- First Online:
- Cite this article as:
- Siddiqui, M.A.A. & Scott, L.J. Drugs (2005) 65: 2179. doi:10.2165/00003495-200565150-00014
- 136 Views
Infliximab (Remicade®) is a chimeric monoclonal antibody against tumour necrosis factor (TNF)-α that has shown efficacy in Crohn’s disease and rheumatoid arthritis with a disease-modifying activity and rapid onset of action. It is administered intravenously, generally in a schedule with initial infusions at 0, 2 and 6 weeks, followed by administration once every 8 weeks.
Infliximab is effective in the treatment of patients with moderately to severely active Crohn’s disease with an inadequate response to other treatment options or those with fistulising disease. In combination with methotrexate, infliximab reduced signs and symptoms and delayed disease progression in patients with active, methotrexate-refractory rheumatoid arthritis and in those with early disease. The drug was generally well tolerated. Recrudescence of tuberculosis infection and worsening of heart failure and demyelinating disease are among some of the concerns with anti-TNFα therapy, requiring cautious use of these agents in high-risk patients.
Current data suggest that infliximab may be cost effective, especially when long-term clinical outcomes and burden of the diseases are taken into account. More robust, prospective pharmacoeconomic studies are required to better ascertain the cost effectiveness of infliximab.
Direct head-to-head comparative trials of infliximab with other biological agents are not yet available and would be helpful in determining with greater certainty the place of infliximab in the management of these diseases. Nonetheless, infliximab, like other biological agents, is a valuable treatment option in patients with moderately to severely active Crohn’s disease (including fistulising disease) or rheumatoid arthritis (including early disease).
Infliximab acts by binding to both the soluble and the transmembrane forms of TNFα and antagonising the proinflammatory actions of TNFα. It also downregulates other proinflammatory cytokines in patients with Crohn’s disease or rheumatoid arthritis. In addition, in in vitro studies, infliximab decreased the migration of inflammatory cells to the site of inflammation by reducing levels of chemokines and endothelial adhesion molecules. Importantly, infliximab does not produce a generalised suppression of cellular immune function.
The pharmacokinetics of infliximab are linear within the therapeutic dosage range. The steady-state volume of distribution, clearance, mean residence time and elimination half-life are all dose independent in patients with rheumatoid arthritis or Crohn’s disease. Infliximab has a long elimination half-life (median 8–12 days) and is still detectable in plasma up to 12 weeks after the last dose. However, repeated treatment does not lead to any clinically relevant systemic accumulation. Concomitant administration of methotrexate in patients with rheumatoid arthritis may sustain serum infliximab concentrations for longer. The pharmacokinetic profile in paediatric patients with Crohn’s disease was similar to that in adult patients.
The usual treatment regimen for infliximab in clinical trials involved intravenous administration as a single infusion, followed by subsequent doses 2 and 6 weeks later and then every 8 weeks (or every 4 weeks in one trial) thereafter. Although a broad range of infliximab doses (1–20 mg/kg) was used in these trials, no dose-response relationship was evident and, generally, no significant added benefit was reported when using doses higher than 5 mg/kg in Crohn’s disease and 3 mg/kg in rheumatoid arthritis patients, which are the currently recommended dosages. However, some patients may benefit from increasing the dose up to 10 mg/kg; increasing the frequency of administration to once every 4 weeks may be useful in some rheumatoid arthritis patients. Infliximab treatment induced a rapid onset of clinical response both in Crohn’s disease (within 2 weeks) and rheumatoid arthritis (48 hours) patients.
In randomised, double-blind trials, single infusions of infliximab (dose range 5–20 mg/kg) were effective in the induction of response in patients with moderately to severely active Crohn’s disease. In addition, repeated treatment with infliximab 5 or 10 mg/kg during longer-term therapy (54 weeks) sustained the clinical remission achieved with single infusions. Clinical response was accompanied by endoscopic evidence of mucosal healing. Infliximab treatment was also associated with significant improvements in the health-related quality of life (HR-QOL) of patients. Similarly, in patients with fistulising disease, repeated treatment (54 weeks) with infliximab 5 mg/kg was effective in maintaining remission in patients who had responded to an initial infusion. Sustained closure and healing of fistulas was seen in these patients.
Limited data available from double-blind and noncomparative trials in paediatric patients with Crohn’s disease with or without fistulas indicate that single infusions of infliximab were effective in reducing disease activity and that repeated treatment may be effective in producing a clinical response or inducing and maintaining remission.
In a randomised, double-blind (54-week, followed by 54-week unblinded extension) trial in patients with active, methotrexate-refractory rheumatoid arthritis, infliximab (3 or 10 mg/kg) plus methotrexate was superior to methotrexate plus placebo in producing clinical response and attenuating radiographic progression of the disease. There was also a significant improvement in functional status and HR-QOL. Infliximab 3 or 6 mg/kg for 54 weeks, in combination with methotrexate, was also significantly more effective than methotrexate plus placebo in 1004 patients with early (≤3 year), methotrexate-naive rheumatoid arthritis with respect to improvements in signs and symptoms, disease progression, physical function and HR-QOL. The results of two smaller, double-blind, placebo-controlled studies using more sensitive imaging techniques (e.g. magnetic resonance imaging) have supported the notion that treatment with infliximab retarded disease progression in patients with early rheumatoid arthritis.
Infliximab was generally well tolerated for up to 102 weeks in clinical trials in patients with Crohn’s disease or rheumatoid arthritis, with the incidence of serious adverse events being generally similar in infliximab and placebo recipients. Infusion-related reactions are the most frequent adverse reactions to infliximab therapy. Although there appears to be no need for regular laboratory monitoring of infliximab therapy, patient monitoring, cautious use and/or discontinuation are recommended for other potentially serious adverse events possibly associated with TNFα antagonists as a class. These include tuberculosis and other serious infections, malignancy (particularly lymphoma), heart failure and CNS demyelination disorders.
Infliximab, like other biological agents, is a costly drug and generally more expensive than other available therapies for Crohn’ s disease. However, preliminary data have shown that it may be cost saving as a single-infusion induction regimen and cost effective as maintenance treatment of moderately to severely active Crohn’s disease. Moreover, pharmacoeconomic analyses have demonstrated that the incremental cost of infliximab maintenance therapy might be substantially offset by a reduction in other costs, such as those related to hospitalisation or surgery. In modelled cost-utility analyses, infliximab plus methotrexate was generally associated with acceptable incremental cost-effectiveness ratios relative to methotrexate monotherapy in patients with refractory rheumatoid arthritis. The cost effectiveness of infliximab in early disease has not been evaluated, but is expected to be improved when reductions with early treatment in the costs associated with the long-term disease outcomes are considered. Nevertheless, more robust pharmacoeconomic studies are needed before any conclusion may be drawn regarding the relative cost effectiveness of infliximab.
Crohn’s disease and rheumatoid arthritis are both chronic, inflammatory, incurable disorders of uncertain aetiology. Although they affect different body systems (gastrointestinal and skeletal, respectively), both are associated with significant morbidity and potentially lifelong treatment.[1,2]
Progress in understanding the pathogenesis of chronic inflammatory disorders has provided further insight into the role of proinflammatory cytokines, such as tumour necrosis factor (TNF)-α and interleukin (IL)-1, and led to the identification of these cytokines as potential therapeutic targets.[3,4] A new era in the management of these disorders began with the development of new biological therapies that selectively blocked the inflammatory cascade by antagonising cytokines. TNFα antagonists have been the most extensively studied biological agents in the above indications.[3,4]
Infliximab (Remicade®)1, a TNFα antagonist and the first biological agent approved for the treatment of Crohn’s disease, has had a significant impact on the standard treatment of the disease and is considered, along with other biological agents, a major advance in the therapy of rheumatoid arthritis. The use of infliximab in the treatment of Crohn’s disease,[5,6] rheumatoid arthritis[5–7] and ankylosing spondylitis has been reviewed previously.[5–8] Although infliximab has also been investigated in several other indications (e.g. ulcerative colitis and psoriatic arthritis; section 7), this review focuses on the use of infliximab in Crohn’s disease and rheumatoid arthritis.
2. Overview of Pharmacological Properties
Infliximab is a high molecular weight (≈149 100Da) chimeric monoclonal antibody against TNFα. The molecule comprises two components: a murine variable region (≈25%) and a human constant region (≈75%). In contrast to murine monoclonal antibodies, whose immunogenicity limits their use in humans, infliximab may have superior efficacy because of reduced immunogenicity and a prolonged half-life.
The pharmacological properties of infliximab have been reviewed previously.[5–7] This section provides a brief overview of the main pharmacological effects of infliximab in patients with Crohn’s disease or rheumatoid arthritis.
2.1 Pharmacodynamic Properties
TNFα is a regulatory and proinflammatory cytokine that plays a central mediator role in chronic inflammatory disorders including Crohn’s disease and rheumatoid arthritis. Infliximab binds with high affinity (association constant of 1010 L/mol) to both the soluble and the transmembrane forms of TNFα[11,12] (in contrast to etanercept that binds to the soluble form only) and inhibits binding of the cytokine to its receptors, thereby neutralising its biological activity. This is thought to be the primary mechanism of action of infliximab. In cultures of T lymphocytes from Crohn’s disease patients, infliximab was significantly (p = 0.004) more efficient than etanercept in downregulating the production of TNFα stimulated by Staphylococcus enterotoxin A alone or in combination with IL-12. TNFβ (lymphotoxin a) is not neutralised by infliximab.
