Clarithromycin in Active Crohn's Disease

WHAT IS THE PROBLEM TO BE ADDRESSED ?

Lack of effective cure for Crohn's disease.

Effective symptomatic treatment of Crohn's disease can be achieved in approximately 60-80% using corticosteroids [1] with a similar response to enteral or intravenous feeding. The benefit is usually short-term however, with more than 50% relapsing within a year [2] and over 90% of patients with ileocaecal disease eventually requiring at least one surgical operation [3].

The cause of Crohn's disease is unknown but it is commonly believed to be due in some way to bacteria: either as an unusual response to the normal intestinal flora or possibly to a specific infection such as an atypical bacterium. It is typified histologically by the presence of granulomata (in over 75% of cases) and closely resembles intestinal tuberculosis, radiologically and histologically. A search for Mycobacteria has produced conflicting results with one centre claiming that Mycobacterium paratuberculosis DNA can be detected in the majority of cases [4] but with others finding Mycobacterial DNA in only about 10% [5]. Studies have shown that a variety of organisms including normal intestinal bacteria and yeasts can be grown from Crohn's disease mesenteric lymph nodes and that the lymphocytes within these lymph nodes are manufacturing antibodies that are directed against a broad range of bacterial antigens so it is a more widely held view that a wide range of micro-organisms invade the mucosa in Crohn's disease, possibly as a result of a breakdown in the normal mucosal barrier [6].

WHAT IS THE HYPOTHESIS TO BE TESTED?

That clarithromycin may be an effective therapy in Crohn's disease. It is a broad spectrum antibiotic that has particularly good penetration into macrophages [7] and may therefore be effective at eradicating the organisms at the centre of the granulomatous reaction in Crohn's disease. There is also some evidence suggesting that macrolide antibiotics can stimulate macrophage proliferation, phagocytosis, chemotaxis and cytocidal activitity [8].

WHY IS A TRIAL NEEDED NOW?

Open label studies have shown promising results with Clarithromycin. A study published in abstract by Present's group in New York reported a response in 7/12 (58%) whose Crohn's disease activity index (CDAI) fell by a mean of 130 points (range 79-130) in response to clarithromycin 500 mg bd for 6 weeks [9].

HAS A SYSTEMATIC REVIEW BEEN CARRIED OUT AND WHAT WERE THE FINDINGS?

A Medline search for " clarithromycin and Crohn's disease" yields only three responses. The first is the report of an open-label study of a combination of clarithromycin 250 mg bd (or azithromycin in 3 patients) and rifabutin 450 mg per day given for a mean of 18 months to 52 patients by the St George's Hospital group who are the major protagonists of the Mycobacterium paratuberculosis hypothesis [10] . They report a response in terms of "significant fall in Harvey Bradshaw Index" in 93% but CDAI was not reported and many of the patients seem to have had rather modest elevations of Harvey Bradshaw index on entry. Ten of the patients also received a quinolone antibiotic and a further 5 received clofazimine. The other two are reports of clarithromycin in Mycobacterium haemophilum infection including at least one in a case of apparent Crohn's disease. Review of American Gastroenterological Association abstracts for the past three years yielded reference [9] and an Australian study of a triple therapy regimen for Mycobacteria that consisted of rifabutin 450 mg per day, clarithromycin 750 mg per day, clofazimine 2 mg per kg. Twelve patients were treated for 8-12 months and 10/12 achieved "near complete control" [11].

HOW WILL THE RESULTS OF THIS TRIAL BE USED? This trial will establish whether clarithromycin is effective in achieving remission in Crohn's disease and would be a significant advance in non-steroid treatment of active Crohn's disease. If a significant positive result is observed then further trials of prolonged treatment would be indicated to assess clarithromycin's effect on the prevention of relapse.

WHAT ARE THE PLANNED TRIAL INTERVENTIONS? Patients will receive either (i) Clarithromycin S/R 1g od or (ii) placebo tablets of identical size, colour and taste.

WHAT IS THE PROPOSED DURATION OF THE TREATMENT PERIOD? Clarithromycin or placebo therapy will continue for three months.

WILL HEALTH SERVICE RESEARCH ISSUES BE ADDRESSED? Not Applicable

WHAT IS THE PROPOSED FREQUENCY/DURATION OF FOLLOW UP? Patients will be reviewed after one, two and four weeks and then monthly for the following two months of treatment. Follow up thereafter will then be in the gastroenterology outpatient clinics.

HOW WILL THE OUTCOME MEASURES BE MEASURED AT FOLLOW-UP? CDAI will be calculated at baseline and at each subsequent follow up from the patient's symptom scoring diary, haematocrit and weight. Similarly the van Hees index will be calculated and serum CRP measured. The IBD quality of life questionnaire will be completed at baseline and at 3 months.

Patients will also have a diary card to record the details of any other symptoms noted during the trial to assess adverse effects of the trial treatment.

