|Home | About | Journals | Submit | Contact Us | Français|
Crohn’s disease (CD) and ulcerative colitis (UC) are chronic inflammatory bowel diseases which can be difficult to control with conventional therapies. A greater understanding of their pathophysiology has led to new therapies that target specific molecules of the inflammatory cascade. Three anti-tumor necrosis factor (TNF) monoclonal antibodies have been developed. Infliximab and adalimumab can induce clinical response and sustained remission in CD. Infliximab is also effective in UC. Certolizumab pegol gives good short-term results but long-term efficacy has yet to be determined in other clinical trials. Therapies that target leucocyte trafficking (anti-integrins) have also been developed and are associated with good clinical response in CD. Natalizumab (anti-α4 integrin antibody) is associated with important side effects and is not used anymore in gastroenterology in Europe but is still used in the USA. Vedolizumab (MLN0002), an anti-α4β7 integrin antibody, has a good efficacy and safety profile. Monoclonal antibodies targeting other cytokines are also under development. For example, ustekinumab (CNTO 1275) inhibits interleukins 12 and 23. It is associated with a good clinical response in CD.
Inflammatory bowel diseases (IBD) are chronic relapsing diseases. Crohn’s disease (CD) is characterized by a transmural inflammation due to an imbalance between pro- and anti-inflammatory processes [Reimund et al. 1996; Reinecker et al. 1993; Stevens et al. 1992]. Only a small number of affected patients have a less severe disease course with a low prevalence of complications and hospitalizations [Beaugerie et al. 2006; Wolters et al. 2006; Munkholm et al. 1995]. The majority of patients develop complications including strictures, penetration of the bowel wall with obstruction, fistulas, and abscess [Cosnes et al. 2002; Louis et al. 2002]. Ulcerative colitis (UC) is only a mucosal disease – stenosis and fistula are uncommon. One quarter of such patients develop a chronic active or a rapid fulminating disease course, and 20% require surgery with ileo-anal pouch or stoma [Hoie et al. 2007]. Moreover, UC and colonic CD are associated with an increased risk of colon cancer linked to disease extent and chronic uncontrolled inflammation. Treatment goals include steroid-free remission, mucosal healing and decreases in hospitalization and surgery; as well as sustained clinical remission and sustained mucosal healing with a reduction of colorectal dysplasia and cancer. Steroids are effective short term but fail long term in most patients and are associated with numerous side effects and low rates of mucosal healing [D’Haens et al. 2008; Faubion et al. 2002]. Mesalazine is not effective in CD but is effective in maintenance of remission in mild UC [Kornbluth and Sachar, 2004]. Immunosuppressive drugs (azathioprine, 6-mercaptopurine, methotrexate) are associated with 40% sustained remission in one year but do not achieve high rates of endoscopic healing and do not modify the natural history of the disease course [Cosnes et al. 2005; Lémann et al. 2005; Bouhnik et al. 1996]. Therefore, chronic active IBD with incomplete mucosal healing under conventional treatment requires the use of biologic treatments. The ideal biologic agent for treating IBD should target a specific event of the inflammatory cascade, induce and maintain a sustained remission, be well tolerated and induce no immunogenicity. A better knowledge of the different inflammatory pathways in IBD has led to the development of new targeted therapies with actions that include inhibition of leukocyte trafficking, inhibition of T-cell activation or T-cell polarization, inhibition of pro-inflammatory cytokines. Moreover, thanks to impressive advances in protein engineering, a technique known as humanization was developed which led to the elimination of immunogenic nonhuman peptide sequences from anti-human antibody, and to the development of humanized (95%) and finally fully humanized (100%) antibodies [Hwang and Foote, 2005].
Development of anti-tumor necrosis factor (TNF) agents dramatically changed the course of IBD. Theses agents induce and maintain remission in patients with moderate-to-severe IBD. Infliximab, adalimumab and certolizumab pegol are able to block both soluble and membrane TNF but also fix complement. Only infliximab [Van den Brande et al. 2003; ten Hove et al. 2002] and adalimumab [Shen et al. 2006; Shen et al. 2005] are able to induce apoptosis of T-cells and monocytes in a caspase-dependent manner. Etanercept is a p75 TNF receptor construct able to bind soluble TNF but failing to induce lysis of cells expressing TNF [Sandborn et al. 2001]. Its efficacy has been proven in rheumatoid arthritis but not in IBD, suggesting that molecular mechanisms underlying the effects of anti-TNF are still obscure.
