Review article: induction therapy for patients with active ulcerative colitis

Transcription

1 Alimentary Pharmacology & Therapeutics Review article: induction therapy for patients with active ulcerative colitis S. P. L. TRAVIS John Radcliffe Hospital and Linacre College, Oxford, UK Correspondence to: Dr S. P. L. Travis, John Radcliffe Hospital, Oxford OX3 9DU, UK. Publication data Accepted 28 June 2006 SUMMARY 5-aminosalicylic acid (mesalazine) is considered first-line therapy for active mild moderate, left-sided or extensive ulcerative colitis. With modern delayed-release formulations, conventional doses are very well tolerated, and there is accumulating evidence that increasing the dose, to >4 g/day, leads to higher response rates and earlier symptom relief in patients with ulcerative colitis. There is evidence that combining oral and rectal (enema) formulations of 5-aminosalicylic acid leads to faster and higher remission rates. Despite the fact that clinical trials in ulcerative colitis have often used different endpoints, making it difficult to make objective comparisons, the evidence now strongly suggests that it is appropriate to start patients with mild or moderately active ulcerative colitis on doses of 5- aminosalicylic acid of at least 4 g/day. For those with extensive disease, adding rectal 5-aminosalicylic acid improves remission rates; the same seems likely to be true for limited disease, but robust evidence is lacking. Rectal 5-aminosalicylic acid may in fact be sufficient on its own for patients with active proctitis. Patients with mild moderate ulcerative colitis who do not improve within 2 weeks of high-dose 5-aminosalicylic acid should have treatment augmented by oral steroids. Aliment Pharmacol Ther 24 (Suppl. 1), ª 2006 The Author

2 REVIEW: INDUCTION THERAPY FOR PATIENTS WITH UC 11 INTRODUCTION The British Society of Gastroenterology (BSG) guidelines recommend 5-aminosalicylic acid (5-ASA) as first-line therapy for active mild moderate, left-sided or extensive ulcerative colitis (UC), unless a prompt response is required that necessitates steroids. 1 Increasing the dose of sulfasalazine to increase its therapeutic effect, however, was found to be limited by side-effects. The advent in 1987 of mesalazine (5-ASA) with its superior side-effect profile allowed higher dosing, showing that increasing the dose could give further benefit. 2 Remission rates achieved in this early study were low, 2 possibly because stringent definitions of complete remission were used with stool frequency, rectal bleeding, flexible proctosigmoidoscopy findings and Physician s Global Assessment (PGA) all scoring zero. The overall response rate (complete and partial response) improved from 27% in patients taking 1.6 g/day to 74% in patients taking 4.8 g/day compared with an overall response rate of 18% in patients taking placebo. Similarly, the remission rate improved from 9% to 24% compared with 5%, respectively. 2 In a subsequent trial of 374 patients with mild moderate active UC, 2 and 4 g/day of 5-ASA were superior to placebo, 3 in terms of both remission and response. There appeared, however, to be a plateau for remission at 29% for both 2 and 4 g/day compared with placebo at 12%. Nevertheless, 4 g/day produced greater improvement in clinical symptoms, greater endoscopic improvement and greater histological improvement than 2 g/day (Figure 1). The study also showed that the efficacy of 5-ASA was independent of the extent of disease or recent steroid use. Adverse effects were independent of dose and, intriguingly, were less frequent in patients in the active-treatment groups than in those in the placebo group. This study demonstrated that dose escalation of 5-ASA without toxicity was possible. A Cochrane meta-analysis has confirmed that 5-ASA is twice as effective as placebo for inducing remission (Figure 2). 4 A systematic review has also examined all nine published placebo-controlled clinical trials of 5-ASA up to The remission rate in patients treated with 5-ASA was indeed twice that in patients given placebo approximately 20% vs. 10%, respectively, at 6 12 weeks and dose-dependent, but the authors identified only a modest improvement in remission as the dose of 5-ASA was increased. They therefore concluded that 5-ASA alone was not % of patients Remission Success Benefit ASA (g/day) Figure 1. Effect of dose of 5-aminosalicylic acid (5-ASA) on remission, success and perceived benefit in patients with active mild moderate ulcerative colitis. Reproduced from Hanauer et al. (1993), 3 with permission. particularly effective in inducing remission; the number needed to treat (NNT) to produce one remission was 10, compared with only two for corticosteroids. 