Biologic Therapy for Budesonide-refractory, -dependent or -intolerant Microscopic Colitis: a Multicentre Cohort Study from the GETAID

Abstract

Background

Budesonide remains the backbone therapy for microscopic colitis [MC]; however, relapses are frequent, and some patients are intolerant or dependent. Anti-TNF therapy is increasingly used to treat these patients, but available evidence is still limited. The aim of this study was to evaluate the effectiveness and safety of anti-TNF therapy in MC patients failing budesonide.

Methods

In a multicentre retrospective cohort study, budesonide-refractory, -dependent, or -intolerant MC patients treated with anti-TNF agents were included. Clinical remission was defined as fewer than three bowel movements per day, and clinical response was defined as an improvement in stool frequency of at least 50%.

Results

Fourteen patients were included. Median age was 58.5 years, median disease duration was 25 months, and median follow-up was 29.5 months. Seven patients were treated with infliximab [IFX], and seven with adalimumab. Clinical remission without steroids at 12 weeks was reached in 5/14 [35.7%] patients; all of these received IFX. Clinical response at 12 and 52 weeks, was obtained in 9/14 [64.3%] and 7/14 [50%] patients, respectively. Five patients switched to another anti-TNF agent. When considering both first- and second-line anti-TNF therapies, 7 [50%] patients were in clinical remission at Week 52. Mild to moderate adverse events were reported in six ptients. Two patients were treated with vedolizumab, of whom one had clinical response; one patient treated with ustekinumab had no response.

Conclusions

This is the first multicentre cohort study showing that half of patients treated with anti-TNF therapy for MC achieved clinical remission in case of budesonide failure.

1. Introduction

Microscopic colitis [MC], including collagenous colitis [CC] and lymphocytic colitis [LC], has emerged as a common cause of chronic diarrhoea. Nowadays, MC is diagnosed in up to 10% of patients undergoing colonoscopy for chronic non-bloody diarrhoea,¹ with a rising incidence and prevalence largely attributed to increased awareness of the disease.^2,3 MC is mainly observed in middle-aged female patients, and is more frequent in smokers and in patients with other auto-immune disorders [e.g., coeliac disease, thyroiditis].^4,5 The aetiology remains unclear but is likely to be multifactorial, involving specific mucosal responses to as yet unidentified luminal agents in predisposed individuals, resulting in an uncontrolled mucosal immune response involving Th1 and Th17 mucosal cytokine profiles.^6,7 The pathogenesis of MC has also been associated with certain drugs [non-steroidal anti-inflammatory drugs, proton pump inhibitors, and selective serotonin reuptake inhibitors], environmental factors [smoking],^1,5 and microbial dysbiosis.⁸ The diagnosis is based on both a clinical history of diarrhoea⁹ and specific histological findings.^10,11 LC is defined as an increase in the number of intraepithelial lymphocytes ≥20%, whereas the diagnosis of CC is based on a thickened subepithelial collagen layer ≥10 µm; both LC and CC exhibit lamina propria inflammation and epithelial damage.^10,11

No curative therapy currently exists, and treatment largely depends on the severity of symptoms. The primary goals are to achieve and maintain clinical remission and to improve patients’ quality of life.^4,12 It is currently unknown whether histological remission is an important goal or not. The search for and withdrawal of a potentially suspect medication is the first step of management. Empirically, antidiarrhoeal agents such as loperamide and cholestyramine can be used for mild disease.^4,13 The best-documented treatment, budesonide, is the treatment of choice for moderate to severe MC.^13–17 However, relapse rates remain high [61–80%] even after maintenance therapy,^16,18,19 and many patients are refractory, dependent, or intolerant. In refractory MC, immunomodulators such as thiopurines and methotrexate can be considered, despite contradictory, retrospective data.^20,21

Anti-TNF agents have radically changed the natural history and management of inflammatory bowel diseases [IBD].^22,23 In severe refractory MC [0.6–6%],²⁴ anti-TNF agents may represent an alternative therapy, given their fast onset of action. However, only 43 refractory MC patients [from nine case series and also one prospective study on three patients] have been reported in total in the literature.^24–33

In order to extend our knowledge of anti-TNF therapy in this setting, we aimed to evaluate the effectiveness and safety of anti-TNF therapy in patients with budesonide-refractory, -dependent, or -intolerant MC, in centres affiliated to the French Groupe d’étude thérapeutique des affections inflammatoires du tube digestif [GETAID].

