ndings suggested that 83% (5/6) of patients with perianal stula noted improvement after the rst four weeks of thalidomide treatment, and three of four patients who completed 12 weeks of therapy showed improvement [16]. The results of another trial indicated that two patients with stulas (13%) had complete closure of all stulas after four weeks of treatment [10]. A total of 69% of the remaining patients with stulas were responders, and 46% were in clinical remission at week 12 follow-up [10]. A case report described two patients (100%) with PCD who had stulas that stopped draining after thalidomide treatment [17]. The ndings of a subsequent study reported that 27% of patients had cessation of drainage, and perianal responses were found in 55% of patients, demonstrating that thalidomide was effective in inducing a clinical response and remission of stulizing CD [18]. However, another study implied that thalidomide seemed to be less effective in patients with stulas in light of a 50% (3/6) relapse rate [12]. Admittedly, fewer patients with PCD were included in these studies. A considerable number of cases would be needed to observe the e cacy of thalidomide in patients with PCD. We also considered whether the e cacy of thalidomide was affected by the abscess with drainage or without drainage. The results showed that the PDAI of patients with perianal abscess was signi cantly lower after thalidomide treatment, regardless of whether the perianal abscess had been surgically intervened. Thus, low-dose thalidomide could improve patients with perianal stulas and abscesses. The guidelines for the medical treatment of PCD also suggested that the e cacy of thalidomide for PCD was good [9].
Several open-label studies have con rmed that thalidomide is effective in inducing and maintaining clinical remission and mucosal healing in children, adolescents, and adults with refractory CD [19][20][21][22][23][24][25]. It was veri ed that thalidomide has anti-in ammatory, immunomodulatory, and angiogenesis inhibitory properties in animal studies [26]. Thalidomide can regulate macrophages and monocytes, resulting in reduced TNF-α production [27]. Thalidomide was effective for treating CD because it increased the degradation of messenger RNA encoding TNF-α, leading to inhibition of TNF-α production [28]. Bauditz et al. [29] indicated that thalidomide played a role in patients with CD by reducing the production of interleukin-12 and TNF-α. A study suggested that thalidomide may inhibit the effect of T cells and TNF-α, which could promote the clinical remission of intestinal symptoms in patients with IL10RA de ciency [30]. Besides, thalidomide may affect other immune cells and in ammatory cytokines to improve CD [12].
Perianal stula and abscess are common complications of CD, which lead to poor quality of life in patients. The risk of perianal stula seems to depend on the location and activity of the disease. The active disease was located in the rectum, while the risk of developing a perianal stula was high [31]. Combination regimens for treating perianal stulas in CD are often adopted by clinicians for better improvement and maintenance of disease remission. In this study, patients with combination regimens were not excluded, and patients with thalidomide in combination with other drugs accounted for 89% of the total (Table 6). Compared with other combination regimens, 25 of 73 patients (34.25%) accepted thalidomide combined with azathioprine, and this was the most common treatment in the study. However, thalidomide was added when azathioprine was no longer effective. Although the response rate of patients with PCD to azathioprine was low, azathioprine may have a bene cial role in combination treatment [9]. Few studies have explored the use of azathioprine in combination with other drugs for PCD. Thus, we compared the e cacy of thalidomide alone and thalidomide combined with azathioprine in the treatment of PCD patients with perianal stula and abscess. The rates of responsive patients were similar between the thalidomide group and thalidomide combined with azathioprine group, indicating that both thalidomide and a combination of thalidomide and azathioprine were effective for treating PCD. It has been reported that patients with PCD can obtain more improvement when patients accept azathioprine in combination with antibiotics or in iximab [2]. To date, there have been no speci c and comparatively large studies designed to investigate the e cacy of thalidomide monotherapy or combination regimens for PCD. Therefore, more studies are needed to con rm the e cacy of thalidomide monotherapy or combination regimens.
Although thalidomide is useful for treating PCD, its toxicity and side effects are the major problems that clinicians have focussed on. There are quite a few side effects of thalidomide, including neuropathy, rash, and sedation. A previous study showed that 34% of patients discontinued thalidomide because of side effects at one year and 46% at two years [20]. The percentage for withdrawal due to adverse events was 35.5% of patients in a recent study [22]. In He et al.'s study [19], the percentage of thalidomide withdrawal due to severe adverse reactions was 10.6% at three months. It was demonstrated that 29% of all patients discontinued thalidomide because of neuropathy [32]. In our study, 31% of patients experienced adverse events, and 15 patients needed surgical intervention. A total of 5.19% of patients discontinued treatment because of the adverse event. The incidence of adverse events was congruent with that reported in previous studies. Discontinuation rates were relatively low, probably because a lower dose (100 mg/day) of thalidomide was used in our study compared with that in some of the previous studies and all adverse events were relieved after the intervention.
This study is probably the original, specialised, and largest retrospective study to evaluate the effect of thalidomide and thalidomide combined with azathioprine for PCD patients with perianal stula and abscess. However, it has certain limitations. First, the bene t of thalidomide over placebo could not be contrasted without a non-controlled study. Second, the e cacy of thalidomide in combination with other drugs, except for azathioprine, was not been studied. The effects of some factors, such as treatment course, lesion regions, sex, and smoking status, were not considered when the e cacy of thalidomide was investigated. Lastly, some inevitable problems including recall bias and data loss occurred because of the lengthy duration of data collection.

Conclusions
This retrospective study's results demonstrated that thalidomide is effective in inducing clinical remission and response in CD patients with perianal stula and abscess. Thalidomide in combination with azathioprine was also effective in these patients which warrants further evaluation. Thalidomide played a role in patients with PCD, and its toxicity and side effects needed to be assessed frequently during treatment follow-up. Thalidomie may be e cacious in patients with paranal CD which and needs further evaluation in dedicated prospective clinical studies.
Declarations article. Neither the entire paper nor any part of its content has been published or has been accepted elsewhere.

Funding
This work was supported by National Natural Science Foundation of China (NSFC grant No.81870384, 81630018, 81670607). There is no disclosure of nancial arrangements related to the research or assistance with manuscript preparation.
Availability of data and materials All data and materials are not available in this study, and are available from the corresponding author on reasonable requests.
Ethics approval and consent to participate The study protocol was approved by the Clinical Research Ethics Committee of The First A liated Hospital of Sun Yat-sen University. Informed consent obtained from all participants, and the Informed consent of participants under 16 years old was obtained from the parent and/or legal guardian of the participant. All methods were carried out in accordance with relevant guidelines and regulations.

Consent for publication
Not applicable.

Competing interests
There are no con icts of interest to declare.