5 The questions that remain are whether dose escalation was sufficient and whether the clinical heterogeneity of mixing mild and moderately active disease, let alone patients with different disease distributions, obscured the benefit. One of the complications of comparing clinical trials is that different study groups have used different definitions of disease severity. There are at least eight ways to define disease severity. 6 There is, as yet, no internationally agreed standard classification of severity of disease and, even more remarkably, no standard definition of remission. If mild and moderately active disease severities are going to be differentiated, then the two strata should mean that there are differences in management or outcome; otherwise clinicians need only discriminate between severe UC and less active disease. This is particularly pertinent now that there are several new trials of 5-ASA compounds and up to 30 new agents in Phase I III trials for inflammatory bowel disease (IBD). 7 Moderately active disease can also be defined by sigmoidoscopy score, with the presence of mucosal friability discriminating moderate

3 12 S. P. L. TRAVIS Figure 2. Results of a Cochrane database meta-analysis of the ability of 5-aminosalicylic acid (5-ASA) to induce remission in patients with mild moderate active ulcerative colitis. Reprinted from Sutherland and MacDonald (2003), 4 with permission. from mild inflammation. Because neither the clinical nor the endoscopic score have been validated in the era of flexible sigmoidoscopy and because interobserver variation is undefined, it can be seen that the definitions are less than precise. This is very likely to have an impact on outcome. EFFECTS OF INCREASING THE DOSE Asacol The most recent well-designed large studies of the effect of 5-ASA were the two Assessing the Safety and Clinical Efficacy of a New Dose (ASCEND) studies (ASCEND of 5-ASA). 8, 9 The objective of these two studies was to evaluate the safety and efficacy of 4.8 g/day of 5-ASA (using a novel high-dose formulation with 800 mg tablets) in comparison with 5-ASA 2.4 g/day (using the conventional 400 mg tablets) for active UC over a 6-week period. The ASCEND studies assessed disease severity with the PGA where 1 is mild and 2 is moderate. The Mayo score was used to evaluate response; this is a composite index measuring rectal bleeding, stool frequency, PGA and sigmoidoscopy, with mild scoring 2 4 and moderate The two studies were double-blind, randomized, multi-site, active control studies in patients with active UC. The ASCEND studies were initially designed to evaluate patients with mild moderate active UC. Once complete, ASCEND I showed that patients with moderately active disease received significant benefit from 4.8 g/day but those with mild disease did not. Thus, the ASCEND II protocol was amended during the study, prior to unblinding, to evaluate efficacy in moderate-uc patients. The primary endpoint was the percentage of patients classified as having an overall improvement ( treatment success ) at the end of 6 weeks. Complete remission was recorded as having been achieved if all clinical assessments and the PGA score were zero. A response was deemed to be an improvement from baseline in PGA and improvement from baseline of at least one clinical assessment, providing that there was no worsening in any clinical assessment. Clinical assessments included stool frequency, rectal bleeding, patient functional assessment and sigmoidoscopy. Overall improvement was defined as the combination of remission and response. The ASCEND I study, which assessed 301 patients, showed only a significant benefit for patients with moderately active disease; 57% of patients treated with 2.4 g/day achieved overall improvement compared with 72% on 4.8 g/day (P ¼ ). 8 This was independently confirmed in the ASCEND II study of identical design. 9 At 6 weeks, patients given the higher dose faired significantly better overall improvement was 59% for the 2.4 g/day dose vs. 72% for the 4.8 g/day dose (P ¼ 0.036; Figure 3). 9 There was, however, no clinically relevant increase in remission rates between 2.4 and 4.8 g/day at 6 weeks (from 18% to 20%).