2. Methods

1.1. Study design and patients

We carried out a retrospective, multicentre, cohort study of all MC patients who were budesonide-refractory, -intolerant, or -dependent and who were treated with anti-TNF agents in GETAID centres. Adult patients [≥18 years old] were eligible if they met the following criteria: histologically established diagnosis of MC, CC, or LC; a history of clinically active disease [with a mean of ≥3 stools per day or a mean of ≥1 watery stool/day over 1 week]; refractory to, intolerant of, or dependent on conventional treatments; treated with anti-TNF therapies, prescribed at the investigator’s discretion. Patients were not eligible if they were suffering from concomitant inflammatory bowel disease.

Budesonide intolerance was defined as the presence of side effects necessitating cessation of the drug; budesonide refractoriness was defined as inadequate control of diarrhoea despite an appropriate course of treatment; and budesonide dependence was defined as the recurrence of diarrhoea after an initial discontinuation or reduction.

2.1. Study endpoints

The primary outcome was the proportion of patients achieving clinical remission without steroids after 12 weeks of anti-TNF therapy. Clinical remission was defined as fewer than three stools daily over 1 week [adapted from Hjortswang et al.^12]. Clinical response [or partial response] was defined as an improvement in stool frequency ≥50%.

The secondary outcomes were: clinical response at Week 12; clinical response and remission [with and without steroids] at Weeks 24 and 52 and at the end of follow-up; histological response and remission; corticosteroid sparing effect; safety; and tolerance. Histological remission was defined as normalisation of intraepithelial lymphocyte count for LC and/or of subepithelial collagen band size for CC; and histological response was defined as a reduction in inflammatory abnormalities or in subepithelial collagen layer thickening, but without normalisation, according to the appreciation of the local histopathologist compared with pre-biotherapy histological findings.

2.2. Data collection

All French GETAID centres were surveyed to enquire about cases of MC that were refractory to, intolerant of, or dependent on budesonide and treated with TNFα antagonists from October 2018 to February 2019.

Demographic and clinical data and outcomes were collected for all patients from the medical records. Patient and disease characteristics, including sex, age, smoking status, type of MC, disease duration, and other inflammatory comorbidities, were recorded. Exposure to previous treatments for MC, their efficacy, and cause of cessation were also collected. The number of daily bowel movements, as well as the physician’s assessment, before treatment and 12, 24, and 52 weeks after treatment initiation were obtained. Adverse events were recorded and histological analyses, when available, were collected. The study was conducted in compliance with the French reference methodology, MR-004, and was supervised by and registered in the GETAID [registration number 2210131].

2.3. Statistical analyses

Statistics are descriptive, with continuous data presented as mean ± standard deviation [SD] or median [interquartile range] [[IQR]], and categorical data as number and percentage. Due to the small number of patients in this study, only descriptive statistics are given. Statistics were calculated using Microsoft^® Excel^® Office 365 MSO Software.

3. Results

3.1. Population characteristics

A total of 14 patients [nine CC and five LC] met the inclusion criteria in five French GETAID centres; 11/14 [78.6%] were female; median [IQR] age at anti-TNF therapy initiation was 58.5 [46–65] years old; six [42.9%] patients had other immune disorders [Table 1]. The median disease duration before anti-TNFα initiation was 25 [16–154] months, and the mean number of daily bowel movements was 7.6 [± 3.3]. The median follow-up time was 29.5 [18–38] months. The characteristics of the population are summarised in Table 1. Regarding previous treatments, all patients were treated with budesonide, with a median treatment duration of 10 [7–24] months of continuous treatment before anti-TNF introduction. All MC patients had failed budesonide: 7/14 [50%] were refractory, 4/14 [28.6%] were dependent, and 2/14 [14.3%] were intolerant; data were missing for one patient [Table 1]. Three patients [21.4%] had previously been treated with immunosuppressants and all of them were intolerant [Table 1].

