Everolimus for refractory Crohnʼs disease: A case report

To the Editor: Approximately three-quarters of patients with inflammatory bowel disease (IBD), Crohn's disease (CD) or ulcerative colitis (UC), will respond to conventional therapy, especially corticosteroids.1 Corticosteroid-refractory patients require treatment with infliximab and immunosuppressant therapies. The immunomodulators azathioprine and methotrexate have been largely used as maintenance treatment, with good results.2 More recently, new strategies for cytokine antagonist therapy have also been developed, including anti-TNF-alpha monoclonal antibody (mAb) therapy (infliximab).2 As the majority of patients will relapse, a long-term therapy with infliximab is necessary. Although episodic therapy can be used, the optimal strategy is systematic maintenance treatment. This results in a reduction of the rate of complications, hospitalizations, and surgeries associated with CD.3 Nevertheless, safety problems with the mAb infliximab treatment are mainly secondary to the formation of antibodies to infliximab, which may lead to infusion reactions and serum sickness-like delayed infusion reactions, leading to loss of response. In addition, some immunosuppressive agents that are routinely utilized in transplantation have been also applied as a “rescue” therapy for CD; for example, calcineurin inhibitors (CNIs) such as cyclosporin (CsA) and tacrolimus (Tac).4,5 Sirolimus and everolimus are new agents of the class of mTOR inhibitors that have been used very recently in organ transplantation. They block a growth-driven transduction signal (IL-2 signaling) in T-cell response to alloantigens.6 In addition, mTOR inhibitors have the major advantage of a much lower nephrotoxicity than CNIs,6 together with potential antitumor effects.7 Until now, this new class has not been evaluated in the field of IBD. We report here the case of a patient who was successfully treated by everolimus after multiple treatment failures or discontinuation, including infliximab, because of adverse side effects. A 19-year-old Caucasian female with a history of arthritis and iritis was diagnosed with probable UC in 1974 at our institution. Diagnosis was established from endoscopic, histological, and clinical criteria. The initial treatment consisted of salazopyrine. A steroid-refractory relapse with severe bleeding and colectasia required a coloproctectomy, with transient ileostomy in 1988. Due to iterative ileo-anal stenosis despite surgical resections, a lateral ileostomy was performed in 1992. Because of multiple perianal and peristomial fistulas, a terminal ileostomy was performed in 1995. Between 1988 and 1999 salazopyrine and low-dose oral steroids were maintained. In 1999 the patient was admitted for upper gastrointestinal hemorrhage with severe anemia. She had suffered from epigastric pain for several days. Endoscopic investigations revealed multiple gastric ulcers without lesions of ileon. Because of recurrent hemorrhage despite several endoscopic adrenaline injections, a surgical exploration was performed, revealing a gastric fistula with the splenic artery. Initial treatment was conservative. Four months later the patient required a subtotal gastrectomy with eso-jejunal anastomosis because of recurrent hemorrhage, due to the recurrence of gastric ulcers with rupture of the splenic artery. A histological analysis of the stomach was not conclusive, but the diagnosis of gastric involvement by CD was suspected. A new ileoscopy disclosed mild ulcerations of the terminal ileon. Histologically, the biopsy specimen of the ileal mucosa revealed marked active chronic inflammation with cryptitis, but no granuloma was detected. In addition to steroids, azathioprine (2.5 mg/kg) was started in 2000. Salazopyrine was replaced with mesalazine. Endoscopic follow-up showed remission between 2001 and 2003. In 2003, the patient was readmitted because of abdominal pain, related to recurrent ileitis. A treatment with local intra-ileal and oral steroids (0.5 mg/kg/d of prednisone) was ineffective and cyclosporine was introduced. The initial estimated glomerular filtration rate (GFR) was 70 mL/min. After 6 months cyclosporine was replaced with tacrolimus because of gingival hyperplasia. The evolution was favorable, without recurrence for 1 year. The treatment consisted of mesalazine (3 g per day), prednisone (15 mg per day), tacrolimus (4 mg per day with trough level at 4 ng/mL), and azathioprine (2.5 mg/kg/d). The level of 6-thioguanine (6-TGN) metabolites was above 280 pmol per 8 × 108 red blood cells from several measurements. In 2004 the GFR decreased to 50 mL/min and the dosage of tacrolimus was progressively reduced. Two months after tacrolimus cessation a new recurrence of ileitis, with abdominal pain, was confirmed by endoscopic examination. The patient was referred for infliximab infusion (5 mg/kg). During the next 2 weeks abdominal pain, nausea, and abdominal distension did not resolve. The patient received 3 courses of infliximab at 0, 2, and 4 weeks. Infliximab therapy was considered not effective. Tacrolimus was reintroduced and azathioprine was replaced by everolimus. Body weight was 38 kg. The white blood cell (WBC) count was 7,500/μL and the hemoglobin level was 10.2 g/dL. Electrolytes were normal except for mildly decreased potassium levels, at 3.2 meq/L. GFR was 45 mL/min. Albumin was 2.0 g/dL; AST and ALT were mildly elevated (AST = 52, N<45 IU/L and ALT = 49, N<45 IU/L); glucose was 133 mg/dL; and serum zinc, copper, and selenium levels were normal. A cholangio-MRI, performed because of a suspicion of concurrent primary