Multicentre randomized-controlled clinical trial of Ipocol, a new enteric-coated form of mesalazine, in comparison with Asacol in the treatment of ulcerative colitis

Authors


Dr A. Forbes, St Mark's Hospital, Watford Rd, Harrow, HA1 3UJ, UK.
E-mail: alastair.forbes@ic.ac.uk

Abstract

Background : 5-Aminosalicylates remain important in the treatment of ulcerative colitis, but it is uncertain if the various preparations currently available are equivalent given the different delivery systems that exist. Generic prescription of mesalazine (mesalamine) is therefore inappropriate. Ipocol has recently become available as an alternative to Asacol-MR.

Aim : To compare the two agents in a controlled trial using a non-inferiority design.

Methods : Eighty-eight ulcerative colitis patients with a mild to moderate clinical relapse were randomized to one of the two drugs at a daily dose of 2.4 g for 8 weeks. Safety was the key concern; the primary measured end-point was efficacy as judged from a colitis activity index.

Results : There were no unexpected adverse events of clinical consequence. The colitis score improved similarly in both patient groups (by 2.3 with Ipocol and by 1.5 with Asacol: not significant), and a similar proportion was in clinical remission at the end of the study (26.1% for Ipocol and 28.6% for Asacol: not significant). Systemic steroids were needed in 11.9% of the Asacol-treated patients compared with 6.5% with Ipocol (not significant).

Conclusion : It appears appropriate to conclude that, while not identical to Asacol-MR, Ipocol offers a safe and similarly effective alternative.

Introduction

Mesalazine and the other 5-aminosalicylates (5-ASA) remain important drugs in the treatment of ulcerative colitis (UC). The oral mesalazine formulations use various coatings or protective characteristics that release the drug under specific conditions targeting the distal gastrointestinal tract.1, 2 Asacol-MR (mesalazine 400 mg; Procter & Gamble, Egham, UK), which has been popular in many countries, releases mesalazine (at a pH of over 7) via its delivery system of Eudragit S coating. The usual sites of release are the terminal ileum and colon. Asacol's patent protection recently expired, and several so-called generic substitutes have consequently been developed. Normally, the introduction of generic medicines is to be welcomed, as there can be advantages in cost control for health care systems with no compromise on clinical efficacy; generic prescription is thus widespread. However, as we have previously indicated, identity of the constituent drug (in this case mesalazine) does not equate to automatic equivalence of clinical effect unless the delivery system is, in all respects, identical also.3

The development of mesalazine preparations with non-identical delivery systems can therefore be seen as a special case, as substitution will not inevitably create an identical effect, even though the principal active agent is the same. This can have positive and negative consequences: positive in that not only additional useful 5-ASA delivery profiles may emerge, but also negative ones if delivery is more proximal to a degree sufficient to reduce the amount of active drug available distally (or even such as to alter toxicity if greater proximal absorption leads to more exposure of a vulnerable kidney). Accordingly, we made a strong recommendation that generic substitution of mesalazine should not be supported without clinical data demonstrating similar efficacy and toxicity profiles for the new preparation.

Ipocol (initially Lagap Pharmaceuticals Ltd and now Sandoz Ltd, Hants, UK), like Asacol, contains 400 mg of mesalazine in a Eudragit S coating. However, the coating is of a different thickness (somewhat thinner in Ipocol), and there appear to be differences in the dissolution characteristics of Asacol and Ipocol in vitro. From unpublished studies (for which the data are held on file by Sandoz Ltd), it was found that at pH 1.0 and 6.5 there are no apparent differences in the behaviour of the two preparations, but at pH 7.5 mesalazine is released more rapidly from Ipocol.

We were therefore pleased to be approached by Lagap Pharmaceuticals for advice on a suitable clinical evaluation of the newer agent before they sought a clinical product licence for the UK. With an identical active agent and the same coating, differences of major clinical consequence between Asacol and Ipocol should not be expected, but we felt that a formal clinical trial was warranted on the basis of our previous published views and on the laboratory evidence alluded to above. A formal equivalence study was not however felt justifiable. Demonstration of equivalence requires that all points within the observed confidence interval for any difference observed correspond to a difference of no clinical significance. Typically, this leads to a need for sample sizes approximately fourfold greater than in an otherwise similar comparative study.4 The more modest proposal of a non-inferiority design (essentially functioning as a safety study) was therefore considered. A non-inferiority design attempts to provide a suitable compromise between the conventional comparative study (one treatment is or is not statistically different from the other) and the equivalence study (no statistical difference is discernable between the two treatments). In the non-inferiority design, equivalence or superiority of the test agent is accepted as a positive outcome.4 A smaller sample size is required than for equivalence and the new drug is shown either (a) to be inferior to the reference agent or (b) to be equivalent or superior. In its most basic analysis, it is simply a one-tailed test. This study is now presented. The objective was to compare oral mesalazine 800 mg three times daily as Ipocol 400 mg, with the same regime using Asacol 400 mg (reference agent), in patients with mildly to moderately active UC. It was hypothesized that the two formulations would not differ therapeutically or toxicologically.