4 REVIEW: INDUCTION THERAPY FOR PATIENTS WITH UC 13 % of patients * n = 130 n = g/day 4.8 g/day *P < 0.05 Table 1. Safety and tolerability data from ASCEND II [data from Hanauer et al. (2005). 9 ] 2.4 g/day [n ¼ 139; n (%)] 4.8 g/day [n ¼ 129; n (%)] Patients with AEs 49 (35) 57 (44) Patients with serious AEs 2 (1.4) 1 (0.8) Patients withdrawn due to AEs 4 (2.8) 4 (3.1) Figure 3. Overall improvement at week 6, Assessing the Safety and Clinical Efficacy of a New Dose (ASCEND) II study programme. Reproduced from Hanauer et al. (2005), 9 with permission. ASCEND, Assessing the Safety and Clinical Efficacy of a New Dose; AE, adverse effect. One of the problems with the disease activity score used (PGA) is that it includes the subjective variables of doctor and patient perception of benefit. These are of course important impressions, but they lack the objectivity of measurable variables. The objective variables in this index are stool frequency and rectal bleeding, both of which are symptoms of active disease that matter to patients. In ASCEND II, doubling the dose of 5-ASA to 4.8 g/day shortened the time to resolution of increased stool frequency, rectal bleeding and a combination of the two. In the population of patients with moderate disease, increasing the dose from 2.4 to 4.8 g/day led to trends towards reduction in the median time to normalization of stool frequency from 13 days to 10 days (P ¼ 0.288), and of the combined endpoint of stool frequency and rectal bleeding from 32 days to 21 days (P ¼ ). However, the reduction in the time to cessation of rectal bleeding was significant from 16 days to 9 days (P ¼ 0.035). Consequently, the results of ASCEND I and II provide good evidence that doubling the dose of 5-ASA from the conventional 2.4 g/day produces moderate, but worthwhile, benefits to patients with active UC. In ASCEND II, the absolute benefit was 13% at 6 weeks in patients taking 4.8 g/day (P ¼ 0.036), while the median time to cessation of rectal bleeding was nearly halved (from 16 days for the lower dose to 9 days for the higher dose). This was achieved without any clinically relevant increase in adverse effects (Table 1), despite the fact that doubling the dose also doubled the serum concentration of 5-ASA and N-acetyl-5- ASA. Furthermore, the tablet burden will decrease once 800 mg Asacol (Egham, Surrey, UK) tablets become available, i.e mg tablets per day compared with mg tablets per day. SPD 476 SPD 476 is a multimatrix formulation of 5-ASA that includes a ph-sensitive coat around a matrix of hydrophilic polymer containing a portion of 5-ASA, together with lipophilic excipients that encapsulate additional 5-ASA. The lipophilic particles are reported to have the ability to stick to membranes, which is intended to create high concentrations of released 5-ASA near the colonic epithelium. The formulation has been evaluated in patients with mild moderate UC and is still under investigation and subject to approval. 10 Patients with mild moderate UC (n ¼ 280) were randomized to receive either SPD g twice daily or SPD g once daily or placebo for 8 weeks. The primary endpoint was remission at the end of the 8 weeks. Remission was defined as a UC disease activity index (DAI) score 1, with a score of zero for stool frequency and rectal bleeding, together with a reduction of at least one point in the sigmoidoscopy score. 10 Once and twice daily dosing produced similar results. Remission rates were 29% and 34%, respectively (4.8 g/day once daily vs. 2.4 g/day twice daily), compared with 13% on placebo (P < 0.01; Figure 4). These are within the range expected in trials of oral 5-ASA, although higher than the mean. 2, 3, 5 Clinical improvement rates of 60% (4.8 g once daily) and 56% (2.4 g twice daily), compared with 26% on placebo, were comparable with those observed in other highdose 5-ASA trials. The most interesting aspect is the efficacy of the once daily dosing. While this is unli-

5 14 S. P. L. TRAVIS % of patients Placebo SPD g bd kely to have a major impact on the treatment of active disease, it is likely to find favour for maintenance if subsequent trials confirm a benefit in remission. In general, the once daily regimen was well tolerated, although two cases of pancreatitis occurred in the 280 patients; both cases were receiving SPD 476 rather than placebo. Any differences in efficacy or adverse events in this trial (compared with other 5-ASA trials) may reflect differences in methodology. The endpoint of remission was tightly defined, but there was no independent histology. The impact of any differences would be expected to disappear in the randomization process, but the trial population included