sclerosing cholangitis, was considered normal. Over 18 months the treatment consisted of mesalazine (3 g/d), prednisone (15 mg/d), tacrolimus (4 mg/d with trough level at 4 ng/mL), and everolimus (4 mg/d with a trough level between 4 and 5.5 ng/mL). Tacrolimus dosage was progressively reduced until complete withdrawal in December 2006. Prednisone dosage was progressively reduced from 15 to 7.5 mg/d. After 12 months of follow-up, no recurrence occurred and a new ileoscopy was normal. The last GFR was 55 mL/min. Despite major progress in the past 10 years, ≈40%–60% of patients presenting with IBD do not benefit from the available treatments, suggesting a significant need for new effective therapies.8 We report here the promising results of everolimus, 1 of the agents from the new class of mTOR inhibitors, in a patient with refractory CD. Different classes of drug have been evaluated for the treatment of CD and many were used in the case of our patient. Together with a high failure rate, all these treatments are associated with numerous side effects that can reduce their long-term efficacy. First of all, steroids are the most effective,9 as reported in our patient, but are associated with unacceptable side effects during prolonged use. Aminosalicylates were used in our case during the whole course of the disease, but they probably play a minor role as maintenance therapy.10 Immunomodulators are considered the treatment of choice in case of steroid-resistant or -dependent CD and for maintenance of remission. The best evaluated is azathioprine (or 6-mercaptopurine). Several randomized control trials have undoubtedly demonstrated that azathioprine use is associated with a two-thirds remission rate (versus one-third for placebo).11,–15 It was used first in our patient but was not able to achieve long-term control of the disease. The therapeutic benefits of thiopurines correlate best with the concentration of active 6-TGN metabolites.16 Several reports have suggested that a therapeutic response can be maximized when patients achieve therapeutic 6-TGN levels. This must be measured in case of thiopurine-based treatment failure, as we did in our case. Methotrexate is an effective short-term treatment for CD and can be considered an alternative to azathioprine for maintaining remission.17 It was not used in our patient because of a preexisting mild elevation in serum transaminase levels. In our patient a favorable outcome was achieved with cyclosporine, and thereafter tacrolimus, both CNIs. The efficacy of this class has been demonstrated for refractory UC but has not been largely evaluated in CD.4 Tacrolimus inhibits the production of interleukin-2 by T-helper (Th) lymphocytes, like cyclosporine,18 but with an immunosuppressive function ≈100 times more potent than cyclosporine.19 In a recent study including patients with CD, the long-term efficacy and safety of tacrolimus were evaluated. Tacrolimus was administered orally, aiming for serum trough levels of 4–8 ng/mL. The majority of patients were receiving concomitant azathioprine. The mean treatment duration was more than 2 years. Steroids were reduced or discontinued in 90% of UC and CD patients taking steroids. Side effects included a transient rise of serum creatinine levels (7.6%), tremor or paresthesias (9.4%), hyperkaliemia (1.9%), hypertension (1.9%), and opportunistic infections (3.8%). In our case the impairment of renal function was the main reason for discontinuing therapy, despite its efficacy. In summary, these results emphasize that long-term tacrolimus therapy is effective in refractory IBD, but with a cumulative risk of life-threatening complications, which have been largely documented in organ-transplanted patients.20 During the past decade the major therapeutic progress in IBD was the development of biotherapies, especially anti-TNFα antibodies. The overall rate of remission with infliximab can be evaluated at 40% for remission induction and remission maintenance.21 Our case illustrates that a majority a patients do not benefit from this treatment. Certolizumab, adalimumab, and natalizumab have also been evaluated but were not available in our case. Until now, the efficacy of mTOR inhibitors has not been well documented in IBD. mTOR is located at a branch point of the signal 3 pathway for T-cell activation6. Like CNIs, sirolimus and everolimus bind to a cytoplasmic binding protein, FKBP (tacrolimus-binding protein). In contrast to CNIs, the siro(evero)limus-FKBP ligand does not block calcineurin, but rather engages mTOR. Therefore, sirolimus and everolimus reduce IL-2-dependent cell proliferation at the G1 to the S phase of the cell division cycle. In a recent animal study the therapeutic effect of everolimus was evaluated in interleukin-10 gene-deficient (IL-10−/−) mice. IL-10−/− mice is a well-characterized experimental model of CD, since they develop severe chronic colitis due to an aberrant Th1 immune response. Everolimus was administered orally for a period of 4 weeks to IL-10−/− mice with clinical signs of colitis. The administration of everolimus resulted in a significant decrease in the severity of colitis, together with a significant reduction in the number of CD4+ T cells in the colonic lamina propria as well as IFN-γ production in colonic lymphocytes. These preliminary results are promising and must be confirmed in a clinical setting. In conclusion, this first case of refractory CD successfully treated by long-term everolimus could be the first step of a further evaluation of mTOR inhibitors in IBD. This could represent a new therapeutic alternative, with specific targets in this field.