Infliximab is also associated with complement- and antibody-dependent lysis of TNF-producing cells and may also cause apoptosis of T lymphocytes and monocytes. The latter may be of particular relevance in Crohn’s disease and may, in conjunction with other pharmacological differences, explain why infliximab, but not etanercept, is effective in this disease.
In general, intravenous infliximab caused downregulation of proinflammatory cytokines in patients with Crohn’s disease or rheumatoid arthritis.[5–7] This included reductions in the production of cytokines in biopsy samples (e.g. TNFα, interferon [IFN]-γ) and cytokine levels in sera (e.g. IL-1 and IL-6),[5–7] with most of the effects correlating well with the clinical response. For example, serum IL-6 levels in rheumatoid arthritis patients treated with infliximab rose above baseline values in patients who had relapsed after an initial response, but not in those who remained in remission.
Infliximab also decreased the migration of inflammatory cells to the site of inflammation by reducing levels of chemokines (e.g. monocyte chemoattractant protein-1) and endothelial adhesion molecules (e.g. intercellular adhesion molecule-1).[5–7]
Although transient changes in numbers of peripheral blood leukocyte subpopulations occur with infliximab therapy,[5–7] generalised suppression of cellular immune function has not been seen. In in vitro studies using cells from infliximab-treated Crohn’s disease patients, the ability of stimulated peripheral blood mononuclear cells to produce TNFα and IFNγ or the proliferative response of T lymphocytes to challenge with antigens and mitogens were not affected. The latter response was reported to be augmented following infliximab administration in rheumatoid arthritis patients.
Infliximab had pharmacological effects on several other biomarkers of Crohn’s disease and rheumatoid arthritis, including a decrease in plasma levels of secretory phospholipase A2 and an increase in nuclear factor-κB inhibitor (IκBα and IκBγ) levels in intestinal mucosa in Crohn’s disease patients, and decreases in serum matrix metalloproteinases-1 and -3, vascular endothelial growth factor and acute phase proteins (e.g. C-reactive protein) in patients with rheumatoid arthritis.
2.2 Pharmacokinetic Properties
Following multiple-dose administration, the median serum concentrations 1 hour post-dose in patients with Crohn’s disease receiving infliximab 5 mg/kg and in those with rheumatoid arthritis receiving infliximab 3 mg/kg were 110.5 μg/mL (week 22; five doses) and 69.7 μg/mL (week 14; four doses with concomitant methotrexate).
Serum infliximab concentrations are still detectable in most patients 10–12 weeks after the last dose (e.g. median 2.2 μg/mL at 12 weeks in one study).[20,21,24] However, there is no clinically relevant systemic accumulation on repeated treatment (for up to 2 years in rheumatoid arthritis patients) with infliximab 3 or 10 mg/kg at 4- or 8-week intervals.[9,25] Steady-state serum concentrations were reached after week 22 in patients with Crohn’s disease receiving maintenance doses of infliximab 5 mg/kg at 8-week intervals.
There are very limited data available on the metabolism and excretion of infliximab. No metabolite of infliximab has been detected in the urine and further studies have not been conducted to characterise the pathways of infliximab elimination.[9,20] It is expected that, like human IgG, infliximab is degraded and reabsorbed, for example, by reticuloendothelial cells in the body.
The median terminal half-life values for infliximab have ranged from 8 to 12 days following administration in patients with rheumatoid arthritis (dose 3–10 mg/kg) or Crohn’s disease (5 or 10 mg/kg).[17,22] Infliximab clearance increases and serum concentrations decrease in patients who develop antibodies to infliximab.[9,27] Concomitant administration of methotrexate may decrease the clearance of infliximab from serum in patients with rheumatoid arthritis, possibly by reducing the immunogenicity of infliximab.
In 21 paediatric patients (aged 11–17 years) with Crohn’s disease, serum infliximab concentrations were proportionate to single doses (1–10 mg/kg) and were detectable through week 8 in patients receiving infliximab 5 mg/kg. The pharmacokinetic profile in these patients was similar to that in adult patients with Crohn’s disease.
There are generally no major effects of age, weight or sex on clearance or volume of distribution of infliximab.[9,20] However, serum infliximab concentrations tend to be increased in Japanese patients with high body mass index. The effects of marked hepatic or renal impairment on clearance or volume of distribution are not known.[9,20]
3. Therapeutic Efficacy
The therapeutic efficacy of infliximab in the treatment of patients with Crohn’s disease[5,6] or rheumatoid arthritis[5–7] has been reviewed previously. This section incorporates data that have become available since the previous reviews. Wherever possible, the discussion focuses on large, randomised, double-blind, comparative trials; less well designed and retrospective studies have generally been excluded.
The usual treatment regimen used in all of these trials involved intravenous administration of infliximab as a single infusion (generally infused over 2 hours), followed, in multiple-dose trials, by subsequent doses 2 and 6 weeks later and then every 8 weeks (every 4 weeks in one trial).
3.1 In Crohn’s Disease
Several randomised, double-blind, placebo-controlled, 4- to 54-week trials have evaluated the efficacy of single[31,32] or multiple[24,27,33,34] doses of intravenous infliximab (dose range 5–20 mg/kg) in adult patients with Crohn’s disease (including those with fistulising disease) [sections 3.1.1 and 3.1.2]. In addition, a randomised, double-blind, single-dose trial has examined the clinical efficacy of infliximab 1–10 mg/kg in paediatric patients; other data in paediatric patients are from prospective, noncomparative, single- or multiple-[36–38] dose studies (section 3.1.3).
Trials in moderately to severely active Crohn’s disease (section 3.1.1) enrolled patients (n = 30–335) showing an inadequate response to prior conventional therapies (refractory disease), with baseline Crohn’s Disease Activity Index (CDAI) scores of 220–400 and a disease duration of ≥6 (or ≥3) months.[24,27,31,32] Patients were excluded if they had received previous treatment with infliximab or any other TNFα antagonist, any monoclonal antibody[24,31] or parenteral corticosteroids (within 4 weeks),[24,31,32] or were currently receiving, or had received within 3 months before enrolment, methotrexate, ciclosporin or experimental agents.[24,31,32] Other exclusion criteria included symptomatic stenosis or ileal strictures, proctocolectomy, total colectomy or stroma.[24,31,32] Those receiving stable dosages of mesalazine, azathioprine or mercaptopurine before enrolment were allowed to continue on stable dosages of these drugs during the trials. Dosages of corticosteroids were kept stable[31,32] or were permitted to be tapered.[24,27]
To be enrolled in trials of fistulising Crohn’s disease (section 3.1.2), patients (n = 94 or 195) were required to have single or multiple perianal (majority),[33,34] enterocutaneous[33,34] or rectovaginal draining fistulas of ≥3 months’ duration as a complication of the disease. Patients were permitted to receive concomitant therapy with stable dosages of an aminosalicylate, a corticosteroid, methotrexate, azathioprine, mercaptopurine or an antibacterial (also mycophenolate mofetil in one trial). Patients were excluded if they were currently receiving ciclosporin or had previously received infliximab.[33,34] Other exclusion criteria included the presence of intestinal stricture or abscess.[33,34]
Paediatric studies (section 3.1.3)[29,35–38] were small (n = 18–30) and involved patients aged 6–19 years who generally had moderate to severe, refractory Crohn’s disease (with or without fistulas; one study excluded patients with fistulas) and, where specified, a modified CDAI score of ≥200[29,37] or a paediatric CDAI score of ≥30.[29,35–37]
Primary efficacy parameters evaluated in all of these trials included the proportion of patients showing a clinical response,[24,31,33,35,37] time to loss of response,[27,34] the proportion of initial responders to therapy still in clinical (or inflammatory) remission at the end of the study[27,35,36,38] and improvement in paediatric CDAI score.
3.1.1 Adult Patients with Moderately to Severely Active Disease
Single infusions of infliximab (dose range 5–20 mg/kg) were effective in the treatment of moderately to severely active Crohn’s disease refractory to conventional therapy.[31,32] In addition, repeated treatment with infliximab 5 or 10 mg/kg during longer-term therapy sustained the clinical remission achieved with single infusions.
A rapid response to treatment with single-infusion infliximab 5–20 mg/kg was seen in this patient population, with symptomatic improvement occurring within 2 weeks. However, no dose-response relationship was evident for single infusions of infliximab in this dose range in two double-blind trials (n = 30 and 107). At 4 weeks, both the clinical response (65% vs 17%; p < 0.001) [primary endpoint] and clinical remission (33% vs 4%; p = 0.005) rates were significantly higher in infliximab (combined groups) than placebo recipients in the larger trial. Both trials demonstrated significant (p < 0.05) reductions (i.e. improvements) from baseline in CDAI scores in infliximab compared with placebo groups at 4 weeks (29–61% vs 6% in both trials).[31,32] In addition, CDEIS scores were significantly improved in patients receiving infliximab compared with placebo recipients (decrease from baseline of 58–61% vs 11%; all p < 0.05), indicating improvement in endoscopic lesions with infliximab treatment. These results were supported in a nonblind, multicentre trial in 25 Japanese patients that reported both clinical and radiological/endoscopic improvements in the disease after a single infusion of infliximab 1–10 mg/kg at 4 weeks.