<<Tables 1 and 2 near here>>

Seven patients received infliximab [IFX] and seven received adalimumab [ADA] [Table 2]. Four patients in the ADA group were treated concomitantly with budesonide or systemic corticosteroids, and none in the IFX group. One patient received IFX and azathioprine in combination. For all patients, the anti-TNF induction regimen was the same as in IBD indications and was followed by a maintenance regimen in all but one patient [Table 2]. The median duration of the first anti-TNF therapy was 10 [6–27] months.

3.2. Efficacy outcomes

Overall, clinical remission without steroids was obtained in five [35.7%] patients at 12 weeks: all of them were in the IFX group [71.4%] [Figure 1]. Clinical remission without corticosteroids was observed in five patients at Weeks 24 and 52. In the IFX group, four [57.1%] patients had clinical remission without steroids at Week 52. In the ADA group, two patients stopped corticosteroids at Weeks 4 and 12, respectively, and only one patient [who had stopped corticosteroids at Week 12] achieved clinical remission without steroids at Week 52 [14.3%] [Figure 1].

Clinical response was observed in nine [64.3%] patients at Week 12 and in 7/14 [50%] at Week 52 [Figure 2]. In patients treated with IFX, 6/7 patients achieved clinical response at Week 12 and 5/7 at Week 52. In patients treated with ADA, 3/7 patients achieved clinical response at Week 12 and 2/7 at Week 52, including one patient who was still on budesonide [Figure 2]. Overall, 5/14 [35.7%] patients had a primary treatment failure [one patient treated with IFX and four with ADA]. In four patients who had a concomitant anti-TNF indication for another inflammatory disease, a 100% response rate was observed for the other inflammatory disease.

Histological data from colon biopsies were available in 9/14 patients during follow-up, namely in five patients treated with IFX [four in clinical remission and one in clinical response] and in four with ADA [one in clinical remission, one in clinical response, and two in clinical failure]. A histological response was observed in 5/9 [55.5%] patients and histological remission was observed in only one patient who received IFX combined with azathioprine. None of the patients with clinical failure achieved histological response. Conversely, all patients in clinical remission had histological response [Supplementary Figure 1].

3.3. Follow-up

Follow-up data are summarised in Figure 3. The median follow-up time from anti-TNF introduction was 29.5 [18–38] months. Anti-TNF treatment optimisation occurred in six [42.9%] patients: in four ADA patients for MC due to efficacy issues and in two IFX patients for reasons other than MC [i.e., symptomatic ankylosing spondylitis in one patient and joint pain in the other one]. Optimisation was effective in the two patients on IFX and in 2/4 patients on ADA.

First-line anti-TNF therapy was discontinued in 11 [78.6%] patients during follow-up: 4/7 IFX patients and in all seven ADA patients [Supplementary Table 1, Figure 3]. The median duration of the first anti-TNF treatment was 10 [6–27] months. The cause of cessation in the four patients treated with IFX was: patient’s choice for two patients in clinical remission, primary failure for one, and intolerance with allergic reaction for one who relapsed after discontinuation. In the seven patients treated with ADA, the cause of cessation was primary failure in four patients [including one with intolerance] and secondary loss of response in three patients.

Seven patients [50%] were treated with a second-line biological therapy [treatment lines are summarised in Figure 3]. Five patients switched to another anti-TNF and in two patients, ADA was swapped for another biologic agent: one patient received vedolizumab [with success] and one received ustekinumab [with failure]. Three patients achieved clinical response with a second-line anti-TNF at Week 12, and two patients at the end of follow-up. In 4/5 [80%], patients, the second-line anti-TNF therapy was discontinued; two because of primary failure and two because of remission. The two patients who did not respond to a second line of biologic treatment were treated with a third line: one response was observed with golimumab and one failure with vedolizumab. Considering all treatment lines together, 12/14 patients were in clinical response at the end of follow-up [Figure 3].

