SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Aliment Pharmacol Ther 2010; 32: 1129–1134

Summary

Background  Tumour necrosis factor-blockade with infliximab has advanced the treatment of Crohn’s disease. While infliximab is efficacious, it remains to be determined whether patients who enter clinical remission with an anti-tumour necrosis factor therapy can have their treatment stopped and retain the state of remission.

Aim  To assess in patients with Crohn’s disease who obtained infliximab-induced remission, the proportion who relapsed after infliximab discontinuation.

Methods  This longitudinal cohort study examined patients from a University-based IBD referral centre. Forty eight patients with Crohn’s disease in full clinical remission and who then discontinued infliximab were followed up for up to 7 years. Crohn’s disease relapse was defined as an intervention with Crohn’s disease medication or surgery.

Results  Kaplan–Meier analysis of the proportion of patients with sustained clinical benefit demonstrated that 50% relapsed within 477 days after infliximab discontinuance. In contrast, 35% of patients remained well, and without clinical relapse, up to the end of the nearly 7-year follow-up.

Conclusions  In patients with Crohn’s disease with an infliximab-induced remission, stopping infliximab results in a predictable relapse in a majority of patients. Nevertheless, a small percentage of patients sustain a long-term remission.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

The introduction of tumour necrosis factor (TNF)-blockade with infliximab, adalimumab, or certolizumab has offered significant advances in the treatment of inflammatory bowel disease and its complications. Indeed, infliximab has been demonstrated in multiple randomized controlled clinical trials to be efficacious for the treatment of both luminal and fistulizing Crohn’s disease and in decreasing the rate of hospitalizations and surgical interventions.1–5 Two recent single centre cohort studies, one from Canada and another from Belgium, demonstrated the efficacy of infliximab in ‘real-life’ clinical practice with induction of clinical improvement rates of 88% and 89%, and maintenance of clinical benefit rates of 83% and 84% at 1 year respectively.6,7

Despite this efficacy, issues relating to cost and long-term safety, in relation to infection and neoplasia, lead to the inevitable question: can the infliximab treatment be discontinued after induction of clinical remission, and/or mucosal healing, and will the patient remain in extended remission on maintenance immunosuppressant therapy? Existing clinical trial data are limited to answer the question whether infliximab can be discontinued and how long the patients will remain in remission thereafter. Infliximab was combined with methotrexate in 120 Dutch rheumatoid arthritis patients with severe disease activity. Seventy seven of the 120 patients discontinued infliximab after reaching a state of clinical remission; of these, 66 (86%) had a sustained remission off infliximab for at least 1 year.8 Similarly, nine patients with rheumatoid arthritis were able to discontinue infliximab after maintenance of remission and sustain their remission on methotrexate alone for a mean of 14 months.9 In patients with Crohn’s disease, the Leuven centre in Belgium followed up a group of 82 patients who had discontinued infliximab therapy for reasons other than loss of response, and found a median sustained remission period of 47 months.6 In a prospective study from France currently underway, but reported as an abstract, patients with luminal Crohn’s disease treated for at least 1 year with combined infliximab and an immunosuppressant were recruited to discontinue infliximab and continue the immunosuppressant. After a median follow-up of 1 year, 45/115 (40%) of patients relapsed.10

The objective of this study was to assess, in patients who had obtained a stable infliximab-induced corticosteroid-free remission, the proportion who relapsed after infliximab discontinuation and to examine factors associated with this relapse.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Patient population

Between July 2000 and July 2007, 354 patients with Crohn’s disease completed a three-dose induction regime with infliximab, were deemed to have responded to infliximab by their individual gastroenterologist after a physician global assessment, and entered into maintenance infliximab dosing every 8 weeks. Of those patients who entered into maintenance infliximab treatment, a total of 48 then discontinued the infliximab for reasons other than loss of response. These patients were identified from an electronic database and with hand searching.

Each of the 48 patients received infliximab in accordance with the Canadian Association of Gastroenterology’s anti-TNF treatment guideline for patients with Crohn’s disease. This guideline specifies a protocol-based ‘step-up’ approach that requires both corticosteroids and immunosuppressive (azathioprine or methotrexate) therapies to have failed, due to lack of response or intolerance, prior to initiation of infliximab therapy.11 Patients who discontinued infliximab while in an infliximab-induced remission were identified from the IBD referral practices of nine Gastroenterologists associated with the University of Alberta. Long-term follow-up for each patient was available until 6th July 2009.

Through a hand-searched chart review, patients were eligible to be included in the cohort for this study if they had1 responded to infliximab (5 mg/kg) induction dosing given at weeks 0, 2 and 6, and2 advanced onto scheduled maintenance treatment every 8 weeks,3 maintained a stable corticosteroid-free clinical benefit for at least 6 months,4 discontinued infliximab therapy for reasons other than loss of response and5 sufficient follow-up that allowed assessment of ongoing wellness and/or disease relapse.