Materials and Methods

Patients

Ninety patients with UC were enrolled. In each case, a prior diagnosis of UC based on standard diagnostic criteria including characteristic endoscopic and histological appearances was required.5 The extent of the disease was defined from its macroscopic relationship to the splenic flexure (distal if only below this, extensive if above also).

Patients were enrolled by a personal approach of the investigators in the gastrointestinal clinics of eight hospitals, which provided a range of specialist and more general colitis services. The inclusion criteria were an acute exacerbation of UC, defined as a deterioration in symptoms (increased stool frequency, with passage of blood and/or mucus) to the extent that the supervising clinician considered it suitable to amend the therapeutic regime. Patients needed to be aged over 18 years and otherwise in good health. Patients were ineligible if there had been exposure to systemic steroids in the 4 weeks prior to enrollment, or to any immunosuppressant or immunomodulatory drug in the previous 3 months. Prior topical therapy up to the date of recruitment was permitted, as was prior therapy with oral 5-ASA at a dose lower than the equivalent of mesalazine 2.4 g/day. All oral gastrointestinal therapies other than the trial agent were disallowed during the study period, but topical therapy was permitted on condition that the dose and frequency had been stable for 4 weeks before the study and was continued at the same level throughout the study. The use of steroids was permitted if the patient's condition deteriorated sufficiently to warrant such treatment, but the patient was then withdrawn from the trial and classified as a treatment failure. This applied both to oral and topical steroid preparations (except for those patients already using topical preparations as above). The usual exclusions in terms of other important medical conditions and expectations for compliance were observed.

Full written consent was obtained from all subjects. The study had the approval of the Northern and Yorkshire Multicentre Research Ethics Committee (Ref MREC/1/3/26), and the local committee relevant to each recruiting institution. At the time of initiation of the study, Lagap Pharmaceuticals Ltd were using the name Pentacol for what is now known as Ipocol; accordingly, the trial protocol and ethics documentation used this previous name for the study product.

Methods

Eligibility was confirmed from a clinical review of all relevant information as above, combined with sigmoidoscopy (rigid or flexible) and rectal biopsy to assess current activity. Full blood count, routine renal and hepatic biochemistry, and C-reactive protein levels were obtained. Additionally, patients were asked to complete the relevant questionnaires to permit calculation of the modified St Mark's Colitis Activity Score,6 and the EuroQoL quality-of-life score.7 The St Mark's score relies on a combination of clinical features from the history, the examination and from the endoscopic assessment of activity. The endoscopic scoring, is on a four-point scale, encompassing ‘normal’ through to the highest score for patients with frank ulceration and spontaneous bleeding. The histological assessment does not itself form part of the score but was separately documented. The EuroQol depends on patient self-ratings in domains including (for example) mobility, ability to lead a normal life, and degree of anxiety; it also includes an overall linear analogue scale.

Eligible patients were then randomly allocated to treatment at the randomization centre maintained by Lagap Pharmaceuticals Ltd. Computer-generated random numbers with stratification for disease extent (distal or extensive) were employed. Study drug was provided in an anonymous blister package with instructions to take two 400-mg tablets three times a day. The tablets themselves were not identical as they are somewhat different in shape. Patients were advised that they might find that they were prescribed a tablet shaped differently from those they had received before, but not that this was or was not Asacol or Ipocol. Clinical investigators took care neither to see nor to enquire of the nature of the tablets. Clinic visits were programmed at every 2 weeks for 4 weeks and then again at 8 weeks (a total of four visits). Those subsequent assessments included clinical review, and repeat blood testing on each visit and a second sigmoidoscopy and biopsy at week 8 only. The patient recorded stool frequency, presence of faecal blood or mucus, abdominal pain and any other symptoms on a standardized daily diary form throughout the study period. There was an option (at week 4) to reduce the number of study doses if clinical remission had occurred, in line with the product information for Asacol.

Efficacy was characterized by the modified St Mark's Colitis Activity Score, the macroscopic and microscopic appearance of the rectum, and by the investigator's overall global assessment. Clinical remission was defined from the latter. Safety parameters comprised the laboratory evaluations together with all adverse events reported. Evaluation was by intention-to-treat for efficacy variables but including all treated patients for safety criteria. Tablet counts were performed to provide a check on compliance; all tablet counting was performed by pharmacy departments to ensure blinding.

