TY - JOUR
AU1 - MD, Mark E. Gerich,
AU2 - MD, Darrell S. Pardi,
AU3 - MD, David H. Bruining,
AU4 - CCRP, Patricia P. Kammer,
AU5 - Becker, Brenda D.
AU6 - MD, William T. Tremaine,
AB - Background Several small retrospective studies have reported encouraging response rates in patients with Crohn’s disease (CD) treated with tacrolimus. Methods We conducted a retrospective study of the use of oral tacrolimus for severe CD refractory to anti–tumor necrosis factor agents. Response was defined as a clinician’s assessment of improvement after at least 7 days of treatment of one or more of the following: bowel movement frequency, fistula output, rectal bleeding, abdominal pain, extraintestinal manifestations, or well-being. Remission required all of the following: <3 stools per day, no bleeding, abdominal pain or extraintestinal manifestations, and increased well-being. Results Twenty-four eligible patients were treated with tacrolimus for a median of 4 months. Approximately 37% were steroid dependent or steroid refractory. Response and steroid-free remission rates were 67% and 21%, respectively, and lasted for a median of 4 months. Approximately 42% of patients were able to stop steroids and 54% of patients ultimately required surgery within a median of 10 months after starting tacrolimus. Patients with mean tacrolimus trough levels of 10 to 15 ng/mL had the highest rates of response (86%) and remission (57%). Surgery seemed to be postponed in this group compared with others. An adverse event occurred in 75% of patients. Eight of these events (33%) required dose reduction and 6 (25%) led to treatment discontinuation. There were no irreversible side effects or deaths attributable to tacrolimus over a median follow-up of 56 months. Conclusions Oral tacrolimus seems to be safe and effective in some patients with severe CD refractory to anti–tumor necrosis factor therapy, particularly at a mean trough level of 10 to 15 ng/mL. Crohn’s disease, tacrolimus, Prograf, anti-TNF, refractory Crohn’s disease (CD) causes transmural inflammation throughout the gastrointestinal tract that can lead to surgery for a variety of reasons, including strictures, fistulas, and medically refractory disease. The relatively recent development of biological agents against tumor necrosis factor alpha (TNF-α) has proven effective for many patients. However, primary nonresponse to anti-TNF agents is not uncommon, and among those who do respond, at least one third fail to maintain a long-term response.1 Although numerous agents that target other pathways are in development, currently available therapeutic options are very limited for patients whose disease is refractory to anti-TNF therapy. Tacrolimus is a macrolide calcineurin inhibitor derived from Streptomyces tsukubaensis that is commonly used to prevent kidney and liver transplant rejection.2 Unlike cyclosporine, a less potent calcineurin inhibitor used for immunosuppression, tacrolimus can be administered orally due to pharmacokinetics that do not depend on the bilioenteric cycle or mucosal integrity, thereby resulting in more predictable absorption.3,–5 Several small retrospective and observational studies have reported encouraging response rates in patients with refractory CD treated with tacrolimus.6,–11 We sought to determine the relationship of tacrolimus blood levels to the response rate in CD refractory to anti-TNF therapy. We hypothesized that higher blood trough levels of tacrolimus would be associated with improved response rates of CD. Materials and Methods This was a single center, retrospective observational study. Potential cases were identified by searching electronic medical records from January 1, 2000 to April 1, 2010, that included the words “Crohn” and either “tacrolimus” or “Prograf.” The diagnosis of CD was made using standard clinical, radiographic, endoscopic, and histologic criteria. Disease location and behavior were categorized according to the Montreal classification scheme.12 Subjects had active CD that was symptomatic, and they were either unresponsive to, or intolerant of, at least one anti-TNF-α agent. Tacrolimus was not initiated to achieve mucosal healing in asymptomatic patients. Exclusion criteria were as follows: indeterminate colitis, perianal fistulas without other evidence of CD, <1 month of follow-up (without adverse effects that required termination of therapy), tacrolimus therapy prescribed primarily for posttransplant immunosuppression or pyoderma gangrenosum, or treatment with tacrolimus as part of an experimental trial for fistulizing CD.13 One subject was treated with 3 separate courses of tacrolimus after long intervals, and only the first course of treatment was included in this analysis. Demographic, clinical, and treatment data were extracted from medical records for those patients who met the entry criteria. Tacrolimus was initially dosed between 0.1 and 0.2 mg/kg/day. Dosage was adjusted