Outcome measures

The primary objective was to assess, in patients with a stable infliximab-induced, corticosteroid-free clinical benefit, the proportion who relapsed after infliximab discontinuation. The secondary objective was to determine factors associated with relapse after infliximab discontinuation.

Definitions

Infliximab-associated induction and maintenance responses were defined as corticosteroid-free clinical benefit, by physician global assessment that resulted in lasting control of disease activity during follow-up and persistent improvement of symptoms. The dates of each initial and final infliximab infusion and date of a Crohn’s disease relapse after the discontinuation of infliximab therapy were determined by hand searching the infliximab infusion database and patient medical records respectively. Crohn’s disease relapse was defined as a physician or hospital visit with documented symptoms of disease activity and a therapeutic intervention with Crohn’s disease medications, or a hospitalization with complications related to active Crohn’s disease. A complete out-patient and in-patient dataset for all 48 patients was available.

Statistical analysis

For this study, spss 16.0 software (SPSS, Chicago, IL, USA) was used to perform all appropriate statistical analyses. Mean values were calculated for continuous data and percentages were computed for discrete data. Infliximab’s sustained clinical benefit was estimated by using both Cox proportional hazards analysis and Kaplan–Meier curves. The outcome variable used was the duration of time between the termination of infliximab therapy and a clinically significant relapse. Right censoring on this duration was performed at the day of the analysis (6 July 2009). Univariate and multivariate survival models were used to assess the effects of gender, location of disease, number of infliximab doses, concomitant drug use, age of disease onset and the period of time on infliximab.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Patient characteristics

The patient’s baseline characteristics are outlined in Table 1. A total of 48 patients with Crohn’s disease were identified who were in steroid-free clinical remission with regular maintenance infliximab therapy, but then discontinued their infliximab for reasons other than infliximab-associated adverse events or loss of response. One hundred percent of the cohort was Caucasian (data not shown). Following a three-dose infliximab (5 mg/kg) induction regime patients received between 2 and 51 regular maintenance infliximab (5 mg/kg) infusions before discontinuance. The median duration of follow-up after infliximab discontinuance was 4.1 years (interquartile range 0.5 to 6.7 years).

Table 1.   Patient baseline characteristics
Male/Female (%)22/26 (46/54)
Mean age at diagnosis (years)25 ± 1.4
Mean duration of disease prior to first infliximab infusion (years)12 ± 1.2
Mean age at first infliximab infusion (years)37 ± 1.8
Median number of infusions8 (2–51)
Median duration of drug therapy (months)15.6 (1.0–67.3)
Mean age at infliximab discontinuance (years)38.0 ± 1.7
Median duration of follow-up (years)4.1 (0.5–6.7)
Location of disease (%)
 Colon15 (31)
 Ileo-colonic22 (46)
 Ileum10 (21)
 Jejunum3 (6)
Concomitant medications (%)
 Azathioprine21 (44)
 Methotrexate9 (19)
 Mesalazine (mesalamine)7 (15)
 Mercaptopurine2 (4)
 Budesonide1 (2)
 None13 (27)

Concomitant immunosuppressive therapy

At the time of infliximab discontinuance, 44% of the patients were receiving azathioprine, 19% methotrexate and 4% mercaptopurine as concomitant immunosuppressive therapy. The rates of concomitant immunosuppressive therapy at infliximab discontinuance were identical to the patient’s concomitant therapy at their first infliximab infusion and at the time of Crohn’s disease relapse following infliximab discontinuance (Table 1).

Discontinuation of infliximab treatment

One hundred percent of patients in this cohort had achieved an infliximab-induced corticosteroid-free clinical benefit as determined by physician global assessment at the time of their infliximab discontinuance. The mean age at discontinuance was 38.0 ± 1.7 years. Of the 48 patients within the cohort, infliximab treatment was discontinued after a median of 8 infusions (2–51) and after a median duration of 15.6 months (1.0–67.3). Information on the reasons for discontinuation of infliximab therapy was available in 43 of the 48 patients. Thirty three percent voluntarily chose to stop infliximab on their own accord or were instructed to stop infliximab by their physicians because they were in a sustained clinical remission. Other reasons for infliximab discontinuance include loss of third party insurance coverage for infliximab (19%), infliximab-associated infusion reactions or serum sickness-like reactions (16%), personal choice due to concerns about drug safety (18%), and stopping therapy because of planned pregnancy (14%) (Table 2).

Table 2.   Reasons for discontinuing infliximab therapy
Reason for discontinuing infliximab therapyN (%)
Physician choice14 (33)
Loss of insurance8 (19)
Patient choice8 (18)
Adverse events7 (16)
Pregnancy6 (14)

Duration of clinical benefit after infliximab discontinuation

The duration of Crohn’s disease clinical benefit after the infliximab discontinuance was assessed by determining the interval between the last infliximab infusion and the next documented episode of Crohn’s disease relapse (see Methods). The Kaplan–Meier curve of the proportion of patients with sustained clinical benefit demonstrated a steady decline with a 50% Crohn’s disease relapse rate at a median of 477 days (95% CI; 290–1339) after infliximab discontinuance (Figure 1). Interestingly, 35% of patients who discontinued infliximab remained in extended clinical remission without requirement for Crohn’s disease therapy. This level of clinical remission was maintained for the duration of follow-up. This implies that within the variable responses to infliximab seen in patients, a subgroup of patients with Crohn’s may achieve indefinite remission. Multivariate analysis of concomitant drug therapy, number of infliximab doses, age, gender, and disease location, and number of years from diagnosis to the start of infliximab therapy did not statistically impact the survivor curve (data not shown).

image

Figure 1.  Kaplan-Meier curve of the proportion of patients with sustained clinical benefit after infliximab discontinuance. Fifty percent of the patients relapsed at a median of 477 days (95% CI; 290 to 1339 days). Approximately 35% of patients who discontinued infliximab remained well with no disease relapse requiring therapeutic intervention.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

While the advent of anti-TNF-agents to manage Crohn’s disease has revolutionized treatment paradigms in the last decade, the duration anti-TNF therapy should be continued in patients in remission and when, and if, anti-TNF agents can ever be stopped remain unknown. This is a particularly relevant clinical question given the cost and safety profile of these agents in conjunction with the lifelong affliction of the disease.12, 13

In this longitudinal cohort study, we followed up, for up to nearly 7 years, patients with Crohn’s disease who had an infliximab-induced corticosteroid-free clinical benefit and then discontinued infliximab for reasons other than loss of response. The patients with Crohn’s disease received infliximab through a protocol-based ‘step-up’ approach whereby, to receive an induction dose of infliximab, patients must have demonstrated remission failure or adverse events to both corticosteroids and immunosuppressive drugs.14

In this current study, Kaplan–Meier analysis of both patients maintaining remission and those with a clinical relapse demonstrated that approximately 50% of patients will have relapsed within 477 days after the discontinuation of infliximab therapy. Interestingly, there is a plateau beginning ∼750 days after the final infliximab infusion and becoming flat at ∼1300 days that demonstrate an extended and stable long-term remission for approximately 35% of the patients.

Similar results were found in a Belgian study where patients with Crohn’s disease received a median of 4 infliximab injections over a median of 6 months and then had the infliximab stopped while in remission. These patients remained in clinical remission for a median period of 47 months (∼1400 days) ranging between 21 and 66 months (∼630 to ∼2000 days).6 In another European prospective study, infliximab was discontinued in steroid-dependent patients with Crohn’s. After 12 months, 40% of the patients with Crohn’s disease had relapsed.10 Moreover, an additional study found that 23% of patients receiving placebo medication after their original loading doses of infliximab at weeks 0, 2, and 6 maintained clinical remission when assessed at 54 weeks.13 While these studies examined clinical remission after infliximab discontinuance, the duration of their assessment was short relative to the lifelong nature of Crohn’s disease. Our study not only reaffirms the short-term duration of clinical remission found in these previous studies but also demonstrates, for the first time, that a significant subset of patients can sustain clinical benefit for long intervals after infliximab was discontinued.

Factors affecting the duration of a clinical benefit after discontinuing infliximab remain speculative. Multivariate analysis of concomitant drug therapy, number of infliximab doses, age, gender and disease location did not statistically impact the survivor curve (data not shown). Instead, it is very likely that a patient’s initial response to infliximab, and in conjunction the sustained response, or not, after discontinuance of infliximab, is based on a patient-unique genetic type of Crohn’s disease.15–18 Indeed, recent studies using microarray data sets to assess gene arrays have identified predictive panels of genes for response and/or nonresponse to infliximab in patients with ulcerative colitis and rheumatoid arthritis.19, 20 Future studies are needed to analyse genetically patients with Crohn’s disease who have discontinued infliximab therapy in attempt to correlate their duration of remission with their disease genotype.

Anti-TNF-induced mucosal healing has been demonstrated to prolong remission and reduce hospitalizations and surgery.21 Furthermore, complete mucosal healing has been found to occur in approximately one quarter of patients, a value similar to the number of patients in our study who had extended long-term clinical benefit after the infliximab was discontinued. It is possible that this same group of patients with complete mucosal healing are also those who have the sustained clinical benefit following infliximab discontinuation. Unfortunately, the current study patients did not have prospective sequential endoscopies to assess mucosal healing and thus future studies will need to address this question.

In summary, this longitudinal cohort study identified that nearly two-thirds of patients with Crohn’s disease who are in an infliximab-induced clinical remission will ultimately relapse after their anti-TNF therapy is discontinued and that in 50% of patients, this relapse will occur within 500 days. It should be noted, however, that one-third of our cohort achieved long-term remission after the discontinuance of infliximab therapy. Extrinsic factors affecting the duration of Crohn’s remission could not be identified making a genetic link plausible.

Acknowledgement

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Declaration of personal interests: None. Declaration of funding interests: This work has been partially supported by an unrestricted grant from the CCFC and CFI.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References