Overall in our cohort, two patients were treated with vedolizumab: one in second line with clinical response, and another one in third line with primary failure. In the only patient treated with ustekinumab, as second-line therapy, failure was both for the MC and for the rheumatological disease; this patient was treated in third line with golimumab, with clinical success [Figure 3].

3.4. Safety

Mild to moderate adverse events occurred in 6/14 patients [42.9%] during follow-up: four with ADA and two with IFX, including only one moderate adverse event in the IFX group [anaphylactoid reaction of moderate severity with the presence of anti-IFX antibodies] [Supplementary Table 2]. Mild adverse events included repeated sinusitis in one patient treated with ADA, mild psoriasis in two patients treated with ADA, arthralgia in one patient treated with IFX, and intolerance with drug fever, asthenia, and pain in a patient treated with ADA. Three patients discontinued anti-TNFα due to side effects, one in the IFX group and two in the ADA group.

4. Discussion

MC is a benign but debilitating disease leading to a poor quality of life. In general, patients respond well to budesonide, which is the best evidence-based treatment and represents the treatment of choice,⁴ with up to 80% of clinical remission after induction therapy.³⁴ However in a recent European multicentre prospective registry, 49% of patients had a chronic active or relapsing disease course.¹⁹ Thus, refractory MC is a challenging situation and existing data are sparse regarding the most suitable second-line options. In our multicentre registry, we report 14 MC patients with IFX or ADA treatment initiated after budesonide failure; 36 % were in clinical remission at Week 12, 50% were in clinical response at 1 year, and 83% were in clinical response, when considering all treatment lines together at the end of follow-up.

Theoretically TNFα antagonists are an appropriate therapeutic option, given that the immunopathology of MC clearly demonstrates a T helper cell type 1 and 17 mucosal cytokine profile.^6,7,35,36 Nevertheless, data on clinical outcomes after these biologic agents are scarce in the literature to date. Although small, our series of cases collected from French centres dedicated to the management of IBDs is the second largest case series reported so far in the literature and is the first multicentre study. The favourable results we observed in patients with anti-TNFs are consistent with previously reported short case series [Table 3]. Ten case reports had previously reported a total of 43 MC budesonide-refractory patients treated with anti-TNFα,^24–33 with response rates ranging from 70% to 80% for case series reporting more than two patients. Although mainly retrospective, there is a growing body of evidence in favour of this treatment option in case of budesonide failure [refractoriness, intolerance, high dose-dependence]. Eight weeks of budesonide induction treatment are usually recommended in MC.⁴ In our cohort, which reflects the situation observed in real life, patients with budesonide failure had received treatment continuously for a median duration of 10 [7–24] months before anti-TNF introduction, which is actually very long for an inadequate treatment, emphasising the need for new validated strategies.

Histological assessment, available for the first time in nine patients in our series, yielded substantial evidence of anti-TNF efficacy on mucosal damage and could be used in an IBD-like way to target strategy. Unfortunately, faecal calprotectin was not assessed in our study. Finally, reported anti-TNF tolerance in this patient population was in accordance with what is already known about this drug class, and appears quite acceptable regarding the benefit of the biologic therapy to these patients’ quality of life. Ultimately, cost and availability must be considered. In this regard, biosimilars of infliximab and adalimumab are potential game changers for this indication.