Regular maintenance infliximab therapy was more effective than episodic retreatment in patients in ACCENT I, with response rates at week 30 of 62% (237/382) and 52% (98/188) [p = 0.024]; the clinical remission rates approached, but did not reach, a statistically significant difference (40% vs 32%).
Infliximab recipients were generally able to reduce the dosage of, or discontinue, their corticosteroid treatment. The proportion of infliximab (5 or 10 mg/kg) recipients who discontinued corticosteroids while in remission was more than three times higher than placebo recipients in remission at week 54 (29% vs 9%; p = 0.004). Among the week 2 responders, the reduction in corticosteroid dose was more rapid in infliximab 5 or 10 mg/kg than in placebo recipients, with the median corticosteroid doses reaching 0, 0 and 10 mg/day, respectively, by week 22 (statistical significance not reported).
Infliximab treatment was also associated with significant improvements in the HR-QOL of patients.[27,40] At 54 weeks, the mean IBDQ scores were significantly increased in infliximab 5 or 10 mg/kg maintenance groups compared with that in the placebo maintenance group (table III). These results were generally supported by the results of the SF-36 scores; the changes from baseline in the physical component summary scores for placebo and infliximab 5 or 10 mg/kg (2.5 vs 6.1 and 7.2; both p < 0.05 vs placebo) showed generally greater improvements than those in the mental component summary scores (2.0 vs 5.1 [nonsignificant] and 5.8 [p < 0.05]).
3.1.2 Adult Patients with Fistulising Disease
At 18 weeks in the smaller of the two trials, patients receiving infliximab 5 (n = 31) or 10 (n = 32) mg/kg as a three-dose regimen had significantly higher clinical response rates (primary endpoint) [68% and 56% vs 26%; both p ≤ 0.02] and complete response rates (55% and 38% vs 13%; both p ≤ 0.04) than placebo recipients (n = 31). Among patients achieving the primary endpoint, the onset of response in both infliximab 5 and 10 mg/kg recipients was three times faster than that in placebo recipients (median time to onset 14, 14 and 42 days, respectively); the median duration of response was broadly similar (84, 99 and 86 days, respectively).
Results of ACCENT II (A Crohn’s disease Clinical trial Evaluating infliximab in a New long-term Treatment regimen in patients with fistulising Crohn’s disease II) established the clinical benefits of maintenance treatment with infliximab (figure 2) in patients with fistulising Crohn’s disease who had responded to a three-dose induction regimen.
Patients receiving infliximab 5 mg/kg (n = 96) maintenance treatment experienced a significantly prolonged time to loss of response (primary endpoint; intent-to-treat analysis) compared with that in placebo maintenance recipients (n = 99) [median 40 vs 14 weeks; p < 0.001]. At 54 weeks, the clinical response (46% vs 23%; p = 0.001) and complete response (36% vs 19%; p = 0.009) rates were also significantly higher in infliximab (n = 91) than placebo (n = 98) recipients.
The median decreases from baseline in CDAI scores were significantly greater in infliximab than in placebo recipients at 30 (42 vs 16; p = 0.004) and 54 (15 vs 40; p = 0.04) weeks. In addition, significant improvements in HR-QOL were observed in the infliximab group compared with the placebo group, as evaluated using the IBDQ scores at 30 weeks (increases from baseline of 14 vs 4 units; p = 0.002) and 54 weeks (10 vs 5 units; p = 0.03).
3.1.3 Paediatric Patients
Single infusions of infliximab were effective in reducing disease activity in paediatric patients with Crohn’s disease with or without fistulas. In a double-blind trial (n = 21), paediatric CDAI scores in infliximab 1–10 mg/kg recipients had decreased by a median of 50% at week 2; the median percentage reductions ranged from 28% to 65% in the first 8 weeks (values estimated from graph; p-values not reported). In a noncomparative trial, infliximab 5 mg/kg produced a clinical response in 14 of 15 patients with refractory Crohn’s disease; 10 patients were in clinical remission at week 10.
In three noncomparative trials,[36–38] multiple-dose infliximab infusions (generally 5 mg/kg) were effective in producing a clinical response or inducing and maintaining remission in paediatric patients with active Crohn’s disease. Initial clinical response rates were as high as 92% (12/13 patients) at 4 weeks in one trial, with half of the responders sustaining the response at 20 months. In two trials, clinical remission rates were 90% (19/21; at day 45) or 56% (10/18; at 8 weeks); the latter also reported inflammatory remission in 67% (12/18) of patients. Among patients with fistulas, 56% (9/16) of patients showed good clinical response (fistula closure or cessation of drainage maintained for >4 weeks) in one trial.
3.2 In Rheumatoid Arthritis
The efficacy of multiple intravenous infusions of infliximab (3–10 mg/kg) in combination with methotrexate in the treatment of patients with rheumatoid arthritis has been evaluated in several randomised, double-blind, placebo-controlled trials (the 1 mg/kg dosage, found to be ineffective in a preliminary study, was not used in the later studies).[28,30,42–44] These were 26–54 weeks in duration and involved 20–1004 patients;[28,30,42–44] one pivotal study also had an unblinded, 54-week extension phase. Patients had methotrexate-refractory[28,30,44] or methotrexate-naive[42,43] rheumatoid arthritis; those enrolled in early-disease trials were also required to have a disease duration <1 or ≤3 years.[42,44]
Intravenous infliximab was administered at 0, 2 and 6 weeks, followed by infusions every 8 weeks (or every 4 weeks) for up to 102 weeks.[25,28,30,42–44] Patients were generally allowed to continue treatment with stable dosages of NSAIDs and oral corticosteroids, but other disease-modifying antirheumatic drugs (DMARDs) were not permitted during the study period. Concomitant methotrexate was used at specified dosages (7.5 mg/week), stable previous dosages or initial (7.5 mg/week[42,43] or stable previous) dosages uptitrated later on (maximum 20 or 25[43,44] mg/week).
Primary efficacy parameters included: clinical response rates defined as the proportion of patients with a 20% clinical improvement in signs and symptoms according to American College of Rheumatology criteria (ACR 20); ACR numeric index of overall response (ACR-N); duration of response on the Paulus 20% index; magnetic resonance imaging (MRI)-measured synovitis scores according to the Outcome Measures in Rheumatology Clinical Trials definitions; van der Heijde modification of the total Sharp score (vdH-Sharp); and the score on the Health Assessment Questionnaire (HAQ).[25,42] Other efficacy parameters included ACR 50 and ACR 70 clinical response rates, erosion and joint-space narrowing radiographic scores and SF-36 questionnaire scores.
Efficacy data are also available (as abstracts/posters) from a large (n = 1082), randomised, double-blind, placebo-controlled 54-week study (START [Safety Trial for rheumatoid Arthritis with Remicade Therapy]) designed to evaluate the safety of infliximab 3 or 10 mg/kg in combination with methotrexate compared with methotrexate alone in patients with active rheumatoid arthritis of >3 months duration despite treatment with methotrexate (mean duration 9.9 years).[45–47]
3.2.1 Patients with Methotrexate-Refractory Disease
In randomised, double-blind trials in patients with active, refractory rheumatoid arthritis, infliximab plus methotrexate was superior to methotrexate plus placebo in producing a clinical response and attenuating radiographic progression of the disease.[28,30] These data were supported by large (n > 500) noncomparative studies.[48,49]
In an early, well designed trial in 101 patients receiving background therapy with methotrexate, multiple doses of infliximab 3 or 10 mg/kg (but not 1 mg/kg) were significantly more effective than placebo in reducing the signs and symptoms of the disease. Furthermore, in 47 Japanese patients with methotrexate-refractory rheumatoid arthritis, infliximab 3 mg/kg produced significantly higher ACR 20 response rate than that seen in placebo recipients at 14 weeks (61% vs 23%; p = 0.001) in a double-blind study reported in the manufacturer’s prescribing information. The response rate in infliximab recipients was maintained at 54 weeks (53%). Patients received treatments at 0, 2 and 6 weeks and were evaluated at 14 weeks; infliximab recipients received four additional infusions at 8-week intervals.
Subsequently, a large double-blind trial (Anti-TNF Trial in Rheumatoid Arthritis with Concomitant Therapy [ATTRACT]) evaluated the efficacy of infliximab in patients with active rheumatoid arthritis despite treatment with methotrexate (for ≥3 months and at a stable dosage for ≥4 weeks before screening). At baseline, the median disease duration in the enrolled patients was 8.4 years and they were receiving methotrexate at a median dosage of 15 mg/week. At the end of first year (54 weeks), 259 patients continued in a 1-year extension of the study in which treatment allocation, but not infliximab dose and regimen, was unblinded; 94 of these patients had a gap of >8 weeks between treatments. Total duration of treatment, including the gap, was 102 weeks.
Clinical and Radiographic Responses
At all timepoints (30, 54 [table IV] and 102 weeks) in ATTRACT, ACR 50 and ACR 70 response rates were significantly (all p ≤ 0.011) better in infliximab than placebo groups. Although no clear dose-response relationship could be established, the ACR 50 response rates at 54 weeks in the 8-weekly infliximab 3 mg/kg group were significantly lower than those in 4- or 8-weekly infliximab 10 mg/kg groups (table IV).