patients from centres from a range of countries not widely used in other UC trials, and for which no historical remission rates with other formulations of 5-ASA are available. The population in the study also included a higher proportion with mild disease and 12% had no rectal bleeding at entry. Nevertheless, the study clearly shows that once daily dosing with 5-ASA is effective, hence there is a need to re-examine the requirement for multiple dosing regimens. Combined oral and rectal 5-ASA SPD g od Remission Response Figure 4. Comparison of multimatrix 5-aminosalicylic acid (5-ASA) 2.4 g twice daily with 4.8 g once daily (bd) in patients with mild moderate ulcerative colitis (od). Data from Lichtenstein et al. (2005). 10 High doses of the slow-release rectal formulation of 5-ASA, Pentasa (Lucerne, Switzerland), have also been assessed in a clinical trial in patients with extensive, mild moderate active UC. 11 Patients (n ¼ 127) were randomized to receive either Pentasa 4 g/day (twice daily dosing) oral mesalazine for 8 weeks plus 1 g/day Pentasa enema, or 4 g/day plus a placebo enema. At 4 weeks, 89% of the patients given both the oral dose and enema had improved, compared with 62% in the placebo enema group (P ¼ ). The remission rate was also greater in the active-enema group 64% compared with 43% in the placebo-enema group (P ¼ 0.03) at 8 weeks. These rates are extremely high and among the highest for any trial of mesalazine. Remission was defined as a DAI <2, so traces of blood or double the normal stool frequency were technically acceptable as remission if the sigmoidoscopy was normal. The main value of this study is that other trials involving enemas have traditionally excluded patients with extensive disease. It clearly demonstrates the added value of rectal 5-ASA when treating patients with extensive UC, and this applies to distal disease as well. 12 IMPACT OF THE DEFINITION OF REMISSION In UC, there are at least three definitions of remission, which may be termed clinical, regulatory or complete remission. Clinical remission is what is used in practice, meaning cessation of rectal bleeding and normal stool frequency. This is not the same as regulatory remission (the one currently, but not exclusively, favoured by the FDA), which means cessation of rectal bleeding and a sigmoidoscopy score of 0 or 1 (normal appearance, or erythema only). This in turn is not the same as complete remission, where there is normal stool frequency, no rectal bleeding and normal or quiescent appearances of the mucosa at sigmoidoscopy. The potential impact of these three definitions is considerable, but many trials simply use an arbitrary threshold (0, 1 or 2) of one of the disease activity indices, or <150 in the complex Seo index, to define the remission endpoint. 6 Obscured in these definitions can be symptoms (such as bleeding or increased stool frequency) that clinicians and their patients would not recognize as remission (Table 2). In a pooled analysis of the ASCEND studies, which included a total of 687 patients with mild moderate active UC, the remission rate can vary from 22% [complete remission (score ¼ 0) of stool frequency, rectal bleeding, sigmoidoscopy findings, patient functional assessment and PGA] to 28% (DAI 1 with no bleeding and normal frequency and at least a 1-point

6 REVIEW: INDUCTION THERAPY FOR PATIENTS WITH UC 15 Table 2. Outcome criteria for remission Clinical remission decrease in sigmoidoscopy score), and 50% (DAI 2, with no individual subscore >1). 13 When the interobserver variation in sigmoidoscopy score is factored in (and Baron et al. originally based their score on >60% agreement 14 ) the rates of remission defined in trials are open to considerable differences in interpretation. This is a crucial issue for regulatory authorities, since the currently favoured definition depends substantially on sigmoidoscopy. In the future, trials should have independent evaluation of sigmoidoscopy using video imaging, together with a histological record of disease activity. This is because a normal, pre-treatment rectal biopsy excludes a diagnosis of active UC, even though there is no clear correlation between clinical, endoscopic and histological disease activity. This means that any patient with a normal, pre-treatment biopsy should be excluded from the trial and many patients with mild disease might well have symptoms provoked by an irritable bowel as much as by active colitis. RAPIDITY OF RESPONSE Regulatory remission Rectal bleeding score Stool frequency score 0 0 Sigmoidoscopy score 0/1 0/1 Complete remission Speed of response is an often-neglected aspect of treatment, despite its importance to patients. The