In our cohort, we observed a striking difference in clinical remission rates between IFX and ADA [71% vs 0% clinical remission rates at Week 12, respectively]. In the cohort reported by Daferera et al., the difference was lower [63% vs 40% clinical remission rates, respectively].³¹ A difference between these two molecules has also been observed in ulcerative colitis, an inflammatory bowel disease that shares immunological pathways with MC and in which IFX outranks ADA for clinical remission induction in network metanalyses.^37,38 This remains to be further investigated in dedicated studies. As in IBD, an initial combination of immunosuppressants with IFX should be appropriate, as the only patient in deep remission received a combination therapy in our series and loss of response due to antibodies against the biologic agent was observed in the Swedish study.³¹

Finally, other biologics were used by investigators in our cohort during the follow-up of their patients: vedolizumab was successful in 1/2 patients in second-line therapy. In a series of highly refractory MC patients, vedolizumab showed efficacy in 5/11 subjects.³⁹ Ustekinumab failed in one patient from our cohort. No other data are available on anti-IL12–23 in MC, with the result that no conclusions can be drawn.

This study has some limitations, notably its retrospective design and the small sample size. Moreover, health-related quality of life was not evaluated with specific tools. Nonetheless, it has been shown that patients with a mean of three or more daily stools or a mean of one or more watery stools per day have a significantly impaired health-related quality of life.¹² Assessment of daily bowel movements was systematically obtained and used for clinical endpoints in our study, to limit evaluation bias and as a surrogate marker of quality of life.¹² However, stool consistency was not adequately reported in most patient files, and therefore we chose to limit the definition of remission in our study to les than three stools/day [without the component ‘and less than one watery stool/day’ from the definition by Hjortswang et al.^12]. Unfortunately, in the five centres including patients, the total number of MC patients seen over the study period was unknown, preventing us from drawing conclusions about the proportion of patients requiring biologic therapy.

To the best of our knowledge, this is the largest multicentre study available to date, with 14 cases of MC sharing the condition of refractoriness, dependence, or intolerance to budesonide. Moreover, we assessed histological efficacy in the majority of patients. Our study adds to the growing body of evidence regarding anti-TNF efficacy in MC, although prospective and controlled data are lacking. Since further, larger, prospective, randomised, controlled trials to confirm these results are unlikely to be performed for this relatively rare condition, case series such as this one, although retrospective, are essential to guide decision making.

In conclusion, anti-TNF agents, especially infliximab, should be considered in routine practice in case of budesonide-refractory, -dependent, or -intolerant MC. Their safety is similar to that observed in IBD patients.

Funding

None.

Conflict of Interest

CZ has received fees from Abbvie, Takeda, Janssen, and Pfizer. FZ has received lecture fees from Janssen. LP has received lecture fees from Abbvie, MSD, Takeda, Janssen, and Pfizer. AA has received consulting fees from Abbvie, Hospira, Takeda, Gilead, and Biocodex as well as lecture fees and travel accommodation from Abbvie, Janssen, Biocodex, Hospira, Ferring, Takeda, and MSD; and advisory board fees from Gilead, Takeda, and Abbvie. SK has received fees from Abbvie, Mayoli, and Gilead. LP-B has received personal fees from Galapagos, AbbVie, Janssen, Genentech, Ferring, Tillots, Celltrion, Takeda, Pfizer, Index Pharmaceuticals, Sandoz, Celgene, Biogen, Samsung Bioepis, Inotrem, Allergan, MSD, Roche, Arena, Gilead, Amgen, BMS, Vifor, Norgine, Mylan, Lilly, Fresenius Kabi, OSE Immunotherapeutics, Enthera, Theravance, Pandion Therapeutics, Gossamer Bio, Viatris, Thermo Fisher. LV has received fees from Abbvie, Amgen, Celltrion, Galapagos, Gilead, Ferring, Mayoli, Mylan, MSD, Pfizer, Janssen, and Takeda.

Author Contributions

GB was involved in the concept of the study, data collection and interpretation, and drafting and revision of the manuscript. LV conceived of the study and supervised the interpretation of the data; she was involved in the drafting of the manuscript and its revision. CZ, FZ, LP, SK, AA, and LB recruited patients and recorded data. All authors contributed to the critical revision and the final approval of the manuscript.

References