The improved clinical response rates with infliximab were associated with significant beneficial effects on progression of joint damage as assessed by radiographic studies.[25,30,50] All four infliximab groups had no significant changes from baseline in the Sharp total scores at 54 weeks, whereas placebo recipients experienced a 9–10% increase; thus, a significant between-group difference in the change in scores from baseline was evident at 54 weeks (table IV). The changes in total scores remained significantly smaller in infliximab than in placebo recipients after 102 weeks’ treatment (−0.4 to +1.1 vs +12.6; all p < 0.001). Of note, interpretation of radiographic data was not expected to be affected in the unblinded period because the radiograph readers were kept blinded to the treatments and infliximab doses.
Similarly, the changes in erosion and joint-space narrowing scores showed significantly slower progression of disease in infliximab than in placebo recipients at 54 weeks (table IV), with the difference maintained at a significant level (p ≤ 0.002) at 102 weeks.
In the START trial, the 28-item disease activity score (DAS-28) in infliximab (3 or 10 mg/kg) recipients (n = 709) was significantly decreased compared with that in placebo recipients (n = 354) at week 22 (3.4 vs 4.4; p < 0.001); remission (DAS-28 score <2.6) rates were also significantly higher in infliximab than in placebo recipients (31% vs 14%; p < 0.001).
Effects on Functional Status and Health-Related Quality of Life
In patients receiving infliximab 3 mg/kg every 4 weeks or 10 mg/kg every 4 or 8 weeks in ATTRACT, the increase in HAQ and the physical component of the SF-36 scores were greater than those in placebo recipients at 54 weeks, indicating a significant improvement in functional status; the results with infliximab 3 mg/kg every 8 weeks did not reach statistical significance (table IV).
At 102 weeks, all infliximab groups demonstrated significant improvements in HAQ (median change 0.3–0.4 vs 0.1; p ≤ 0.006) and the physical component of the SF-36 (4.6–6.9 vs 2.8; p ≤ 0.011) scores compared with those in placebo recipients. Significant (all p ≤ 0.026) improvements were also observed in all the individual SF-36 subscales related to the physical aspects of HR-QOL (physical functioning, role-physical, bodily pain, general health) with all infliximab regimens compared with placebo. Additionally, vitality and social functioning subscales were also significantly improved (p ≤ 0.007). However, role-emotional and mental health subscales were not significantly affected, probably because their baseline values were not markedly abnormal in this patient population.
At week 22 in the START trial, there were significantly (all p < 0.05) greater improvements in SF-36 summary scores and all subscores in infliximab (3 or 10 mg/kg) than in placebo recipients. These improvements in the infliximab groups were maintained at 54 weeks, at which time the improvements in patients who switched (at week 22) from placebo to infliximab were similar to those in infliximab-initiated patients.
3.2.2 Patients with Early Disease
Post hoc analyses of ATTRACT data demonstrated that the baseline physical function of patients and the improvement after treatment were a direct function of the extent of baseline joint damage and that early treatment with infliximab plus methotrexate might be beneficial in the long term as it prevented disease progression. These findings prompted the evaluation of infliximab in the treatment of early rheumatoid arthritis in the ASPIRE (Active-controlled Study of Patients receiving Infliximab for treatment of Rheumatoid arthritis of Early onset) [table IV]. Patients were not eligible for inclusion if they had received DMARDs other than methotrexate within the last 4 weeks (or 6 months for leflunomide). The median disease duration at baseline was 0.6 years.
Two smaller trials have also examined the efficacy of infliximab in the treatment of methotrexate-naive, poor-prognosis (according to Persistent Inflammatory Symmetrical Arthritis scoring system) or methotrexate-refractory, early rheumatoid arthritis. Patients received infliximab 3 or 5 mg/kg or placebo at weeks 0, 2 and 6, and then every 8 weeks until week 46.
Clinical and Radiographic Response
In ASPIRE, the median improvements from baseline in ACR-N (co-primary endpoint) at week 54 were significantly higher in patients receiving infliximab 3 or 6 mg/kg than in placebo recipients (39% and 47% vs 26%; both p < 0.001). Similarly, the ACR 20, ACR 50 and ACR 70 response rates were significantly increased in infliximab compared with placebo recipients (table IV).
Infliximab recipients generally experienced attenuated progression of radiographic evidence of the disease compared with those receiving placebo. The change in total radiographic score (co-primary endpoint) was significantly smaller at 54 weeks in patients receiving infliximab 3 or 6 mg/kg than that in placebo recipients (table IV). The changes in erosion and joint-narrowing scores analysed individually were also significantly in favour of infliximab treatment, with the exception of the change in joint-space narrowing score with infliximab 6 mg/kg that did not reach statistical significance (table IV).
The results of two smaller, double-blind studies using more sensitive imaging techniques supported the notion that treatment with infliximab was beneficial in patients with early rheumatoid arthritis.[43,44] In 20 patients with very early (<12 months), methotrexate-naive disease, the reduction in the median total synovitis score (assessed using MRI scans) in the infliximab 3 mg/kg group was significantly greater than that in the placebo group at 14 weeks (2.1 vs 0.3; p < 0.05) [primary endpoint]; the difference was maintained at 54 weeks (decrease of 1.7 vs increase of 0.4; p < 0.05). In the other study in 24 patients with methotrexate-refractory, early disease (duration <3 years), the mean percent decreases from baseline in total synovial thickness (measured by high-frequency ultrasonography) [55% vs 14%; p = 0.014] and colour Doppler area (79% vs 27%; p = 0.017) at 18 weeks were significantly greater in infliximab 5 mg/kg than in placebo recipients.
Effects on Functional Status and Health-Related Quality of Life
Infliximab treatment led to significant improvements in physical function in patients with early rheumatoid arthritis in ASPIRE. Changes in HAQ (co-primary endpoint) scores in both infliximab groups showed significantly greater improvements than with placebo at 54 weeks (table IV). In addition, the proportion of patients experiencing a clinically significant improvement in HAQ score (i.e. ≥0.22 units) was significantly higher in infliximab 3 or 6 mg/kg than in placebo recipients (76% and 76% vs 65%; both p ≤ 0.004). The physical component score of the SF-36 was significantly improved in infliximab 6 mg/kg, but not 3 mg/kg, recipients compared with that in placebo recipients (table IV).
The tolerability profile of infliximab has been comprehensively reviewed previously in Drugs[7,8] and elsewhere.[5,6,53] This section provides an overview and focuses on the tolerability of infliximab as evaluated in four large, randomised, double-blind, placebo-controlled trials[27,30,34,42] (section 3). Additional information has been taken from other studies, where relevant, and from the manufacturer’s prescribing information that included data on 2779 patients (1304 with rheumatoid arthritis, 1106 with Crohn’s disease and 369 with other conditions).
The incidence of serious adverse events (see also sections 4.2 and 4.3) in patients receiving infliximab was similar to that in placebo recipients in four double-blind, 54-week trials in patients with Crohn’s disease or rheumatoid arthritis (11–28% vs 11–29%).[27,30,34,42] Serious adverse events after a single infliximab dose occurred with a frequency similar to that after multiple doses in ACCENT I (12% vs 10%).
In the paediatric double-blind trial (section 3.1.3), adverse events considered reasonably related to infliximab were noted in 4 (19%) patients: upper respiratory tract infection (n = 2), pancreatitis (n = 1) and sinusitis/appendicitis (n = 1). There was no infusion reaction and none of the patients developed antibodies to double-stranded DNA or to infliximab. A similar tolerability profile was seen in other paediatric trials discussed in section 3.1.3, with infusion reactions generally being the most common adverse events.[35–38] In one trial, reactivation of Epstein-Barr virus and development of antinuclear antibodies (ANA) and antibodies to double-stranded DNA were frequent, but disappeared within 6 months after infliximab discontinuation.
In Japanese patients with rheumatoid arthritis (n = 193) or Crohn’s disease (n = 24) who received infliximab in clinical trials, the most frequent clinical adverse reactions included fever (8%), increase in blood pressure (7%), headache (6%) and hot flushes (6%); major laboratory test abnormalities included urinary sediment (14%) and increases in liver enzymes (12–14%).
4.1 Infusion Reactions and Other Immune-Related Phenomena
4.1.1 Infusion-Related Events
Infusion-related events occurring during or within 1–2 hours after an infusion were the most frequent adverse drug reactions with infliximab.[9,20] These reactions occurred in approximately 20% of infliximab recipients compared with approximately 10% of placebo recipients. Nonspecific symptoms (e.g. fever or chills), cardiopulmonary symptoms (e.g. chest pain, hypotension or dyspnoea) and urticaria/pruritus were seen in 3%, 1% and <1% of infliximab infusions.[9,20]
Infusion reactions led to discontinuation of treatment in approximately 3% of patients and were considered serious (e.g. anaphylaxis) in <1% of infliximab recipients.[9,30,34,42] However, there were no long-term sequelae and symptoms were controlled with drug treatment (e.g. antihistamines), by slowing down the infusion rate or by discontinuation of infliximab.[9,20,54]
During the extended (102-week) ATTRACT trial, approximately 3% and 2% of infliximab or placebo infusions were associated with an infusion reaction. Furthermore, a 3-year follow-up of patients with rheumatoid arthritis or Crohn’s disease involved in clinical trials showed that infusion-related reactions occurred in 4.8% (0.5% serious) of 4797 infliximab infusions and 2.1% (0% serious) of 2121 placebo infusions; 1.9% and 0% of infusions were discontinued because of these reactions.