Infliximab is a chimeric monoclonal anti-TNF IgG1 antibody with proven efficacy in induction and maintenance of remission in patients with refractory luminal and fistulizing CD [Hanauer et al. 2002; Present et al. 1999; Targan et al. 1997]. It also causes rapid and profound endoscopic healing. It has been in use since the end of the 1990s. In the ACCENT 1 trial [Hanauer et al. 2002], infliximab induced a clinical response at 5mg/kg administered at weeks 0, 2 and 6. At week 54, with a scheduled treatment every 8 weeks, a remission rate of 29% was observed compared with 5% in the placebo group. Moreover, mucosal healing was obtained in 44% of patients compared with 18% in the placebo group. The ACCENT 2 study proved that infliximab (administered at weeks 0, 2 and 6) improved fistulizing disease in 55% and 38% of patients treated respectively with 10 and 5mg/kg compared to the placebo group (13% of response) [Sands et al. 2004]. At week 54, a sustained response was observed in 69% and 46% of patients treated respectively with 10 and 5mg/kg infliximab compared to the placebo group (23%).
Because of its residual murine fragment, some patients developed anti-drug antibodies which could increase the risk of acute and delayed hypersensitivity reactions and induce a secondary loss of response [Hanauer et al. 2004; Baert et al. 2003]. Thirty to 60% of patients receiving episodic treatment developed anti-drug antibodies compared to 7 to 10% in patients receiving a scheduled treatment [Hanauer et al. 2004]. This risk is lower if patients are treated concomitantly with immunosuppressive drugs. ACCENT 1 and 2 trials showed a good tolerability to the drug with a low rate of side effects. Similar rates of infections in placebo and infliximab groups were observed. Development of active tuberculosis in patients with latent tuberculosis was observed [Keane et al. 2001], so that the presence of latent tuberculosis has to be evaluated before treatment [Rutgeerts et al. 2004]. The TREAT trial showed similar mortality rates in infliximab and placebo groups [Lichtenstein et al. 2006]. In this trial, the risk of infections was associated with corticosteroid treatment.
The efficacy of infliximab in inducing and maintaining remission in 110 children with recently diagnosed CD has been demonstrated in the REACH study [Hyams et al. 2007] with 88.4% response and 58.9% remission at week 10, and 63.5% response and 55.8% remission after one year of treatment.
Infliximab also showed its efficacy in UC. ACT1 and ACT2 trials studied the efficacy of infliximab at a dose of 5 or 10mg/kg administered at weeks 0, 2 and 6 in moderate-to-severe UC compared with placebo group during 54 weeks (ACT1) or 30 weeks (ACT2) [Rutgeerts et al. 2005]. A clinical response was observed at week 8 in 61% and 69% of patients treated respectively with 5 and 10mg/kg versus 29% and 37% in the placebo group. Clinical response to infliximab in patients with corticosteroid resistance in UC was also assessed and the rate of colectomy was 67% in the placebo group to 27% in the treated group [Järnerot et al. 2005].
Adalimumab is a fully humanized anti-TNF monoclonal IgG1 antibody that may be given subcutaneously. Its immunogenicity is very low (only 3% of anti-drug antibodies were observed at week 2 compared with 14% with infliximab 5mg/kg at week 56). The CLASSIC 1 trial [Hanauer et al. 2002] was the first multicenter, randomized double-blind controlled trial studying the response to adalimumab by 299 moderate-to-severe CD patients naïve to anti-TNF treatment [Hanauer et al. 2002]. At week 4, remission was observed in 36% of patients receiving an induction dose of 160mg and 80mg at week 0 and 2 respectively compared with 24% in patients receiving an induction dose of 80 and 40mg, and 12% in the placebo group. Fifty-five patients from the CLASSIC 1 study were enrolled in the CLASSIC2 trial. The CLASSIC 2 trial showed remission maintained at week 54 in 74%, 83% and 44% of the patients treated respectively with adalimumab 40mg every 2 weeks, every week and placebo. The CHARM trial evaluated the clinical response to adalimumab in moderate-to-severe CD having received or not received other anti-TNFs previously [Colombel et al. 2007]. The remission rate was higher with adalimumab every week and every other week compared to the placebo group at week 26 (respectively 47%, 40%, 17%) and at week 56 (respectively 41%, 36%, 12%). Interestingly, the response rate was higher in patients who had never received other anti-TNFs before.
There is no prospective randomized placebo-controlled trial studying adalimumab in fistulizing CD but the presence of a fistula was not an exclusion criteria in the CHARM study and post hoc data have been published. In this trial, a complete closing of fistula was observed in 36% of patients treated with adalimumab every other week, 46% in patients treated every week, and 14% in the placebo group [Rutgeerts et al. 2005].
Interestingly, the GAIN study [Sandborn et al. 2007b] showed remission in 301 patients who were secondary nonresponders or intolerant to infliximab at week 4 in 21% of patients treated with adalimumab, at an induction dose of 160/80mg and then with 40mg every other week, compared with only 7% in the placebo group. A decrease in C-reactive protein (CRP) levels was also observed – elevated CRP may be considered prognostic of good response to adalimumab therapy. The safety of adalimumab was comparable to that of infliximab.
Preliminary data from open-label trials showed efficacy of adalimumab in mild-to-moderate UC with loss of response or poor tolerability to infliximab [Trinder and Lawrance, 2009, Peyrin-Biroulet et al. 2007]. These data should be confirmed by a large ongoing prospective trial.
Certolizumab pegol (CDP-870) is a PEGylated Fab fragment of a humanized anti-TNF antibody. It has shown efficacy in patients with refractory CD. The addition of two polyethylene glycols to the antibody fragment enhances its plasma half-life and permits its subcutaneous administration every 4 weeks. The preliminary placebo-controlled phase II study by Schreiber and colleagues studied certolizumab pegol at different doses (100, 200, 400mg) at weeks 1, 4, 8 in patients with moderate-to-severe CD [Schreiber et al. 2005]. All doses produced clinical benefit over placebo, mainly at week 2. At week 10, the optimal dose was 400mg. The PRECISE 1 [Sandborn et al. 2007a] and PRECISE 2 [Schreiber et al. 2007] trials used an induction dose of 400mg at weeks 0, 2 and 4 and then a scheduled treatment every 4 weeks. The trial showed response by two-thirds at week 6 as compared with placebo but no significant improvement in remission at weeks 6 and 26.
The number of inflammatory cells in the bowel wall depends on cell proliferation, apoptosis and also on leukocyte recruitment. Adhesion molecules, mainly integrins (α4β7, α4β1), located at the surface of endothelial cells, play a crucial role in the migration of leukocytes from blood vessels to intestinal tissue in IBD.
Natalizumab is a humanized IgG4 monoclonal antibody directed against α4-integrin inhibiting the adhesion and migration of leukocytes into inflamed tissues [Sandborn and yednock, 2003]. Different pilot studies have suggested that natalizumab may be effective in CD without major adverse effects [Ghosh et al. 2003; Gordon et al. 2001]. The ENACT1 trial included 905 patients receiving either placebo or 300mg natalizumab at weeks 0, 4 and 8. Similar response rates were observed at week 10 in treated and placebo groups (56% and 49% of response and 37% and 30% of remission, respectively), which suggested no benefit of natalizumab as an induction treatment in moderate-to-severe CD. The patients who responded to the induction treatment were enrolled in the ENACT2 study (354 patients treated with natalizumab and 74 patients treated with placebo). Continuing natalizumab resulted in a higher rate of sustained response (61% versus 28%) and remission (44% versus 26%) at week 36. Although data from the clinical trial indicate that natalizumab seems to be safe, it has been associated with progressive multifocal leukoencephalopathy (PML). PML is a fatal opportunistic infectious, demyelinating brain disorder induced by the JC virus [Van Assche et al. 2005]. This complication has also been described in six patients treated with natalizumab for multiple sclerosis. A follow up of 3100 patients who received natalizumab in different clinical trials showed no additional cases. A pilot study of natalizumab in UC showed no major adverse effects [Gordon et al. 2002]. There are now seven cases of PML associated with the use of natalizumab. The incidence of PML is estimated at one per thousand [Yoursy et al. 2006]. However, the medication is now FDA approved for patients who have failed anti-TNF therapy.