The intended total of 90 patients (45 patients per treatment arm) was derived from formal power calculations which indicated that this would provide the study with a power of 80% to detect a difference in efficacy of 30% (approximately 15% absolute difference). The investigators considered that differences in efficacy were likely to be of clinical meaningfulness only if they exceeded 15% as based on the measurements to be obtained. Differences of this magnitude were not expected – the essence of the study being one of safety and investigating for no therapeutic difference. All statistical testing therefore was targeted at the demonstration of non-inferiority, aiming (null hypothesis) to demonstrate that Ipocol was inferior to Asacol or that it was either equivalent or superior.

Results

Ninety patients were screened and randomized to be treated – 44 with Asacol and 46 with Ipocol. Two patients (both allocated to Asacol) were withdrawn from the trial after randomization, but before administration of the trial medication, as a result of withdrawal of consent. The safety population therefore included 88 patients (Table 1). During the course of the study, 11 patients withdrew from the Asacol group, and nine withdrew from Ipocol; thus 68 patients completed the trial as per protocol. The trial allocation presented a reasonably well-balanced randomization for the standard demographic and clinical parameters at study entry (Table 1).

Table 1.  Clinical data for the two groups at entry to the study period
 Patients allocated AsacolPatients allocated Ipocol
  1. Mean values with standard deviations are given other than in the case of platelet count and C-reactive protein (CRP), where there were obviously non-normal distributions, and for which medians and ranges are also supplied.

Number4246
Age (years; mean ± s.d.)44.8 ± 13.747.9 ± 15.3
Male sex (%)59.558.7
Weight (kg; mean ± s.d.)77.1 ± 15.177.2 ± 18.1
Extensive disease (%)3938
Haemoglobin (g/L; mean ± s.d.)136 ± 15.6137 ± 13.3
Haematocrit (mean ± s.d.)0.40 ± 0.0420.41 ± 0.035
Platelet count
 Mean ± s.d.336 ± 153279 ± 92
 Median (range)310 (171–865)269 (120–654)
Albumin (g/L; mean ± s.d.)41.5 ± 4.641.3 ± 3.6
CRP (mg/L)
 Mean ± s.d.16 ± 38.79 ± 7.4
 Median (range)6 (0.2–211)6 (0.3–39)
Sigmoidoscopy score
 of 1 (%)2934
 of 2 (%)5235
 of 3 (%)1926
St Mark's Score (mean ± s.d.)5.1 ± 2.325.4 ± 2.09

Thirteen patients in the Asacol group and 14 patients in the Ipocol group were on mesalazine at permitted levels/routes prior to the trial. Most other pre-study drugs were for unrelated conditions. Concomitant medications used without deviation from the study protocol were few, and the most frequent new oral agent was prednisolone taken because of inadequate efficacy of the trial medication – this was needed in 9.1% overall. There was a small excess need for oral prednisolone in the Asacol group (11.9% vs. 6.5%), which did not reach significance. By contrast, topical steroids, which were needed overall in 15.7% of individuals, were used somewhat less in the Asacol patients (11.0% vs. 17.4%; ns).

The only protocol violations were supplementary mesalazine taken by two patients (4.8%) in the Asacol group and one patient (2.2%) in the Ipocol group as a consequence of prescription by nontrial physicians. Compliance was judged to be >90% in the remaining patients at all stages in the trial, and was similar between the treatment groups.

The modified St Mark's Colitis Activity Score fell in both treatment groups. Although Ipocol yielded a slightly greater reduction between baseline and 8 weeks (−2.3 vs. −1.5 for Asacol), there were no significant differences between the treatment groups (Figure 1). The EuroQoL score also changed similarly in both treatment groups with improvements of 0.7 and 0.5, respectively (ns). There was no significant difference in the clinical remission rate between Asacol and Ipocol at 4 weeks (28.6% for Asacol vs. 26.1% for Ipocol; ns) or at 8 weeks (Figure 2).

Figure 1.

Serial estimations of the modified St Mark's Score in the two groups (lower scores indicate improvement).

Figure 2.

Proportion of the two groups in clinical remission through the study period.

The sigmoidoscopy score improved by 8 weeks in 54.8% of patients receiving Asacol compared with 50.0% of patients on Ipocol (ns). Nonetheless, most completing patients still had abnormal appearances at sigmoidoscopy at 8 weeks (73.8% vs. 73.9%); there were no clear differences between the two agents under evaluation. Corresponding improvements occurred in the histological assessments; grades were improved in 31% of Asacol patients and 30.4% of those on Ipocol (ns). Most patients still had clinically significant abnormalities in the biopsies at 8 weeks, and there were no differences between the two agents under evaluation (22.6% normal vs. 21.6% normal for Asacol and Ipocol, respectively).