Repeated infusions were generally not associated with an increase in the incidence of reactions;[9,27] instead, reactions were less likely to occur with subsequent than with initial infusions.[20,50] However, in ATTRACT, one patient experienced a serious infusion reaction after receiving his fourteenth infliximab infusion, although this patient had not shown any reaction to the previous infusions.
Patients were at an approximately 2- to 3-fold increased risk of experiencing an infusion reaction if they developed antibodies to infliximab[9,27,34,56] (section 4.1.3), whereas concomitant use of immunosuppressant agents was associated with a decrease in the incidence.[9,27,34,56,57]
4.1.2 Delayed Hypersensitivity
Delayed hypersensitivity reactions were seen in 10 of 40 patients with Crohn’s disease 3–12 days after they were retreated with infliximab 5 mg/kg following a gap of 2–4 years from their initial infliximab treatment (data available as an abstract). These included myalgia, polyarthralgia, fever, rash, pruritus, oedema of the face, hands or lips, dysphagia, urticaria, sore throat and headache, and required hospitalisation and medical therapy in six patients. Of note, the previous infliximab treatments in nine of these patients involved an investigational liquid formulation and only one patient received a lyophilised formulation.
Serum sickness-like reactions have been reported with infliximab.[9,20,27] The incidence of these events (generally after 1–5 maintenance infusions) in placebo or infliximab 5 or 10 mg/kg recipients was 2%, 3% and 3%, respectively, in the ACCENT I trial. A similar incidence (3%) of serum sickness-like reactions at least possibly related to infliximab was reported in 500 patients with Crohn’s disease who received the drug in a clinical practice safety study.
4.1.3 Immunogenic Reactions
Infliximab treatment may be associated with the production of autoantibodies and antibodies to infliximab. In patients receiving maintenance infliximab therapy following the induction regimen, ANA and antibodies against the double-stranded DNA developed in a higher proportion of infliximab (34–68% and 7–34%) than placebo (11–35% and 0–11%) recipients after 54 weeks (all p ≤ 0.013; p-values not reported in one trial).[27,30,34,42] Between week 54 and 102 in the ATTRACT trial, the proportion of patients with ANA or anti-DNA antibodies increased by approximately 13% and 5% in infliximab recipients compared with increases of 6% and 0% in those receiving placebo. Approximately 10% of patients who received a three-dose induction regimen followed by maintenance treatment through 1–2 years of infliximab treatment developed antibodies to infliximab. In Crohn’s disease patients, retreatment after a gap of >16 weeks was associated with an increase in the incidence of antibodies to infliximab. Nevertheless, the antibody titre has generally been low.[9,30]
An antibody titre of ≥8.0 μg/mL before an infusion was associated with a reduced duration of response compared with a titre of <8.0 μg/mL (35 vs 71 days; p < 0.001) in a study in 125 consecutive patients with luminal or fistulising Crohn’s disease. Patients had received a single- or three-dose induction regimen of infliximab 5 mg/kg followed by further infusions on relapse in responders over a mean period of 10 months. The risk of the antibody titre being ≥8.0 μg/mL was at least 2.4-fold lower in patients receiving than in those not receiving concomitant immunosuppressants (p < 0.001). Similarly, among patients not receiving methotrexate in a double-blind study, 53%, 21% and 7% of infliximab 1, 3 or 10 mg/kg recipients developed antibodies compared with 15%, 7% and 0%, respectively, of patients who also received methotrexate. It should be noted that the incidence of antibody development was inversely related to infliximab dose, which was also the case in the ASPIRE trial, with 15% and 7% of infliximab 3 or 6 mg/kg recipients developing antibodies.
4.2 Serious Infections
In clinical studies, 36% and 25% of infliximab or placebo recipients (average follow-up periods of 51 and 37 weeks) had infections that required treatment. Respiratory (e.g. sinusitis) and urinary tract infections were the most common infections reported. The frequency of serious infections (e.g. pneumonia, abscess, sepsis and tuberculosis) in patients receiving infliximab (≤8%)[27,30,34,42] was generally similar to that in placebo recipients in the double-blind, 54-week trials,[27,30,34] with the exception of the ASPIRE where serious infections were more common with infliximab 3 or 6 mg/kg than with placebo (6% and 5% vs 2%; both p ≤ 0.04). Fatalities have occurred as a result of these infections.[9,20]
Infliximab treatment may be associated with the development of tuberculosis.[9,20,60] In an analysis of all reports of tuberculosis received by the MedWatch spontaneous reporting system of the US FDA, 70 reports of tuberculosis during or after infliximab therapy were received after 147 000 patients had been treated (121 000 in the US); 56% had extrapulmonary, including disseminated (24%), disease. Seventeen reports were from within the US, 45 from Europe and eight from the rest of the world. The median time from initiation of infliximab therapy to detection of tuberculosis was 12 weeks (range 1–52 weeks). A later update in a review increased this figure to 172 after 198 235 patients had been treated with infliximab, corresponding to 227 559 patient-years of exposure; 97% of cases occurred within 7 months of treatment initiation. Although some of the tuberculosis infections have been seen in patients with no known history of the disease, most occurred in patients with a previous history of tuberculosis, suggesting recrudescence of latent disease.
Opportunistic infections (e.g. histoplasmosis, Pneumocystis jiroveci [previously carinii] pneumonia and aspergillosis) have also been reported in patients receiving infliximab.[9,20,60,61] In the 54-week ACCENT II in patients with fistulising Crohn’s disease, 12% of infliximab versus 17% of placebo recipients developed a new fistula-related abscess.
Preliminary data from the START trial (section 3.2) have shown the risk of serious infections (primary safety variable) with infliximab 3 mg/kg to be similar to that with placebo; infliximab 10 mg/kg was associated with a higher relative risk than placebo.
According to published reports of the double-blind, 54-week trials that included long-term follow-up periods (e.g. 3 years in ATTRACT), 6[27,34] and 13[25,30,42] malignancies were seen in infliximab-treated patients with Crohn’s disease or rheumatoid arthritis compared with two and one in placebo recipients. Nevertheless, this and further clinical experience has shown that, with the exception of lymphomas, the incidence of malignancies in infliximab recipients has been similar to what would be expected in an equivalent population not receiving the drug.
Although not certain, there is some speculation that lymphomas may be associated with the use of TNFα antagonists (including infliximab); more robust studies are needed for a definite conclusion.[9,62,63] Among infliximab recipients, lymphomas developed with an incidence of 0.08% and 0.12% of patient-years of follow-up in patients with rheumatoid arthritis and in the combined rheumatoid arthritis and Crohn’s disease patients, corresponding to a 3- and 5-fold increased incidence compared with an equivalent general population. In addition, noncutaneous malignancies were seen in 23 patients, with breast and colorectal cancers and melanoma being most common. These data were based on a long-term follow-up (median duration 1 year) of 3469 patients with rheumatoid arthritis or Crohn’s disease who received infliximab in clinical trials.
4.4 Other Adverse Events
The use of TNFα antagonists was associated with heart failure in 47 cases (18 with infliximab) reported to the FDA MedWatch programme; 38 of these (12 in infliximab recipients) were new-onset heart failure. Furthermore, in a randomised, double-blind trial in patients with moderate-to-severe (New York Heart Association class III to IV) heart failure, the risk of all-cause mortality or heart failure-related hospitalisation was significantly higher in infliximab 10 mg/kg (but not in 5 mg/kg) recipients than in those receiving placebo (hazard ratio 2.84; 95% CI 1.01, 7.97; p < 0.05). In this trial, patients received three infusions (at 0, 2 and 6 weeks) of placebo (n = 49), or infliximab 5 (n = 50) or 10 (n = 51) mg/kg; 13, 10 and 20 patients, respectively, were hospitalised for any reason after 28 weeks, with an all-cause mortality of four, four and eight after 1 year.
Hepatic events, ranging from mild or moderate increases in ALT and AST levels to severe hepatic reactions (e.g. acute liver failure) that were fatal or necessitated liver transplantation, have been seen in infliximab-treated patients.[9,66] Severe hepatic reactions occurred in approximately 38 patients (three in controlled clinical trials and 35 in the voluntary postmarketing reports) after approximately 576 000 patients had been treated with infliximab worldwide since its first approval in the US in 1998. However, a causal relationship with infliximab has not been established.[9,66]
Rare cases of neurological adverse events have been reported in association with infliximab and other TNFα antagonists.[9,67] These included optic neuritis, seizure and new onset or exacerbation of clinical symptoms and/or radiographic evidence of demyelinating disorders, including multiple sclerosis.
The outcomes of pregnancy in women with Crohn’s disease or rheumatoid arthritis after infliximab exposure was similar to that in the general population. In an analysis of pregnancy outcomes for 96 women who were directly exposed to infliximab (including 53 within 3 months of conception; 30 during the first trimester), 67% had live births, 15% miscarried and 19% had to undergo therapeutic termination of pregnancy. In comparison, the incidences in a cohort of general population were 67%, 17% and 16%, respectively.
5. Pharmacoeconomic Issues
This section provides a brief overview of the pharmacoeconomic analyses of infliximab used in the treatment of Crohn’s disease and rheumatoid arthritis. These have been comprehensively reviewed previously.[69,70]
Limited data are available on the pharmacoeconomics of infliximab therapy in Crohn’s disease (section 5.1), whereas a large number of studies have been conducted based on trials in patients with rheumatoid arthritis (section 5.2). Nevertheless, prospective, well designed pharmacoeconomic studies for infliximab are not available. Also, in the absence of direct comparative data, studies comparing infliximab with other therapies have generally used data from different trials, with differences in patient populations and outcomes. Most of the studies are available as abstracts only.[71–79]
5.1 In Crohn’s Disease
Two modelled studies, conducted by the same group of investigators, concluded that single-infusion infliximab was cost saving in certain scenarios and maintenance treatment with infliximab was likely to be cost effective in patients with refractory Crohn’s disease. Compared with placebo, single-infusion infliximab therapy was cost saving, assuming remission in infliximab recipients was similar to that in surgically treated patients; marginal cost-effectiveness ratios would be $US14 200 and $US40 000 per quality-adjusted life-year (QALY) gained (year of costing not stated) if infliximab produced remission rates similar to medically induced remission rates or those seen in patients with mild disease. Infliximab maintenance treatment among initial responders was associated with incremental cost-effectiveness ratios of $US47 000 (dose 5 mg/kg) and $US183 200 (10 mg/kg) per QALY gained in Canada (year of costing not stated). Both incorporated data from randomised, double-blind trials[27,31] into a Markov model; discounting was done at a rate of 3% annually.[71,72]
The incremental cost-utility ratios (direct costs only) for infliximab (5 mg/kg; episodic reinfusions for relapse or maintenance therapy every 8 weeks) relative to a strategy involving surgery followed by conventional therapy were €63 701 and €784 057 per QALY (year of costing not stated) gained in patients with severe Crohn’s disease. The lifetime cost-utility analysis, conducted from the third-party payer perspective, used a Markov model and was based on the results of a single-dose study and other published data, and experts’ opinion. From the same perspective, a three-infusion regimen of infliximab was found to be more expensive than a combination of mercaptopurine and metronidazole in a modelled cost-utility analysis involving a hypothetical cohort of patients with fistulising Crohn’s disease (reviewed previously).
Infliximab treatment of fistulising or nonfistulising Crohn’s disease patients was associated with decreases in healthcare resource utilisation. Post hoc analyses of the ACCENT I and II trials showed that, among initial responders to infliximab, maintenance treatment (infliximab 5 mg/kg every 8 weeks) was associated with significantly (p < 0.05) lower incidences (33 vs 20 and 31 vs 14 per 100 patients) and durations (2.3 vs 1.6 and 2.4 vs 0.8 days) of hospitalisation than those with placebo over a 1-year period. In addition, compared with placebo in ACCENT II, maintenance infliximab treatment led to significant (all p < 0.05) reductions in the per 100 patients rates of all (169 vs 83) and inpatient (45 vs 10) surgeries/procedures and major surgeries (13 vs 2) among initial responders. In another analysis by the same investigators, among unemployed patients at baseline, employment rate improved in patients who were in clinical remission at week 54 compared with those who were not (31% vs 16%; p < 0.05).
5.2 In Rheumatoid Arthritis
Several cost-utility analyses of infliximab plus methotrexate relative to methotrexate alone in patients with active refractory rheumatoid arthritis have been conducted incorporating efficacy data from ATTRACT into HAQ-based Markov models.[74–77,84,85] In addition, a few cost analyses are also available, most of which[75–77,84–87] have been reviewed previously.
The cost-utility analyses in the US,[75,84] Europe (country not reported), Portugal, Spain and the UK used the same model, which was different from the one used in the UK/Swedish concurrent analysis. Analyses included direct healthcare costs with or without indirect costs of lost productivity; discounting was done at a rate of 3% annually. Therapies were assumed to be cost effective if they had an incremental cost per QALY gained of <$US50 000,[75,84] <€50 000[76,77] or <€30 000.
When only direct healthcare costs were considered, infliximab 3 mg/kg every 8 weeks in combination with methotrexate was cost effective relative to methotrexate when given over 54 or 102 weeks or a lifetime.[74,76,77,88] The incremental costs per discounted QALY gained were $30 500 (54 weeks) and $38 200 (102 weeks) in the US, $US38 700 (lifetime) in Europe (country not reported), €39 500 (lifetime) in Portugal, €27 300 (lifetime) in Spain and £33 618 (lifetime) in the UK. The year of values was 1998 in the US (54-week) and 2000 in the UK analyses and not reported in the others. There was further improvement in the incremental cost per QALY gained when, in addition to direct costs, societal (loss of productivity) costs were included in the analysis[74,77,84] or when the benefits of infliximab on radiographic progression were taken into account. For example, in one of the US analyses, the incremental cost per QALY was reduced by two-thirds (to $US9 100) when indirect costs were included in the analysis in addition to the direct costs.
In UK/Swedish analysis, the incremental costs of infliximab (3 mg/kg every 8 weeks) plus methotrexate per QALY gained relative to methotrexate monotherapy for 1 or 2 years were €28 600–56 100 and €3440–48 200 (year of values not reported) when direct costs and direct plus indirect (loss of productivity) costs were considered over a 10-year time horizon. The infliximab regimen was cost saving (considering both direct and indirect costs) relative to methotrexate alone in the Swedish study when an alternative model was used that eliminated the placebo effect, whether or not a loss of effect after discontinuation was assumed to occur.
In a large state Medicaid programme in the US, the median actual cost per patient for infliximab therapy was $US16 399 per year. From the healthcare payer perspective, the total 1-year costs related to rheumatoid arthritis care were higher with infliximab than with etanercept ($US20 263 vs $US16 534 per patient; 2001 values) among 4069 patients in a US managed-care database. Similarly, in a 1-year cost analysis conducted from a societal perspective, the total cost of treatment with infliximab (3 mg/kg every 8 weeks) plus methotrexate was greater than that of etanercept 25mg twice weekly ($US18 046 vs $US12 648 per patient; 1999 values) in The Netherlands. The annual per-patient drug costs being similar for infliximab and etanercept ($US12 610 and $US12 534), the difference in total costs was mainly driven by the additional costs associated with administration of intravenous infliximab in an outpatient facility.
By contrast, a cost-analysis conducted from the perspective of the Spanish Public Healthcare System showed that infliximab therapy was cost saving versus etanercept therapy over a time horizon of 2 years. The total costs per patient associated with infliximab and etanercept treatments were €11 238 and €13 066 in the first year and €8 166 and €13 520 in the second year (2002 values).
6. Dosage and Administration
In the US, EU and Japan, the recommended initial doses of infliximab in patients with rheumatoid arthritis or Crohn’ disease (moderately to severely active or fistulising) are 3 or 5 mg/kg, respectively, given as an intravenous infusion. The usual recommended regimen involves administration of the infusions at 0, 2 and 6 weeks followed by infusions every 8 weeks. Maintenance treatment in Crohn’s disease beyond the initial treatment (single dose in moderately to severely active and a three-dose regimen in fistulising disease) is not mentioned in the Japanese prescribing information. Infliximab should be given only in combination with methotrexate in patients with rheumatoid arthritis.[9,20,26]
Readministration of infliximab within 16 weeks of the last infusion may be considered if the signs and symptoms of disease recur after an initial response.
The US prescribing information recommends that the doses may be increased to 10 mg/kg in rheumatoid arthritis patients who have an incomplete response and in Crohn’s disease patients who respond and then lose their response. In addition, increasing the frequency of maintenance treatment to administration every 4 weeks may be considered in rheumatoid arthritis patients.
For comprehensive information on dosage and administration, precautions and contraindications, the local manufacturer’s prescribing information should be consulted.
7. Place of Infliximab in the Management of Crohn’s Disease and Rheumatoid Arthritis
7.1 Efficacy Considerations
7.1.1 Crohn’s Disease
Crohn’s disease is a chronic, debilitating disease that generally affects individuals in the second and third decades of life. Northern Europe, the UK and North America are the regions historically associated with a high incidence and prevalence of Crohn’s disease. According to an estimate, 10 000–47 000 cases of Crohn’s disease occur annually in North America and Europe, affecting approximately 1 and 1.4 million people in the US and the EU. Nevertheless, other areas of the world, including Asia, have shown an increasing incidence and prevalence of Crohn’s disease. In Japan, the disease occurs at a rate of 1.2 per 100 000 person-years, affecting 13.5 per 100 000 individuals (1998 estimate).
Crohn’s disease potentially affects any part of the gastrointestinal tract and classically follows a relapsing and remitting course. Since the disease is not curable, the goal of therapy is to eliminate symptoms and promote well-being with the least possible unwanted effects while avoiding long-term sequelae. Disease severity, disease location and the presence of complications generally determine the choice of medical therapy.