Because of the major adverse effects of natalizumab, represented by the JC-induced PML, new selective adhesion molecule-inhibiting agents are in development. Vedolizumab (MLN0002, MLN-02) is a recombinant humanized IgG1 monoclonal antibody to the α4β7 integrin. Two randomized control trials evaluating the efficacy and safety of vedolizumab in 181 active UC patients [Feagan et al. 2005] and 185 CD patients with moderate-to-severe CD showed no serious adverse effects. Clinical remission rates in UC at week 6 were 33, 32 and 14% in the groups receiving 0.5mg/kg vedolizumab, 2mg/kg vedolizumab and placebo, respectively. For the minority of patients who developed vedolizumab antibody titers greater than 1/250, no benefit of treatment was observed because of incomplete saturation of the α4β7 receptor on circulating lymphocytes. Data from a phase II study using vedolizumab [Feagan et al. 2008] were recently published. The primary endpoint was clinical response on day 57 and the secondary endpoints were the number of patients with clinical remission and with an enhanced clinical response. Clinical response rates at day 57 were 53%, 49% and 41% in the 2.0mg/kg vedolizumab, 0.5mg/kg vedolizumab and placebo groups, whereas clinical remission rates at day 57 were 37%, 30% and 21%, respectively. No major adverse effects were reported except one hypersensitivity reaction.
Interleukin (IL)-12 is a key cytokine that drives the inflammatory responses in CD. IL-12 is found at high concentrations in the bowel wall of CD patients, is mainly produced by antigen-presenting cells and takes a crucial role in the Th1 orientation of the lymphocyte system [Berrebi et al. 1998; Gately et al. 1998; Monteleone et al. 1997]. Mice with trinitrobenzene sulfonate-induced colitis have a Th1-mediated gut inflammation characterized by greatly increased production of IL-12, interferon (IFN) gamma and TNF alpha. Neutralization of endogenous IL-12 in CD mucosal cell cultures resulted in significant decreases in the number of IFN gamma producing cells [Parronchi et al. 1997]. Administration of recombinant anti-IL-12 in mice can resolve the induced colitis or prevent its development if given at the time of induction [Davidson et al. 1998; Neurath et al. 1995]. IL-12 is a heterodimeric cytokine constituted by two linked subunits, p35 and p40 [Mattner et al. 1993]. IL-12 and IL-23 share a common p40 subunit and, therefore, IL-23 biological activity is fully inhibited by the recombinant anti-IL-12p40. The role of IL-23 in T-cell response and its increased expression in CD has been described more recently [Schmidt et al. 2005]. The IL-23-driven inflammation appears to be mediated by the production of IL-17 and IL-6. Moreover, administration of IL-23 can increase the development of colitis lymphocyte-deficient recombinant mice [Yen et al. 2006]. Two fully humanized IgG1 monoclonal antibodies targeting the IL-12/IL-23 p40 subunit have been developed: ABT-874 (Abbott) and CNTO 1275 (Ustekinumab, Centocor). A randomized, controlled study of 79 moderate-to-severe CD patients treated with ABT-874 (1 or 3mg/kg) was performed. This study proved the efficacy of anti-IL-12/IL-23 p40 in the induction of response at week 7 in the group receiving the higher dose with 75% of response compared to 25% in the placebo group [Mannon et al. 2004]. A high proportion of patients receiving the drug developed injection-site reactions. Only three out of 79 patients developed anti-drug antibodies. More recently, a double-blind crossover trial studied the efficacy of ustekinumab against the subunit p40 of IL-12, IL-23 [Sandborn et al. 2008]. Clinical response rates for the combined groups given ustekinumab and placebo were 53% and 30%, respectively at weeks 4 and 6, and 49% and 40%, respectively at week 8. In a subgroup of 49 patients who were previously given infliximab (neither primary nor secondary nonresponders), clinical response to ustekinumab was significantly greater than in the group given placebo (p<0.05) through week 8.