number of days to relief of objective symptoms (increased stool frequency, rectal bleeding) is a useful measure that is easier to interpret than reductions in scores of disease activity indices. These data should be provided in reports on clinical trials and include the breakdown of the disease activity indices with numbers of actual patients. Otherwise it can be impossible to determine whether an improvement in the index has been driven by the subjective PGA, the patient s functional assessment or an objective measure of symptoms. In the ASCEND II study the median time to cessation of rectal bleeding was just 9 days in those receiving 4.8 g/day. 9 In a study of combined oral and enema treatment with Pentasa in patients with extensive mild moderate active UC, which had a much higher apparent remission rate, the time to cessation of rectal bleeding in 50% of those with frank bleeding at entry was 21 days in the combined oral and enema group. 11 In addition, among patients with any rectal bleeding at baseline, bleeding had ceased within 56 days of treatment in 73% of mesalazine enema-treated patients compared with 38% of placebo enema-treated patients. 11 Analyses of stool frequency data were not reported in this study. An important message for clinical practice is that there is no need to wait the 6 8 weeks duration of clinical trials to determine the response to high-dose 5-ASA induction therapy. Patients who do not improve within 2 weeks should be considered for decisive treatment with steroids. CONCLUSIONS High-dose (>4 g) 5-ASA is appropriate as initial treatment for patients with mild or moderately active UC. Adding rectal therapy with 5-ASA (i.e. as an enema) has added benefits compared with oral 5-ASA alone even in those patients with extensive disease, although rectal 5-ASA alone may be sufficient for some patients with active proctitis. Patients with mild moderate UC who do not improve within 2 weeks of high-dose 5- ASA should have treatment augmented by oral steroids. REFERENCES 1 Carter MJ, Lobo AJ, Travis SP. Guidelines for the management of inflammatory bowel disease in adults. Gut 2004; 53 (Suppl. 5): V Schroeder KW, Tremaine WJ, Ilstrup DM. Coated oral 5-aminosalicylic acid therapy for mildly to moderately active ulcerative colitis. A randomized study. N Engl J Med 1987; 317: Hanauer S, Schwartz J, Robinson M, et al. Mesalamine capsules for treatment of active ulcerative colitis: results of a controlled trial. Pentasa Study Group. Am J Gastroenterol 1993; 88: Sutherland L, MacDonald JK. Oral 5-aminosalicylic acid for induction of remission in ulcerative colitis. Cochrane Database Syst Rev 2003; 3: CD Bebb JR, Scott BB. How effective are the usual treatments for ulcerative colitis? Aliment Pharmacol Ther 2004; 20: Irvine EJ. Assessing outcomes in clinical trials. In: Satsangi J, Sutherland L, eds. Inflammatory Bowel Diseases. London: Churchill Livingstone, 2003:

7 16 S. P. L. TRAVIS 7 Travis SPL, Hanauer SB, Richter J. Advances in therapeutic approaches to ulcerative colitis and Crohn s disease. Curr Gastroenterol Rep 2005; 7: Hanauer SB, Sandborn WJ, Archambault A, Dallaire C, Yacyshyn B, Regalli G. Efficacy and safety of Asacol 4.8 g/day (800 mg tablet) compared with 2.4 g/ day (400 mg tablet) in treating moderately active ulcerative colitis. Can J Gastroenterol 2006; 20 (Suppl. A): A Hanauer SB, Sandborn WJ, Kornbluth A, et al. Delayed-release oral mesalamine at 4.8 g/day (800 mg tablet) for the treatment of moderately active ulcerative colitis: the ASCEND II trial. Am J Gastroenterol 2005; 100: Lichtenstein G, Kamm M, Sandborn W, Boddu P, Gubergrits N. SPD476, a novel once-daily, high dose mesalazine formulation is well tolerated and effective for the induction of remission of mildto moderate ulcerative colitis: a Phase III study [abstract]. Gut 2005; 54 (Suppl. VII): A Marteau P, Probert CS, Lindgren S, et al. Combined oral and enema treatment with Pentasa (mesalazine) is superior to oral therapy alone in patients with extensive mild/moderate active ulcerative colitis: a randomised, double blind, placebo controlled study. Gut 2005; 54: Safdi M, DeMicco M, Sninsky C, et al. A double-blind comparison of oral versus rectal mesalamine versus combination therapy in the treatment of distal ulcerative colitis. Am J Gastroenterol 1997; 92: Katz S, Kane S, Higgins P, Yacyshyn B, Eusebio R. Different definitions of remission for ulcerative colitis result in large variations of clinical outcome scores [DDW abstract]. Gastroenterology 2006; 130 (suppl. 2): A (Abstract). 14 Baron JH, Connell AM, Lennard-Jones J. Variation between observers in describing mucosal appearances in proctocolitis. BMJ 1964; 5375: