In the first placebo-controlled trial by Present et al.30 in fistulizing Crohn's disease, a series of three infusions of 5 mg/kg of infliximab at weeks 0, 2 and 6 resulted in a response (≥50% reduction of draining fistulae) at two consecutive evaluations in 68% of patients and in 56% with 10 mg/kg with complete cessation of drainage in 55% and 38% of the patients, respectively. The placebo response rate was 26% and complete response rate 13%. The median time to loss of response without re-treatment was 12 weeks.
The median time to loss of response was significantly longer for patients who received infliximab maintenance (>40 weeks) when compared with those who received placebo maintenance (14 weeks, P < 0.001). At week 54, 23% of patients on placebo maintained a response and 19% a complete response when compared with 46% and 36%, respectively, for patients maintained on infliximab 5 mg/kg every 8 weeks (both P < 0.01).
Can we predict response to infliximab in patients with Crohn's disease?
The efficacy to induce response and remission both in luminal disease and in fistulizing Crohn's disease reported in the placebo controlled trials has been confirmed in many open cohorts of patients throughout the world.36–42 About 30% of patients with refractory Crohn's disease have consistently been found to be resistant to infliximab therapy. Moreover, not all responders display a full response.
It is important to identify reliable predictors for response to infliximab in order to optimize the use of the drug and to gain insight in the differences in pathogenesis of different disease forms. A number of clinical, biological and genetic risk factors have been investigated. We report only these factors which have been shown by at least two independent randomized-controlled trials or large cohort studies to predict the outcome of infliximab therapy.
Patients with biologically active inflammation as witnessed by increased CRP have the best chance of responding to infliximab therapy.29, 43 Patients with non-stricturing disease and with pure colonic disease respond better to infliximab therapy.44, 45 Patients treated with concomitant immunosuppression 45–47 are more likely to respond both short- and long-term to infliximab therapy. The Odds ratio to respond when patients are taking immunosuppressants was 3 (95% CI 2–5) in the large cohort by Vermeire et al.45 and 7 (95% CI 3–10) in the study by Arnott et al.47 Also, in the cohort of Parsi et al., 46 74% of patients taking immunosuppression responded to infliximab compared with 39% of those not on immunosuppressive therapy (P = 0.007). Whether this additive effect is because of improved combined efficacy or to suppression of the formation of antibodies to infliximab (ATI) is not clear. Smokers are less likely to respond to infliximab.46,47 The value of smoking and immunosuppression has not been identified as clinical risk factors in all studies.48
Infliximab is highly effective in achieving improvement and inducing and maintaining remission in paediatric patients with Crohn's disease in fistulizing as well as luminal disease.49–53 It seems that children with early Crohn's disease have a higher chance of prolonged response to infliximab than children with longstanding Crohn's disease.52 This suggests that early introduction of the drug might be particularly efficacious to change the long-term course of Crohn's disease. Borrelli et al.53 also showed in their cohort of patients that treatment with infliximab induces weight gain and reverses growth failure in children especially when they are maintained on the drug.
Besides CRP, other biomarkers have been proposed but not confirmed as predictors of response to infliximab. Specific predictors of response in fistulizing disease have not been identified. In an open study from Calgary, the presence of a rectovaginal fistula was a poor prognostic indicator for successful infliximab therapy 54, but this was not suggested by the Accent II study. In the large Italian cohort study48, the efficacy of infliximab was less in the treatment of other fistulae when compared with perianal fistulae.
Several genetic factors have been studied and interesting data have been gathered but genetic predictors need to be confirmed in different cohorts of patients.
Infliximab for ulcerative colitis
In Crohn's disease, the inflammatory response is characterized by the presence of T-helper 1 cytokines including IL-12, IFN-γ and TNF whereas in ulcerative colitis there seems to be an excess of T-helper 2 cytokines including IL-5 and -13 while IL-12 is lacking. However IFN-γ is also abundantly present as well as TNF. The results of open studies on the use of the monoclonal chimeric IgG1 antibody to TNF, infliximab in ulcerative colitis, were conflicting. In the small placebo-controlled study by Sands et al., 55 four of eight patients (50%) treated with infliximab responded at 2 weeks and none of the three placebo-treated patients.
In the study by Probert et al., 56 remission was achieved with infliximab in three of 23 (13%) patients with steroid-refractory UC and in 1/19 (5%) treated with placebo at 2 weeks and the rates were 39% and 30%, respectively, at 6 weeks (NS).
In a randomized study 57, Ochsenkuhn et al. investigated the efficacy of infliximab 5 mg/kg at weeks 0, 2 and 6 to induce remission in patients with acute severe ulcerative colitis not refractory to glucocorticosteroids in comparison with high doses of glucocorticosteroids (2 mg/kg). Only 13 (six received infliximab and seven prednisolone) patients were included in this study. Five of six patients responded to infliximab and six of seven to prednisolone at 3 and 13 weeks, respectively.
Definitive evidence for the efficacy of infliximab in the treatment of UC was offered by the two large placebo-controlled clinical trials ACT-1 58 and ACT-2.59
In ACT-1, 69% and 62% of patients receiving infliximab 5 and 10 mg/kg, respectively, at weeks 0, 2 and 6, were in clinical response at week 8, compared with 37% of those who received placebo (P < 0.002 for both comparisons). In ACT-2, 65% and 69% of patients receiving infliximab 5 and 10 mg/kg, respectively, were in clinical response at week 8, compared with 26% of those who received placebo (P < 0.001 for both comparisons).
In both studies, the proportions of patients who achieved clinical response at weeks 8, 30 or 54 (ACT-1) or clinical remission at weeks 30 or 54 (ACT-1) were several-fold higher among infliximab-treated patients than placebo-treated patients. The full publication of the data is awaited.
Infliximab seems also efficacious as rescue therapy to avoid colectomy in severe to moderately severe ulcerative colitis. In a placebo-controlled pilot study, Jarnerot et al. found that 71% of patients resistant to steroids escaped colectomy in 3 months when treated with infliximab vs. 33% of the patients treated with placebo (P = 0.017).60
Data are scarce also for treatment of indeterminate colitis with infliximab. In an open retrospective study reported by Papadakis et al. 61 of infliximab treatment in medically refractory indeterminate colitis, 14 of 20 patients (70%) had a complete response to infliximab 5 mg/kg and two patients a partial response. Four of the 14 responders had to step up to 10 mg/kg infliximab because of attenuation of the response to 5 mg/kg. It was striking that 10 patients were in the course of follow-up reclassified as Crohn's disease and two as ulcerative colitis.
Infliximab therapy for systemic manifestations of inflammatory bowel diseases
Extra-intestinal manifestations occur in up to 40% of patients with inflammatory bowel disease. These include mainly skin lesions, eye disease, joint manifestations and primary sclerosing cholangitis (PSC). Infliximab seems efficacious for all systemic manifestations of inflammatory bowel disease whether associated with Crohn's disease or ulcerative colitis or even if there is no evidence of underlying inflammatory bowel disease.
The only placebo-controlled trial of infliximab therapy for pyoderma gangrenosum showed short-term efficacy of infliximab 5 mg/kg at 2 weeks in 6/13 (46%) of patients vs. 1/17 (6%) of placebo-treated patients (P = 0.025).62 Infliximab was found to be efficacious for the management of peristomal pyoderma gangrenosum a particularly disabling complication of Crohn's disease or ulcerative colitis.63
Eye abnormalities associated with inflammatory bowel disease are episcleritis, scleritis and uveitis. Uveitis associated with Crohn's disease or ulcerative colitis is predominantly bilateral, posterior, insidious in onset but chronic in duration. This contrasts with the uveitis associated with spondylarthropathy, which is mostly anterior, unilateral, sudden in onset and limited in duration.64 Infliximab is efficacious for treating both types of uveitis.65–66 The spectrum of joint disorders associated with inflammatory bowel disease includes peripheral arthropathy, which evolves in parallel with the activity of the bowel disease and sacroiliitis or ankylosing spondylitis which, although associated with inflammatory bowel disease, has a rather independent course. Infliximab relieves peripheral joint problems together with the bowel symptoms. Axial manifestations or spondylarthropathy associated with Crohn's disease also respond well to infliximab.67 Moreover, both infliximab and etanercept are very efficacious to improve disease activity, functional indices and quality of life in patients with active ankylosing spondylitis.68–70 There are also data suggesting that these strategies prevent functional loss and progress of spine lesions.