The treatment strategy for Crohn’s disease consists of two phases: the induction of remission and the maintenance of remission. The conventional choice of mesalazine and antibacterials for remission induction in patients with mild-to-moderate disease, as recommended in US guidelines, was challenged based on published evidence, and a new treatment algorithm was proposed. Budesonide or sulfasalazine have been recommended as agents of choice, depending on the location of the disease, with conventional corticosteroids as second-line options in cases of treatment failure. However, recent British guidelines still recommend mesalazine for mild ileocolonic disease. Oral corticosteroids (e.g. budesonide or prednisone) are the first-line treatment options in patients with moderate-to-severe[97,99] or mesalazine-resistant mild-to-moderate ileocolonic disease, but are associated with well known adverse effects. Immunosuppressive agents (e.g. azathioprine, mercaptopurine or methotrexate) have shown adjunctive benefits in patients who become corticosteroid-resistant or -dependent and azathioprine and mercaptopurine are recommended as first-line agents of choice for chronic, corticosteroid-dependent, active Crohn’s disease in British guidelines. However, these drugs have a slow onset of action and are associated with serious adverse effects in some patients.
While pharmacotherapy is the mainstay of treatment, nutritional therapy is often indicated as supportive therapy, especially in patients who are unable to maintain nutritional requirements. Surgery is largely palliative and generally reserved for patients with drug-refractory, severe-to-fulminant disease or those with complications such as fibrotic strictures, or who are not able to tolerate adverse effects of drug treatment. However, despite the fact that up to two-thirds of patients eventually require surgical intervention, relapse after a resection is common and improvement of symptoms is usually temporary.
The most widely encountered complication of Crohn’s disease is the development of perianal fistulas (e.g. 92% of patients with colonic disease and rectal involvement), with incidences of up to 23% reported in population-based studies. Antibacterials, such as metronidazole with or without ciprofloxacin, are the mainstay of treatment of both simple (as a first-line option) and complex (in combination with other medical agents or surgery) perianal fistulas, while azathioprine or mercaptopurine are the other important options.[99,101] Surgical drainage is required in perianal/perirectal abscesses.
There is accumulating evidence that ongoing maintenance therapy is beneficial in many patients with active Crohn’s disease (including fistulising disease) who have responded to initial therapy. Despite being very effective in inducing remission, corticosteroids are not effective at maintaining remission;[97,99,100] >50% of patients who receive corticosteroids eventually become corticosteroid-resistant or -dependent.[97,100] After remission induction with the corticosteroids or ileocolonic resections, azathioprine or mercaptopurine may be used for tapering corticosteroid dose and maintenance of remission.[97,99] Toxicity is a potential limitation of this therapy.
The advent of infliximab has provided a useful second-line treatment option in patients with Crohn’s disease (including fistulising disease). It is the first biological agent recommended for use in this indication. Therapy with intravenous infliximab is recommended in patients with moderately to severely active, refractory Crohn’s disease.[97,99] It may be used as an adjunct, or as an alternative, to corticosteroids in patients in whom these are ineffective or contraindicated. For the treatment of fistulising disease, the British guidelines restrict its use, in conjunction with immunomodulation and surgery, to patients with refractory disease. By contrast, US guidelines have found insufficient evidence for recommending a preference for any of the treatment options for simple perianal fistulas and suggest that infliximab may be the initial treatment of choice in complex perianal fistulas. Nevertheless, in double-blind trials, infliximab 5 mg/kg was effective in inducing a clinical response in patients with moderately to severely active, refractory Crohn’s disease after a single infusion (section 3.1.1) and in those with fistulising disease with a three-dose regimen (section 3.1.2). Infliximab had a rapid onset of action, with symptomatic improvement observed within 2 weeks of treatment. This is an important advantage over immunosuppressive agents, which require 3–6 months for any visible signs of improvement.
Furthermore, repeated treatment with infliximab 5 mg/kg every 8 weeks prolonged the duration of clinical response in patients with refractory Crohn’s disease, including those with fistulising disease (who experienced a sustained closure and healing of fistulas) [sections 3.1.1 and 3.1.2]. Infliximab therapy demonstrated a corticosteroid-sparing effect, and was more effective, with a lower incidence of antibody formation to infliximab, in a regular maintenance regimen than in an episodic treatment schedule. Importantly, infliximab treatment resulted in reductions in hospitalisations and surgical procedures and improvements in HR-QOL. In addition, endoscopic studies showed that infliximab had a beneficial effect on mucosal healing. Thus, recent guidelines have included infliximab as one of the treatment options for maintenance of remission in patients with Crohn’s disease (including those with fistulising disease) who have responded to an initial infusion.[99,101] However, patients who do not show a response by week 14 are unlikely to respond to further infusions; discontinuation of infliximab therapy should be considered in such patients.
An understanding of the factors predicting a response to infliximab is essential to rationalise therapy and thereby increase the effectiveness of treatment. Current evidence indicates that concurrent immunomodulator therapy may increase the response to infliximab therapy. This may be an indirect effect since concomitant immunosuppressant agents have been shown to decrease the risk of an antibody titre being ≥8.0 μg/mL, a factor associated with reduced duration of response (section 4.1.3). Also, there is preliminary evidence that infliximab plus ciprofloxacin may be more effective than infliximab alone in patients with fistulising disease, but this needs to be confirmed in larger trials. Furthermore, although there was no clear dose-response relationship in clinical trials, remission rates were numerically increased with an increase in dose in patients with moderate-to-severe Crohn’s disease (section 3.1.1). This, in turn, is reflected in the US prescribing information that recommends that the infliximab dose may be increased from 5 to 10 mg/kg in patients who lose their response (section 6).
Cost of therapy is an important consideration in the management of Crohn’s disease, as the disease is an enormous burden on the healthcare system. The average annual medical cost of Crohn’s disease was estimated to be >$US9000 per patient (1996 values), resulting in an estimated cost of over $US1.7 billion to the entire US population annually. Of note, a large proportion of that cost (80%) was due to surgery and hospitalisation, while drugs and outpatient care accounted for 10% and <3% of the cost. Infliximab is a costly drug and generally more expensive than other available therapies for Crohn’s disease (section 5.1). However, it is a second-line option and is reserved for use in patients where conventional, less expensive treatment options have been exhausted or are contraindicated. Also, it may be cost saving as a single-infusion induction regimen and cost effective as maintenance treatment of moderate-to-severe Crohn’s disease. Moreover, the cost of infliximab therapy may be justified if other costs, such as those related to hospitalisation, can be reduced. Indeed, analyses of the ACCENT I and II trials have demonstrated that the incremental cost of infliximab maintenance therapy might be substantially offset by a reduction in these aspects of care.[73,82] Nevertheless, more robust pharmacoeconomic studies are needed before any conclusion may be drawn regarding the relative cost effectiveness of infliximab in Crohn’s disease.
Although the safety and efficacy of infliximab have not been established in large, well controlled trials in paediatric patients with Crohn’s disease, data from small studies, including a double-blind trial (section 3.1.3), suggest that infliximab may be an important second-line treatment option in these patients.
7.1.2 Rheumatoid Arthritis
Rheumatoid arthritis, the most common inflammatory polyarthropathy, affects ≈1% of the world’s adult population. In the US alone, over two million people are affected by the disease. Generally, the onset of rheumatoid arthritis is seen in the middle age (fourth to seventh decade of life), with a 2.5-fold higher prevalence in women than in men.
Early diagnosis and treatment is an integral part of the management of rheumatoid arthritis, with the aims of reducing and controlling symptoms of joint pain and inflammation, minimising loss of function and reducing joint damage and disability. The ultimate goal of treatment of rheumatoid arthritis is to induce a complete remission, which is rarely achieved in clinical practice. Current guidelines have adopted an aggressive approach, recommending DMARDs for upfront use within 3 months of the onset of disease if symptoms persist despite NSAID treatment or if there is radiographic evidence of joint damage. NSAIDs and low-dose oral corticosteroids are not recommended as sole therapy, but can be used as adjuncts to reduce joint pain and inflammation.
Conventional DMARDs are used both as monotherapy and in combination therapy, but the latter has proved extremely valuable in controlling the disease as monotherapy often fails in this respect. Sulfasalazine, hydroxychloroquine and methotrexate have generally been the first choice agents; methotrexate or combination therapy is preferred for patients with very active or poor-prognosis disease. However, there are several drawbacks of the traditional DMARDs, including a slow onset of action (1–6 months), lack of adequate long-term response to therapy in most patients and the need for monitoring the therapy because of adverse effects that eventually lead to discontinuation in many patients.
More recent additions to the DMARDs are leflunomide and the biological agents (the TNFα antagonists infliximab, etanercept and adalimumab, and the IL-1 inhibitor, anakinra).[105,107] Leflunomide is now well established as an alternative to methotrexate monotherapy and, in cases of incomplete response to full-dose methotrexate, has also shown benefits in combination with methotrexate (although the combination is associated with a several-fold higher risk of hepatotoxicity than with leflunomide alone). However, leflunomide also has some of the disadvantages of traditional DMARDs (slow onset and need for monitoring) and is a potential teratogen.
Biological agents, considered to be a major advance in the therapy of rheumatoid arthritis, offer a further treatment option for this disease and may be used alone or in combination with other DMARDs.[105,107] In contrast to the other DMARDs, TNFα antagonists have a rapid response (1–2 weeks). Infliximab, for example, has shown beneficial effects on symptoms within 48 hours of administration in patients with rheumatoid arthritis. The IL-1 inhibitor anakinra has a delayed onset of action (3 months) compared with TNFα antagonists. Treatment guidelines recommend the use of biological agents, such as infliximab, in patients with active rheumatoid arthritis when an adequate trial of the standard therapy (including methotrexate) has failed or is not tolerated.[105,107,108] Head-to-head comparative trials of biological agents are not yet available. Indirect evidence suggests that anakinra may be less effective than TNFα antagonists in reducing the signs and symptoms of rheumatoid arthritis.[109,110] In the absence of direct comparative trials, it is not possible to determine if there are any differences in efficacy among anti-TNF agents.
Infliximab, when used in combination with methotrexate over 2 years (ATTRACT), improved signs and symptoms in patients with methotrexate-refractory rheumatoid arthritis, and also reduced radiographic progression and improved physical function and HR-QOL (section 3.2.1). In addition, data from a large, double-blind trial (ASPIRE) have shown that infliximab plus methotrexate given for 1 year was effective in early (≤3 years from onset), methotrexate-naive rheumatoid arthritis (section 3.2.2). Importantly, patients in this trial were not required to be refractory to DMARDs. Furthermore, double-blind studies have shown that infliximab reduced progression of early disease as assessed by techniques more sensitive than radiography, such as MRI (which, for example, can identify joint damage in rheumatoid arthritis as early as 4 months) and high-frequency ultrasonography (section 3.2.2). A long-term (2-year) follow-up of patients with very early rheumatoid arthritis in one of these studies indicated that there was a potential for maintaining the beneficial effects of infliximab even after the discontinuation of treatment. These results were supported by the 2-year follow-up data (available as an abstract) from the BeST (Behandel-STrategieen) study in which half (52%) of the 128 patients with early rheumatoid arthritis (duration ≤2 years) were able to discontinue infliximab after a mean of 1 year without disease flare. However, more robust studies with longer follow-up periods are required to substantiate any claims of sustained disease remission after infliximab discontinuation.
Given the aggressive approach currently recommended in treatment guidelines in view of the rapid disease progression, it is likely that infliximab and other TNFα antagonists, with their significant disease-modifying activities, may eventually become a first-line treatment option in patients with early disease. However, cost considerations and lack of long-term safety/efficacy data have meant that the guidelines recommend only a limited use of these agents as the first DMARDs in treatment-naive patients. Recent British guidelines anticipate that circumstances necessitating the use of TNFα antagonists as a first-line therapy would be rarely encountered in clinical practice (of note, these guidelines have not included evidence from the ASPIRE trial). Nevertheless, infliximab, in combination with methotrexate, has been approved as first-line therapy for the treatment of patients with active and progressive, DMARD-naive disease and, along with other TNFα antagonists, provides another useful treatment option in these patients.
Infliximab should be used only in combination with methotrexate, whereas etanercept and adalimumab can also be used as monotherapy. Thus, the latter two may be attractive options in patients who do not tolerate methotrexate well. Results of the iRAMT (infliximab Rheumatoid Arthritis Methotrexate Tapering) trial have suggested that it may be possible to gradually taper the dosage of methotrexate once an adequate response to infliximab plus methotrexate combination is achieved. Nevertheless, studies have shown that the efficacy of TNFα-antagonist therapy is increased when these agents are administered in combination with methotrexate. Moreover, immunosuppression with concomitant methotrexate also helps attenuate the formation of antibodies to infliximab (section 4.1.3). There is some evidence that infliximab may be combined with leflunomide (reviewed previously); however, most of these small, nonblind studies are available as abstracts only. In the absence of data from large, randomised trials, methotrexate remains the preferred DMARD for use in combination with infliximab.
Limited data have shown that, despite similar mechanisms of action (TNFα antagonism), subtle differences among TNFα antagonists may result in their use as an alternative to each other. For example, infliximab can be effective in patients who have not responded to etanercept and vice versa. Convenience of use is another consideration in making a choice between these agents. Infliximab is administered intravenously in a clinic/hospital, whereas etanercept and adalimumab are given subcutaneously, thus potentially making it possible for the patient to self-administer a dose. On the other hand, infliximab is given every 8 weeks (after an initial regimen at 0, 2 and 6 weeks),[9,20] whereas etanercept (once weekly) and adalimumab (once every two weeks) have a higher frequency of administration.
Although the initial recommended dose of infliximab in rheumatoid arthritis is 3 mg/kg, the US prescribing information indicates that the dose may be increased up to 10 mg/kg or the frequency of administration could be increased. Indeed, results of the ATTRACT trial (section 3.2.1), supported by a large open-label study, have shown that an increase in dose may be associated with a higher response rate. In the START trial, less than one-third (30%) of patients required an increase in infliximab dose.
Rheumatoid arthritis is a substantial economic burden, with the annual total medical costs averaging approximately $US6000–$US8500 per patient in the US, despite the fact that the disease therapy has traditionally utilised relatively inexpensive options, such as methotrexate. With the recent advances in therapy and the availability of newer DMARDs, the widespread use of combination therapies and aggressive and early institution of the most effective strategies has meant that the cost of treatments is increased greatly. For example, the annual costs per ACR 20 responder for biological agents range between $US25 000 and $US65 000 per year in the US, which are much higher than those with conventional DMARDs (when only direct costs are considered). However, the medium- to long-term usage of the healthcare resources (including hospitalisation or orthopaedic surgery) may be reduced by the use of biological agents, which may also lower indirect costs (e.g. those associated with lost productivity due to disability).[70,110] Indeed, modelled cost-utility analyses indicate that infliximab plus methotrexate was generally associated with acceptable incremental cost-effectiveness ratios relative to methotrexate monotherapy in patients with refractory rheumatoid arthritis (section 5.2). The cost effectiveness of infliximab in early disease has not been evaluated; however, it can be anticipated that early initiation of therapy may further reduce the costs associated with the long-term disease outcomes.
7.2 Tolerability Considerations
Infliximab was generally well tolerated in patients with Crohn’s disease or rheumatoid arthritis with treatment durations of up to 2 years and follow-up periods of up to 3 years (section 4). In these trials, it was almost always administered in combination with one or several other agents (sections 3.1 and 3.2). In general, serious adverse events occurred with similar frequency in infliximab and placebo recipients. Importantly, there appears to be no need for regular laboratory monitoring of infliximab therapy for adverse effects. Infusion-related reactions are the most common adverse drug reactions, but are usually mild and transient and can be minimised with concomitant immunosuppressive therapy; such therapy also reduces antibody formation against infliximab, thereby attenuating the risk of loss of response to the drug (section 4.1.3). Nevertheless, because of the central role of TNFα in several pathological states, concerns regarding the safety of TNFα antagonists as a class remain.
As with other TNFα antagonists, patients must be screened before and during infliximab therapy for infections. In particular, recrudescence of latent tuberculosis has been seen in patients receiving TNFα antagonists (section 4.2). Of note, most of these patients were receiving concomitant therapy with immunosuppressive agents, which itself could predispose them to the risk of infections. Infliximab is contraindicated in patients with tuberculosis or other serious infections.[20,26] Malignancy is another problem postulated to be related to TNFα antagonism. However, although the incidence of lymphomas has been found to be increased in patients receiving TNFα antagonists, this needs to be confirmed; there was no increase in other malignancies with the use of infliximab or other TNFα antagonists.
Finally, the use of TNFα antagonists in two potential indications has provided some insight on the safety aspect of these agents. Heart failure and CNS demyelination disorders were both shown to have worsened when TNFα antagonists were used. Infliximab is contraindicated in patients with heart failure (in Europe [moderate-to-severe failure] and Japan) or demyelinating disease (Japan).
Although some of the above adverse events could be potentially fatal, a clear association between the use of TNFα antagonists and an increased incidence of mortality has not been established. On the contrary, a recent analysis of 63 811 patient-years of follow-up in rheumatoid arthritis patients receiving TNFα antagonists (available as an abstract) has reported a decreased risk of mortality (e.g. hazard ratio of 0.69 [95% CI 0.55, 0.87] with vs without infliximab-methotrexate combination). Data from the TREAT (Therapy Research Evaluation and Assessment Tool) registry in 5000 patients with Crohn’s disease (available as an abstract) have also suggested a reduced risk of mortality within the first 3 months of an infliximab infusion.
Infliximab is effective in the treatment of patients with moderately to severely active Crohn’s disease refractory to other treatment options or those with fistulising disease. In combination with methotrexate, infliximab reduced signs and symptoms and delayed disease progression in patients with active methotrexate-refractory rheumatoid arthritis as well as in those with early disease. Infliximab had disease-modifying activity and rapid onset of action and was generally well tolerated. Recrudescence of tuberculosis infection and worsening of heart failure and demyelinating disease are among some of the concerns with anti-TNFα therapy, requiring cautious use of these agents in high-risk patients.
Current data suggest that infliximab may be cost effective, especially when long-term clinical outcomes and burden of the diseases are taken into account. More robust, prospective pharmacoeconomic studies are required to better ascertain the cost effectiveness of infliximab.
Direct head-to-head comparative trials of infliximab with other biological agents are not yet available and would be helpful in determining with greater certainty the place of infliximab in the management of these diseases. Nonetheless, infliximab, like other biological agents, is a valuable treatment option in the treatment of patients with moderately to severely active Crohn’s disease (including fistulising disease) or rheumatoid arthritis (including early disease).
The use of trade names is for product identification purposes only and does not imply endorsement.