according to trough level, response, and side effects. Trough levels, blood cell counts, and comprehensive metabolic panels were first measured 2 to 3 days after treatment initiation, then 1 to 2 weeks later, and then every 2 to 3 weeks thereafter. After reaching steady-state trough levels, these laboratory data were checked monthly. Acute kidney injury (AKI) was defined as a rapid increase in serum creatinine ≥0.3 mg/dL. Response was defined as improvement after at least 7 days of treatment of 1 or more of the following: bowel movement frequency, fistula output, rectal bleeding, abdominal pain, extraintestinal manifestations, or overall well-being. Remission required all of the following: <3 stools per day (excluding subjects with a previous resection), no bleeding, no abdominal pain, no extraintestinal manifestations, increased well-being, and no ongoing use of steroids. Loss of response was defined as either the discontinuation of tacrolimus or the first documentation of symptoms that progressed and led to tacrolimus discontinuation, the addition of corticosteroids, or surgery for CD.8,10 Descriptive statistics were used to analyze demographic and clinical characteristics, treatment response, and adverse events. The study was approved by the Institutional Review Board of the Mayo Clinic. Results Seventeen female and seven male patients (median age 38 years, range 19–57 years) were treated with tacrolimus after failing at least one anti-TNF-α therapy. All patients had clinically active CD involving more than the perianal region. Baseline demographic and clinical characteristics are shown in Table 1. Table 1. Baseline Characteristics of Patients with CD Treated with Tacrolimus     View Large Table 1. Baseline Characteristics of Patients with CD Treated with Tacrolimus     View Large As a group, patients receiving tacrolimus had long-standing, complex, and relatively refractory disease. Median disease duration was 9.5 years (range, 1–30 years). Most patients (75%) had ileocolonic disease and 21% had upper gastrointestinal involvement. Half of the patients had nonperianal penetrating disease and 63% had perianal penetrating disease. Treatment of fistulae was the primary indication for tacrolimus in 8 patients (2 perianal; 6 nonperianal). The majority of patients (63%) had already had at least 1 surgical resection and 37% were either steroid dependent or refractory. All but 1 patient (96%) had failed (either intolerant or unresponsive to) previous treatment with infliximab. Five patients (21%) had failed adalimumab. No patients had received certolizumab pegol. Three patients (13%) failed an investigational anti-TNF-α agent (CDP-571) as well as infliximab. One patient had failed natalizumab. Similarly, nearly all the patients (96%) had failed thiopurine monotherapy and 46% patients had failed methotrexate. Only 1 patient was actively smoking during tacrolimus therapy. Clinical response and steroid-free remission rates were 67% and 21%, respectively, over a median treatment duration of 3.8 months (Table 2). Of 16 responders, fistula improvement was the primary driver of response in 50% (3 perianal; 5 nonperianal). Tacrolimus seemed to have equivalent efficacy regardless of disease behavior. For instance, the response rates for patients with penetrating disease (n = 12) and nonpenetrating disease (n = 12) were both 67%, with no significant difference in the duration of response between the 2 groups. The median time to response was 21 days (range, 7–65 days) with an overall median duration of response of 3.7 months (range, 1–30 months). For the 11 responders who did not achieve remission, the median duration of response was 1.8 months (range, 1–30 months). Reasons for ultimate discontinuation among responders included loss of response (5)—one of which occurred after dose reduction because of AKI, adverse events (n = 2; both AKI), insufficient response (n = 2), and loss of insurance (n = 1). One patient maintained a partial response for 30 months after a dose reduction due to paresthesias and subsequent re-induction with infliximab that was continued as concomitant therapy. After stopping tacrolimus, 5 responders went to surgery, 3 were lost to follow-up, 1 tried adalimumab and then certolizumab pegol but ultimately went to surgery, and 1 tried adalimumab and then certolizumab pegol and is currently on natalizumab. Table 2. Treatment Variables and Outcomes of Tacrolimus Therapy     View Large Table 2. Treatment Variables and Outcomes of Tacrolimus Therapy     View Large Five patients achieved remission within a median of 2 months (range, 0.5–7 months). The duration of remission while still on tacrolimus was a median of 3.8 months (range, 1–21 months). Of the 5 patients entering remission, 1 was maintained on methotrexate and another on azathioprine monotherapy after stopping tacrolimus because of the desire to avoid potential long-term toxicity of tacrolimus as maintenance