Clinical laboratory evaluation was reassuring throughout the study period for both formulations, with no major changes seen in any of the major parameters, which remained in the normal range in all patients apart from those in whom these were abnormal at recruitment, and which improved. For no parameter and for no visit were significant differences observed, and there were no numerical trends.

Overall, employing the criteria of the International Conference on Harmonization, there were 233 adverse events (AEs) in 65 patients (73.9%).8 There was a similar percentage of patients with at least one AE in each treatment group – 31 patients (73.8%) had a total of 93 AEs in the Asacol group compared with 34 patients (73.9%), who had a total of 140 AEs, in the Ipocol group. The great majority of the adverse events were mild and ‘unrelated’ or ‘likely to be unrelated’ to the study drug (70.1%). There were however two patients (2.3%) with an AE which the International Conference classifies as severe.8 Both of these patients were on Asacol, and in each case the problem stemmed from worsening UC. The relevant clinicians considered the problems ‘unrelated’ and ‘unlikely to be related’ to therapy and therefore representative of failure of therapy rather than toxicity. Aminosalicylate therapy was continued in these cases; in the latter patient, there was full resolution, but the former underwent interval colectomy. Two other patients on Asacol in whom abdominal pain became prominent were withdrawn from the study, but no other patient was withdrawn on the basis of known or suspected drug toxicity.

Discussion

There were no major differences in observed toxicity between the two preparations of mesalazine. Although Ipocol yielded a slightly greater reduction in the St Mark's score at 8 weeks and a lesser need for systemic steroids, there were no significant differences between treatment groups by intention-to-treat (or indeed by treatment received). In line with this, the EuroQoL score also improved modestly during the study in both groups. The relatively poor response rates reflect the relatively low potency of mesalazine in active UC, and perhaps also that some of the patients were effectively mesalazine failures prior to entry (albeit with the prior drug administration necessarily at a lower dose).

This cannot be considered an equivalence study, but we believe we have satisfied the challenge for a new mesalazine product that we set ourselves in our previous review,3 and that the product licence now issued by the Medicines and Healthcare products Regulatory Agency (MHRA) for Ipocol in UC is appropriate. However, it is to be remembered that while major therapeutic differences have not been identified, and while safety seems satisfactory, the behaviour of the two preparations cannot be assumed to be identical at the level of an individual even if the net effect in a population is apparently the same. Our concern that a greater proximal availability could lead to overall loss of efficacy for the same total oral dose remains (if diminished for this particular mesalazine pairing) given the predominantly distal nature of UC. At a more hypothetical level, we cannot rule out a potentially greater danger for the vulnerable kidney if more of the drug is absorbed more proximally. Ipocol has such pharmaceutical and in vitro characteristics that would tend to predict more rather than fewer concerns in these areas in comparison with Asacol. It is therefore reassuring that the small numerical differences seen do not suggest any adverse trend. Only a substantially larger study could address these questions more authoritatively, and we do not believe that the resources required for such a study would now be appropriately deployed. Research in inflammatory bowel disease is at an exciting stage and there is a clear need for evaluation of several of the more novel agents in UC. We believe that such studies now have legitimate primacy over further comparisons between Asacol and Ipocol, even if other 5-ASA comparisons may still be needed.

We continue to recommend that the clinician prescribing mesalazine selects a particular formulation and that (assuming this to be effective and well tolerated) this specific agent is used thereafter without unplanned substitution of other agents. Financial considerations may very legitimately play a part in the initial deliberation.

In summary, there was no apparent difference between Asacol and Ipocol in treating mild exacerbations of UC in terms either of safety or efficacy. It appears appropriate to conclude, that, while not identical to Asacol-MR, Ipocol offers a safe and similarly effective alternative.

Acknowledgements

We gratefully acknowledge the contributions made to the study by: Dr J. Booth, Reading, Prof J. Brown, Gloucester, Dr C. Probert, Bristol, and Dr A. Thillainayagam, London, who each helped in the enrollment of patients. We also thank our various assistants and of course the patients themselves. Lagap Pharmaceuticals Ltd provided all the drugs used for the study, and arranged the blinded packaging and the telephone randomization service. The company also monitored the conduct of the trials and appropriate documentation in the various centres to comply with Good Clinical Practice, together with providing modest running expenses. This is acknowledged in the authorship of the paper, but the analysis and conclusions are those of the clinical authors and have not been modified by the company.

Ancillary