In an unpublished pilot trial, infliximab did not improve liver abnormalities in patients with primary sclerosing cholangitis and inflammatory bowel disease.
How important is the immunogenicity problem with infliximab?
Infliximab is a chimeric antibody and may be associated with the formation of ATI formerly called human antichimeric antibodies (HACA). The detection of the antibody depends on the sensitivity and on the specifications of the assay used, the timing of measurement of the antibodies. With the present assays, ATI can mostly not be measured as long as infliximab is present in the serum of the patient.
In general, it can be stated that ATI interfere with the safety and efficacy of the drug. Immunogenicity leads to clinical problems especially when infliximab is administered episodically.
In the episodic re-treatment arm of the Accent I study, 71 the cumulative incidence of ATI as measured with the Centocor assay amounted also to 30% through 72 weeks, which was significantly higher than the 10% and 7% in the group of patients treated with systematic treatment with 5 or 10 mg/kg infliximab infusion every 8 weeks. In the episodic treatment arm, ATI impaired the kinetics of the drug with a more rapid reduction in serum infliximab concentrations from postinfusion peak levels and a reduction in the magnitude and duration of clinical response. The presence of ATI was associated with a 12% increase in infusion reactions. Overall patients receiving immunomodulators had a lower incidence of ATI through the groups (10% vs. 18%) than patients not receiving immunosuppressants. However, in the overall populations of the trial similar proportions of antibody-positive, -negative and inconclusive patients achieved clinical response (64%, 62% and 65%) and clinical remission (41%, 39% and 48%) at 54 weeks.71
In open cohorts of patients treated with infliximab in an episodic ‘on flare’ manner, the formation of ATI was found to be an important clinical problem.
Baert et al. 72 detected ATI in 38 of 125 (61%) patients and Farrell et al. 73 reported ATI in 19 of 53 patients (36%). In the studies by Baert et al. 72 and Farrell et al. 73, it was clearly demonstrated that the formation of ATI was associated with the occurrence of infusion reactions, lower postinfusion infliximab serum levels and with a shortened duration of response. The formation of ATI was decreased by concomitant therapy with immunosuppressants. Azathioprine and methotrexate are equally effective to counteract the formation of ATI.74
How are we using infliximab in the clinic?
The present indications for infliximab in inflammatory bowel disease are summarized in Table 1. For control of luminal disease, we use one 5-mg/kg i.v. infusion followed by maintenance infusions every 8 weeks. A three dose induction regimen with i.v. infusion of 5 mg/kg infliximab at 0, 2 and 6 weeks is only marginally more efficacious than a single infusion. If a patient does not respond to the first infusion, we will mostly repeat the infusion after 4 weeks. If he does not respond to the second infusion or if he did not respond to a loading schedule, we will not use the drug further in this patient. For fistulizing Crohn's disease, a loading dose of infliximab 5 mg/kg at 0, 2 and 6 weeks is generally used. Fistulizing disease necessitates always a medical surgical approach from the onset. Many patients will need drainage of abscesses, placement of setons, fistulotomy, advancement flap plasty and sphincter repair. If setons are in place, they should be kept until the first infusions have been administered and can then be removed when maintenance therapy is carried out. Drainage of all sepsis prior to start of infliximab therapy is mandatory. It is strongly recommended that patients treated with infliximab continue their immunosuppressants or are started on one of these drugs.
Table 1. Indications for induction and maintenance infliximab in patients with inflammatory bowel disease
|Refractory luminal (including upper GI) Crohn's disease Steroid-dependent Crohn's disease Refractory fistulizing Crohn's disease Ulcerative colitis Systemic manifestations of inflammatory bowel disease Ankylosing spondylitis and sacroiliitis Pyoderma gangrenosum Chronic uveitis Metastatic Crohn's disease||First line therapy Indeterminate colitis Refractory pouchitis||Fibrostenosis only Refractory anemia in inflammatory bowel disease Primary sclerosing cholangitis|
Infliximab can be used in different settings long term. These include (i) infliximab every 8 weeks maintenance as monotherapy, (ii) infliximab every 8 weeks with concomitant immunosuppression maintenance, (iii) infliximab episodic with immunosuppression maintenance and (iv) infliximab as a bridge to immunosuppression.
There is no doubt that after induction with infliximab maintenance with infliximab 5 mg/kg every 8 weeks is the optimal regimen for patients treated for refractory luminal or fistulizing Crohn's disease. There is no need to wait for a new flares to re-administer, there is lower risk of immunogenicity, there is better mucosal healing and the improvement in quality of life is maintained long term. There is also evidence from the Accent I study that systematic maintenance therapy with infliximab is superior to episodic therapy to reduce hospitalizations and surgeries over the period of 1 year of treatment. In all patients who tolerate the drugs, an immunosuppressant should be maintained when the patient is already on the drug or added at the start of infliximab. It is not clear whether the combination should be maintained throughout. Preliminary data from an immunosuppression discontinuation study seem to indicate that immunosuppression may be stopped after 6 months without risk for earlier relapse.75
From the Accent I and II studies, we know that up to 30–50% of patients treated with infliximab every 8 weeks lose response in the course of 1 year following the successful induction of improvement or remission. This loss of response may be because of the formation of ATI, to rapid clearance of the drug unrelated to immunogenicity or maybe because of the increase of other pro-inflammatory molecules. This loss of response can be overcome by the increase of the dose of infliximab. In the Accent I study, increase to 10 mg/kg in patients with luminal Crohn's disease restored response in 90% of the patients who lost response to 5 mg/kg. Similarly, in the Accent II study, increase to 10 mg/kg in patients with fistulizing Crohn's disease restored fistula response in 57% when draining of fistulae recurred during the maintenance therapy with 5 mg/kg.
The efficacy of dose increase to overcome loss of response has been largely confirmed in clinical practice. Another way to deal with the problem of loss of response is to decrease the interval between infusions and administer 5-mg/kg infliximab every 4–6 weeks instead of every 8 weeks. The strategy will be decided based on the practicalities of drug acquisition.
Infliximab can also be used for induction of remission as a bridge to the action of an immunosuppressant. In that setting, infliximab is used for induction in patients who are mostly resistant to or dependent on glucocorticosteroids and who have not failed to respond to immunosuppressants in the past. There is not much evidence that this is an effective strategy.
The GETAID group 76 has investigated the potential of infliximab as a bridge to immunosuppression in patients with steroid dependence for more than 6 months. Patients were treated with infliximab at 0, 2 and 6 weeks or placebo together with azathioprine or 6-MP and followed for 1 year. The results of the study have not been published in full but the data show that although the combination of infliximab induction and long-term azathioprine is more effective than azathioprine alone there is loss of response over time and that the majority of patients will require rescue therapy also in the infliximab-azathioprine arm. The rates for remission and off steroids were 75% (infliximab) vs. 38% (placebo) P < 0.0001 at 12 weeks, 57% (I) vs. 29% (P) P = 0.003 at 24 weeks and 40% (I) vs. 22% (P) (P = 0.04) at 52 weeks.
The preliminary data from the Benelux step-up/top-down trial seem to suggest a role for infliximab for bridging to immunosuppression. Induction therapy with infliximab with maintenance with azathioprine resulted in a remission rate without steroids in 75% at 6 months vs. 41% for step-up therapy (P = 0.006).77 The details of this study are awaited for.
An important argument used to advocate episodic treatment is the lower cost. When we look at series however where episodic therapy is used, it appears that the mean interval between episodic infusions is between 9 and 14 weeks in patients that need re-treatment. This is not so very different from the 8 weeks used in the systematic maintenance schedule. The decreased cost with episodic therapy is frequently offset by the problems of infusion reactions, of loss of response with the necessity to increase the dose of the drug or (and) to decrease the dosage interval. Moreover successful therapy with infliximab may decrease the direct as well as the indirect costs of the disease not including the cost of drug acquisition. This has been suggested by three retrospective analyses.78–80 There are a number of contra-indications to the use of infliximab. The drug should not be used in the presence of active sepsis or abscess or in patients with active tuberculosis. Also, multiple sclerosis and severe heart failure and a history of optical neuritis or malignancy especially lymphoma are contraindications. The drug should be stopped in case of uncontrolled infusion reactions or intestinal obstruction.
Infliximab for ever and ever?
There are no stopping rules for infliximab therapy.
There is also no controlled study data on infliximab therapy beyond 1 year but experienced centres have many patients who are treated with infliximab for up to 7 years with excellent efficacy and safety.
We use the following strategies. In patients who were not on immunosuppression, we will attempt to use infliximab as bridge therapy and give only induction therapy. If the patient does well, we will not repeat infliximab. If a patient relapses, we will retreat and maintain with infliximab 5 mg/kg every 8 weeks.
In patients who have failed immunosuppression, we will induce with infliximab and concomitant immunosuppression and continue infliximab every 8 weeks for at least 1 year.
When in such patient the treatment goal is reached, e.g. the discontinuation of glucocorticosteroids with maintained remission or complete external healing of fistulae, we may try to discontinue infliximab and continue immunosuppression and see whether the disease stays in remission. If the disease relapses, treatment is resumed with infusions every 8 weeks long term.
For extra-intestinal manifestations of the disease, we will use infliximab until the signs have disappeared in the case of skin or eye manifestations and then we will discontinue therapy while keeping the patients on immunosuppression. In the case of joint problems, we will always maintain with infliximab therapy every 8 weeks.
Safety of infliximab in the treatment of inflammatory bowel disease
The most common adverse effects with infliximab are related to the formation of ATI. Acute infusion reactions are mostly easy to deal with and are treated with slowing of the infusion and administration of antihistamines and (or) hydrocortisone. In patients who have suffered infusion reactions, we will use prophylactic hydrocortisone 250 mg i.v. 30 min before each subsequent infusion.
Delayed infusion reactions also called serum sickness-like reactions are treated with high doses of steroids for 4–7 days. Patients having suffered delayed reactions can be re-treated with infliximab but need prophylaxis with prednisolone 40 mg/day for 2 days prior to and during 5–7 days after each infusion. However, the response to therapy is mostly lost in these patients and these patients will benefit from switching to the fully human antibody. After a drug-free interval of more than 14 weeks, prophylaxis with hydrocortisone prior to infliximab infusion is recommended in all patients to avoid infusion reactions.
Infections occur commonly in patients with chronic diseases treated with several potentially toxic drugs. In clinical studies, 36% of patients treated with infliximab and 28% of patients treated with placebo had infections that required therapy (PSUR 11).81
Respiratory tract infections are more common in patients treated with infliximab than with standard therapies, but these respond well to supportive therapy and antibiotics.
The rate of serious infections in placebo-controlled clinical trials in Crohn's disease and RA in patients treated with infliximab (6%) was not higher than in patients treated with placebo (7%). In a large cohort study82 in 500 consecutive patients treated at the Mayo Clinic, infliximab-related infections were found in 8% including 20 patients with serious infections: two patients had fatal sepsis, eight had pneumonia (two were fatal), six viral infections, two abdominal abscess, one cellulitis and one histoplasmosis.
Infliximab can be used safely in the perioperative period in patients undergoing bowel resections. There is no increased risk for severe infections or other complications.83, 84
There is an increased risk of activation of latent tuberculosis and less of other opportunistic infections associated with therapy with infliximab. By February 2005, 709 cases of tuberculosis have been reported including 591 spontaneous reports, 86 trial reports and 32 registry reports. The median time from the first infliximab infusion to the onset of symptoms was 123 days. There were a total of 62 reports with tuberculosis as the underlying cause of death.81 Educational programs have resulted in a gradual decrease in the reporting rate of tuberculosis. All patients who will undergo treatment with an anti-TNF agent should be screened for latent tuberculosis. A careful interview for a history of tuberculosis or exposure to tuberculosis as well as a chest X ray are mandatory prior to start of infliximab therapy. There is a lot of discussion on the usefulness of the skin test for screening. A positive skin test mostly is indicative of latent tuberculosis even if the patient has received BCG vaccination. False negative tests occur in patients treated with steroids or immunosuppression. Patients with positive skin test should receive prophylactic therapy with isoniazide 300 mg/day for 6 months and pyridoxin once weekly. This therapy should be given for 2–4 weeks before the first infusion of infliximab is given. Physicians should adhere to the recommendations for TB screening and treatment as proposed by national scientific organizations and authorities.
Mild-to-moderate increases in liver enzymes have been rarely seen in patients treated with infliximab. Severe hepatic disease has been observed in <0.08/1000 patients and a causal relationship with infliximab has not been established.81
Infliximab may worsen congestive heart failure, and should be used in patients with this concomitant condition only after the consideration of other treatment options. Infliximab and other agents that inhibit TNF have been associated in rare cases with optic neuritis, seizure and new onset or exacerbation of clinical symptoms and/or radiographic evidence of central nervous system demyelinating disorders, including multiple sclerosis. Infliximab therapy does not predispose to progressive multifocal leucoencephalopathy. Infliximab therapy is frequently associated with autoimmunity with the formation of antinuclear antibodies and antibodies to double-stranded DNA. Drug-induced lupus reactions without end organ damage occur rarely. Antinuclear antibodies are associated with female sex and with the occurrence of papulosquamous or butterfly rash.85 The development of antinuclear antibodies (ANAs) is no reason to discontinue therapy with infliximab.
In the CD and RA trials, six lymphomas were diagnosed for a follow-up of 4148 patient years vs. 0 for 691 placebo patient years. All lymphomas occurred in patients treated with concomitant immunosuppression. Four of these lymphomas occurred in patients with RA. In RA, the background incidence of lymphoma is increased in comparison with the general population. In a prospective study in 18 572 patients with RA who were enrolled in a US National Data Bank (NDB) for rheumatic diseases, a standardized incidence ratio (SIR) of 3 (95% CI 1–5) in comparison with the Surveillance, Epidemiology and End Results (SEER) data base was reported for the use of infliximab86 whereas the overall SIR was 2 (95% CI 2–5).
Data for Crohn's disease are scarce. There is lack of evidence at this stage that there is an increased risk of lymphoma associated with the use of infliximab in Crohn's disease. It is also unlikely that good quality controlled data will ever be available making treatment registries especially important.
Mortality in patients with inflammatory bowel disease treated with infliximab is not increased in comparison with patients not treated with this drug and mortality is mostly the consequence of complications of the disease.
The mortality rate in the infliximab trials with 3 years poststudy follow-up was 2% for patients treated with infliximab and 4% for placebo-treated patients. In postmarketing, surveillance infection accounted for 37% of deaths and neoplasms for 23%.81 Alarming reports on mortality because of infections or lymphomas must be interpreted with caution as a number of patients in such observational studies have often been treated off label for life-threatening conditions.87–88
Data on the safety of infliximab during pregnancy are scarce. Large antibodies do not pass through the placenta during the first two trimesters of pregnancy and therefore the risk for the foetus seems small. Small series of women who gave birth to a child after having been exposed to infliximab during pregnancy 89 do not show excess pregnancy complications or peri-partal problems. No increase in congenital abnormalities was observed.
Infliximab has also been intentionally used to control Crohn's disease during pregnancy.90 The authors expect that the benefits of infliximab in achieving and maintaining remission in mothers with Crohn's disease may outweigh the risk to the foetus of exposure to the drug. There is no data on the effects of infliximab therapy on lactation.
In our practice, we will continue systematic infliximab q8 during the first 20 weeks of the pregnancy and then interrupt therapy while maintaining azathioprine in women with aggressive Crohn's disease. After delivery, the decision to restart infliximab is taken on an individual basis. In women with history of disabling peri-anal disease, therapy with infliximab will always be restarted after the delivery.
Important data concerning the safety of infliximab therapy will be gathered using large registries. The TREAT registry in North America includes already more than 6000 patients. Half of the patients are treated with infliximab and the other half with Crohn's disease medications not including biologics. A preliminary analysis shows that serious infections and mortality in Crohn's disease is related especially to the use of steroids and not to the use of infliximab. Up to present, the rate of lymphoma or overall cancer was not higher in the cohort of patients treated with infliximab than in the patients treated with conventional treatments.
A similar registry in Europe (ENCORE) is actively recruiting patients, but data are not yet available.