therapy. A third patient was maintained on combination azathioprine and tacrolimus but symptoms as a result of CD recurred after stopping tacrolimus due to tremor. The fourth patient was treated for 4 months and then maintained clinical remission off all therapy for CD for 6 months after resolution of anal ulcerations. The fifth patient maintained remission for more than 2.5 years on tacrolimus monotherapy but, after a dose reduction because of tremor, had disease recurrence that ultimately led to discontinuation. Using these various techniques to maintain remission after stopping tacrolimus, median remission was extended to 32 months or 28 months beyond the remission achieved while still on tacrolimus. The mean initial and maintenance doses of tacrolimus were 0.16 and 0.13 mg/kg/day, respectively, resulting in a mean trough concentration of 9.3 ng/mL. Twenty patients (83%) received some form of concurrent therapy, including thiopurines (n = 14), methotrexate (n = 2), and/or antibiotics (n = 11). Twelve patients (50%) were on steroids at the start of treatment, of whom 2 were able to wean and 5 were able to discontinue steroids. Surgery was ultimately required in 13 patients (54%) within a median time of 10 months (interquartile range 4–17 months). Most surgeries were resections, although 2 were diverting loop ileostomies that were performed because of persistent perianal fistula output. Surgical rates were the same (50%) for patients who responded within 30 days (n = 12) and those who responded later (n = 4). One or more adverse events occurred in 18 patients (75%) treated with tacrolimus (Table 3). Eight of these events (33%) required dose reduction and 6 (25%) led to treatment discontinuation. AKI occurred in 7 patients (29%). Tacrolimus-associated AKI was typically mild and reversible but did nevertheless result in the majority of dose reduction and drug discontinuation. Seven patients (29%) developed paresthesias and 3 of those patients required dose reduction. Headache and tremor occurred in 4 patients (17%). Headache required discontinuation of tacrolimus in 2 patients. Three patients with tremors needed dose reduction. There were no deaths attributable to tacrolimus over a median follow-up of 56 months. Table 3. Adverse Events While on Tacrolimus Treatment     View Large Table 3. Adverse Events While on Tacrolimus Treatment     View Large Stratification of clinical outcomes by mean tacrolimus trough level indicated that better therapeutic results may be associated with mean tacrolimus trough levels of 10 to 15 ng/mL (Table 4). This group of patients had the highest rates of response/remission and required less surgery (43%). Compared with all other groups, surgery occurred among patients with mean trough of 10 to 15 ng/mL after a longer interval, with an average time to surgery of 33 months after initiation of tacrolimus, and no surgeries within 1 year of starting treatment. In contrast, 56% of patients with mean tacrolimus trough levels of 5 to 10 ng/mL needed surgery within an average of 7 months of starting tacrolimus. Overall adverse event rates were lower in the 10 to 15 ng/mL mean trough group compared with the 5 to 10 ng/mL group. Subtherapeutic and supratherapeutic trough levels of <5 and >15 ng/mL, respectively, were associated with less favorable outcomes than those for patients with mean trough level of 10 to 15 ng/mL. Table 4. Outcomes Stratified by Mean Tacrolimus Trough Level     View Large Table 4. Outcomes Stratified by Mean Tacrolimus Trough Level     View Large Discussion In our cohort of patients with CD treated with tacrolimus after failing anti-TNF agents, two thirds of patients had at least some response, with about 1 in 5 entering steroid-free clinical remission. Over half of the patients on steroids were able to wean or stop. Additionally, there was an association between trough levels of 10 to 15 ng/mL and avoidance of surgery. These results are encouraging for such a refractory population of patients and are supported by previous studies demonstrating efficacy of tacrolimus for CD.6,8,–11,13,14 A recent systematic review of tacrolimus studies that included 70 patients with luminal CD reported pooled response and remission rates of 37% (range, 14%–57%) and 44% (range, 7%–69%), respectively.15 As with our study, most of these studies were uncontrolled and used differing measures of clinical efficacy, most of which were retrospectively assessed. In a 10-week randomized placebo-controlled trial by Sandborn et al13 in 46 patients with fistulas, 63% of whom had previously received IFX, perianal fistula improvement occurred in 43% of tacrolimus-treated patients compared with 8% in patients treated with placebo (P = 0.004). The odds ratio of fistula response in patients treated with tacrolimus was 8.62 (95% confidence interval: 1.60–46.45; P = 0.01); for the 29 patients who had previously received infliximab, the odds ratio was 9.23 (95% CI: 1.65–51.72; P = 0.01). There was no significant difference in fistula remission between the 2 groups; however, remission was a secondary endpoint for which the study may have been insufficiently powered to detect a significant difference between groups. In our cohort, response occurred within approximately 3 weeks and remission took a median of 2 months. Previous studies have also shown a rapid clinical effect with intravenous or oral tacrolimus. After induction with a goal trough of 10 to 15 ng/mL, 79% of patients in a prospective study by Tamaki et al11 had responded by 20 days, and all patients had responded or entered remission by 40 days. In a retrospective study that included 12 patients with CD treated with tacrolimus to goal troughs of 5 to 10 ng/mL, 83% had responded by 4 weeks and, at a mean follow-up of 5 months, 42% had entered remission.9 There is some evidence, however, that higher troughs may not be required for efficacy, although remission may be postponed. In a retrospective analysis of long-term results for 11 patients with CD treated to a lower trough goal of 4 to 8 ng/mL, Baumgart et al8 reported over 90% response at 30 days but no short-term remission. After a mean treatment period of 25 months, however, remission rose to 55%. Adverse events are common with tacrolimus. They are generally mild, dose-dependent, and reversible with dose reduction. Although three fourths of patients in our cohort had an adverse event, only one third led to dose reduction, and only one fourth led to tacrolimus discontinuation. Tremor (31%), paresthesia (26%), and headache (21%) were the most common side effects reported in a recent systematic review of previous tacrolimus series.15 Together, these adverse events led to discontinuation of tacrolimus in 7% of patients. Nephrotoxicity occurred in 16% of patients in previous series and was always reversible. It is unclear whether the incidence of these side effects correlated with trough levels. In our study, adverse events occurred with similar frequencies in all groups regardless of the mean trough level. In some cases, adverse events led to dose reduction that was then followed by loss of response and discontinuation. Given the potential for dose-limiting long-term toxicity from tacrolimus, identifying successful maintenance strategies is particularly important. Sandborn5 originally reported 3 cases of proximal small bowel CD for which tacrolimus induced a remission that were maintained with either methotrexate or 6-mercaptopurine. In our study, immunomodulator monotherapy was added to tacrolimus treatment in patients in remission on tacrolimus and then tacrolimus was discontinued and remission was maintained. In the 2006 series from Baumgart et al,8 patients with an initial response to tacrolimus were able to be maintained for similar periods of time on either immunomodulator monotherapy or, if that failed, an immunomodulator combined with tacrolimus at relatively low troughs of 4 to 6 ng/mL. Lowry et al16 also previously reported successful maintenance of response for perianal fistulas treated with combination therapy with immunomodulators and tacrolimus. In our cohort, two thirds of patients were receiving concomitant immunomodulator therapy along with tacrolimus. Neither response/remission rates nor duration of response/remission were significantly different in this group from those of patients who were not treated with immunomodulators. This is not surprising because these patients had previously failed immunomodulator therapy before tacrolimus therapy. Finally, another strategy to consider is to continue tacrolimus as maintenance monotherapy but to target lower troughs after achieving remission. As described by Tamaki et al11 in a prospective series of 14 Japanese patients with CD, reduction of the trough target to 5 to 10 ng/mL after achieving remission with trough targets of 10 to 15 ng/mL did not result in any disease recurrence at 120 days. There were no patients in our cohort who were managed according to this protocol. One of the strengths of our study is that it provides evidence for the efficacy of tacrolimus in patients with CD who are refractory to biological therapy. There are a number of shortcomings associated with an uncontrolled retrospective study in a small heterogeneous group of patients. Chief among these is our retrospective assessment of disease activity, which has inherent limitations with respect to the availability of data, and its interpretations. Our records contained insufficient data for the calculation of a retrospective Harvey–Bradshaw Index on all patients. Moreover, as a number of our patients had ostomies, Harvey–Bradshaw Index scores were not applicable. As a result, we chose to adapt definitions of response and remission that have previously been reported in other retrospective studies of tacrolimus for inflammatory bowel disease.8,10 They nonetheless suffer from a lack of validation. We attempted to account for these issues by including the relatively “hard endpoints” of steroid usage and surgery as outcome measures. Nevertheless, it is possible that the lower surgical rates seen with higher mean tacrolimus troughs reflect the fact that tacrolimus was not tried at higher doses, or for similar duration, in more sick patients who ultimately progressed to surgery sooner. In summary, tacrolimus seems to be effective in some patients with severe CD refractory to anti-TNF therapy. It may not ultimately avoid surgery in many of these patients, but it could postpone surgery and thereby avoid urgent operations and their associated worse postoperative outcomes. Moreover, by prolonging the time to surgery, tacrolimus could potentially decrease the overall number of surgeries and loss of functional bowel. References 1. Schnitzler F, Fidder H, Ferrante M, et al.  . Long-term outcome of treatment with infliximab in 614 patients with Crohn's disease: results from a single-centre cohort. Gut . 2009; 58: 492– 500. Google Scholar CrossRef Search ADS PubMed  2. A comparison of tacrolimus (FK 506) and cyclosporine for immunosuppression in liver transplantation. The U.S. Multicenter FK506 Liver Study Group. N Engl J Med . 1994; 331: 1110– 1115. CrossRef Search ADS PubMed  3. Jain AB, Venkataramanan R, Cadoff E, et al.  . Effect of hepatic dysfunction and T tube clamping on FK 506 pharmacokinetics and trough concentrations. Transpl Proc . 1990; 22: 57– 59. 4. Jain A, Venkataramanan R, Todo S, et al.  . Intravenous, oral pharmacokinetics, and oral dosing of FK 506 in small bowel transplant patients. Transpl Proc . 1992; 24: 1181– 1182. 5. Sandborn WJ. Preliminary report on the use of oral tacrolimus (FK506) in the treatment of complicated proximal small bowel and fistulizing Crohn's disease. Am J Gastroenterol . 1997; 92: 876– 879. Google Scholar PubMed  6. Ierardi E, Principi M, Francavilla R, et al.  . Oral tacrolimus long-term therapy in patients with Crohn's disease and steroid resistance. Aliment Pharmacol Ther . 2001; 15: 371– 377. Google Scholar CrossRef Search ADS PubMed  7. Baumgart DC, Wiedenmann B, Dignass AU. Rescue therapy with tacrolimus is effective in patients with severe and refractory inflammatory bowel disease. Aliment Pharmacol Ther . 2003; 17: 1273– 1281. Google Scholar CrossRef Search ADS PubMed  8. Baumgart DC, Pintoffl JP, Sturm A, et al.  . Tacrolimus is safe and effective in patients with severe steroid-refractory or steroid-dependent inflammatory bowel disease–a long-term follow-up. Am J Gastroenterol . 2006; 101: 1048– 1056. Google Scholar CrossRef Search ADS PubMed  9. Ng SC, Arebi N, Kamm MA. Medium-term results of oral tacrolimus treatment in refractory inflammatory bowel disease. Inflamm Bowel Dis . 2007; 13: 129– 134. Google Scholar CrossRef Search ADS PubMed  10. Benson A, Barrett T, Sparberg M, et al.  . Efficacy and safety of tacrolimus in refractory ulcerative colitis and Crohn's disease: a single-center experience. Inflamm Bowel Dis . 2008; 14: 7– 12. Google Scholar CrossRef Search ADS PubMed  11. Tamaki H, Nakase H, Matsuura M, et al.  . The effect of tacrolimus (FK-506) on Japanese patients with refractory Crohn's disease. J Gastroenterol  2008; 43: 774– 779. Google Scholar CrossRef Search ADS PubMed  12. Satsangi J, Silverberg MS, Vermeire S, et al.  . The Montreal classification of inflammatory bowel disease: controversies, consensus, and implications. Gut . 2006; 55: 749– 753. Google Scholar CrossRef Search ADS PubMed  13. Sandborn WJ, Present DH, Isaacs KL, et al.  . Tacrolimus for the treatment of fistulas in patients with Crohn's disease: a randomized, placebo-controlled trial. Gastroenterology . 2003; 125: 380– 388. Google Scholar CrossRef Search ADS PubMed  14. Gonzalez-Lama Y, Abreu L, Vera MI, et al.  . Long-term oral tacrolimus therapy in refractory to infliximab fistulizing Crohn's disease: a pilot study. Inflamm Bowel Dis . 2005; 11: 8– 15. Google Scholar CrossRef Search ADS PubMed  15. McSharry K, Dalzell AM, Leiper K, et al.  . Systematic review: the role of tacrolimus in the management of Crohn's disease. Aliment Pharmacol Ther . 2011; 34: 1282– 1294. Google Scholar CrossRef Search ADS PubMed  16. Lowry PW, Weaver AL, Tremaine WJ, et al.  . Combination therapy with oral tacrolimus (FK506) and azathioprine or 6-mercaptopurine for treatment-refractory Crohn's disease perianal fistulae. Inflamm Bowel Dis . 1999; 5: 239– 245. Google Scholar CrossRef Search ADS PubMed  Copyright © 2013 Crohn's & Colitis Foundation of America, Inc.
TI - Tacrolimus Salvage in Anti–Tumor Necrosis Factor Antibody Treatment-Refractory Crohn’s Disease
JF - Inflammatory Bowel Diseases
DO - 10.1097/MIB.0b013e318280b154
DA - 2013-05-01
UR - https://www.deepdyve.com/lp/oxford-university-press/tacrolimus-salvage-in-anti-tumor-necrosis-factor-antibody-treatment-MXrk3uulGb
SP - 1107
EP - 1111
VL - 19
IS - 6
DP - DeepDyve
ER -