The activation of the CD3 complex on T-cells leads to a severe cytokine-release caused by T-cell activation. A humanized anti-CD3 monoclonal antibody binding the Cd3 chain of the T-cell receptor expressed on activated T-cells has been tested in UC. An open-label phase I trial studied 32 severe steroid-refractory patients with UC who received visilizumab at a dose of 10 or 15μg/kg, administered intravenously on 2 consecutive days. The drug demonstrated an acceptable safety profile at the 10μg/kg dose level, a clinical benefit (66% remission and 87% clinical response) and a rapid decrease in circulating CD4(+) T-cell counts [Plevy et al. 2007]. Transient elevation of blood EBV-DNA (EBV, Epstein -Barr Virus) was also observed. Development of this compound, however, has been halted.
IL-6 is released mainly by antigen-presenting cells and produces its immunological effects mainly by binding to its soluble receptor. This cytokine plays a critical role in the apoptosis resistance of T-cells in CD by enhancing the expression of anti-apoptotic genes like BclXL and by inhibiting the expression of pro-apoptotic genes like Bax. IL-6 acts in synergy with IL-12 and IL-23 leading respectively to a Th1 and to a Th17 orientation of the lymphocyte system [Yen et al. 2006; Schmidt et al. 2005]. In animal models, administration of recombinant anti-IL-6 leads to a regression of colitis by inhibiting the release of pro-inflammatory cytokines, decreasing leukocyte migration and increasing the apoptosis of T-cells. Tocilizumab (namely MRA), a humanized anti-IL-6 receptor monoclonal antibody is under development for the treatment of inflammatory autoimmune diseases such as rheumatoid arthritis [Ding and Jones, 2006]. Tocilizumab showed efficacy at week 12 both in induction of response and remission in 36 patients with moderate-to-severe CD in a phase II study. The drug was administrated intravenously at a dose of 8mg/kg every 2 weeks and every 4 weeks. The response was higher in the group treated every 2 weeks [Ito et al. 2004]. No major adverse effects were observed.
IL-10 is an anti-inflammatory cytokine whose production is reduced in CD. IL-10 suppresses inflammation by various mechanisms including decreasing secretion of IL-2, IL-1, IL-8 and TNF alpha [Moore et al. 2001]. IL-10-deficient mice develop transmural inflammation of the gut which can be prevented by administration of recombinant IL-10 [Kühn et al. 1993]. However, the clinical efficacy of recombinant IL-10 for the treatment of CD has been disappointing. High doses of recombinant IL-10 caused a reduction in TNF alpha production but also an upregulation of IFN-gamma production by peripheral lymphocytes. No clinical efficacy was observed and many side effects occurred [Fedorak et al. 2000; Schreiber et al. 2000].
Problems caused by epithelial permeability have been implicated in the physiopathology of CD. During intestinal inflammation, significant damage to the bowel wall and the epithelial barrier are observed with increased intestinal permeability. Epidermal growth factor (EGF) has been involved in the preservation of mucosal integrity and the regeneration of damaged mucosa [Beck and Podolsky, 1999]. Recombinant EGF has been tested in 24 patients with moderate-to-severe UC in a randomized double-blind placebo-controlled trial [Sinha et al. 2003] – it achieved disease remission in 83% of patients versus 8% in the placebo group.
Granulocyte-macrophage colony stimulating factor (GM-CSF) is able to stimulate the innate immune system and is used for the treatment of disorders due to neutrophil dysfunction. A phase II study in 124 patients with CD evaluated the efficacy of sargramostin (GM-CSF) and showed efficacy for the secondary endpoint of remission. The precise mechanism of GM-CSF’s action in CD is still unclear [Korzenik et al. 2005].
A better knowledge of the physiopathology of CD and UC has led to the development of specific treatments against specific steps of the inflammatory cascade taking place in IBD. Anti-TNFs were the first biologic agents used in CD, and then in UC. New therapies have been developed including drugs inhibiting other cytokines, drugs against leukocyte migration, and drugs promoting leucocyte apoptosis. These recent advances in IBD treatment give us new opportunities to treat IBD refractory to conventional treatment or resistant to biologic agents by switching from one biologic agent to another.