WHAT ARE THE PROPOSED PRACTICAL ARRANGEMENTS FOR ALLOCATING PATIENTS TO TRIAL GROUPS? Randomisation will be allocated by the pharmacy department of the hospital.

WHAT ARE THE PROPOSED METHODS FOR PROTECTING AGAINST OTHER SOURCES OF BIAS? Controls (known only to the Pharmacy Department) will receive placebo tablets which are identical in size colour and taste. Patients will be stratified according to site of disease into 3 groups: colonic CD only, perianal disease only and others (ileocolonic, small bowel only).

WHAT IS THE PROPOSED SAMPLE SIZE? 39 patients in each group (active treatment and placebo) gives a 90% power of excluding a response of 60% (p2) compared to 20% (p1) for placebo at p<0.05 [16]. The published placebo response rate in active CD being 20%.

WHAT IS THE PLANNED RECRUITMENT RATE? 5 patients per month

ARE THERE LIKELY TO BE ANY PROBLEMS WITH COMPLIANCE? Compliance will be assessed by the number of returned tablets and assessed as good (<25% returned, fair 25-50% returned and poor (>50% returned).

WHAT IS THE LIKELY RATE OF LOSS TO FOLLOW UP? 100% follow up should be achievable.

HOW MANY CENTRES WILL BE INVOLVED? One

WHAT IS THE PROPOSED TYPE OF ANALYSIS? For the primary outcome measure the difference in proportions will be calculated with associated 95% confidence intervals. Formal hypothesis testing of the primary outcome will then be compared by chi-square test.

Quantitative variables will be compared using repeated measures analysis taking into account the distribution of each variable [17]

WHAT IS THE PROPOSED FREQUENCY OF ANALYSIS? Once only on completion.

ARE THERE ANY PLANNED SUBGROUP ANALYSES? Data from patients in each stratified group will be examined separately but formal statistical analysis between the subgroups will use a test for interaction (Altman DG. Practical Statistics for Medical Research. Chapman \& Hall. London. 1991). We anticipate though, that the main outcome data will be presented for all the patients together.

WHAT IS THE ESTIMATED RESEARCH COST OF THE TRIAL? To be discussed. Part funding will be required for research nurse and drug trials pharmacist.

IS THERE AN NHS SERVICE SUPPORT COST OF THIS TRIAL, AND IF SO WHAT IS THE ESTIMATED COST? The only NHS cost would be modest, involving only the routine testing of full blood count and CRP which is current practice in the monitoring of patients with relapses of inflammatory bowel disease.

OVER WHAT PERIOD IS FUNDING REQUESTED? Funding is requested for a 2 year period to cover the primary trial.

References:

1. Jarnerot G, Sandberg-Gertzen H, Tysk C. Medical therapy of active Crohn's disease. Ball Clin Gastroenterol 1998;12:73-92.
2. Binder V, Brynskov J. Corticosteroids. in Inflammatory bowel diseases. eds Allan, Rhodes, Hanauer, Keighley, Alexander-Williams, Fazio. Churchill Livingstone 3rd Edition. 1997 pp503-12.
3. Farmer RG, Whelan G, Fazio VW. Long term follow up of patients with Crohn's disease: relationship between clinical pattern and prognosis. Gastroenterology 1985;88:1818-1825.
4. Sanderson JD, Moss MT, Tizzard MLV, Hermon-Taylor J. Mycobacterium paratuberculosis in Crohn's disease tissue. Gut 1992;33:890-6.
5. Fiddler HM, Thurrell W, Rook GA, Johnson NH, McFadden JJ. Specific detection of Mycobacterium paratuberculosis DNA associated with granulomatous tissue in Crohn's disease. Gut 1994;35:506-10.
6. Sartor RB. Current concepts of the etiology and pathogenesis of ulcerative colitis and Crohn's disease. Gastroenterology Clinics of North America. 1995;24:475-507.
7. Fietta A, Merlini C, Gialdroni Grassi G. Requirements for intracellular accumulation and release of clarithromycin and azithromycin by human phagocytes. J Chemother 1997;9:23-31
8. Xu G, Negayama K, Yuube K, Hojo S, Yamaji Y, Kawanishi K, Takahara J. Effect of macrolide antibiotics on macrophage functions. Microbiol Immunol 1996;40:473-479.
9. Rubin PH, Chapman ML, Scherl E, Sachar DB, Stamaty C, Present DH. Clarithromycin in active Crohn's disease: preliminary results of open label pilot study. Gastroenterology 1996;110:A1005.
10. Gui GPH, Thomas PRS, Tizzard MLV, Lake J, Sanderson JD, Hermon Taylor J. Two year outcomes analysis of Crohn's disease treated with rifabutin and macrolide antibiotics. J Antimicrobial Chemotherapy 1997;39:393-400.
11. Borody TJ, Pearce L, Bampton PA, Leis S. Treatment of severe Crohn's disease using rifabutin- macrolide-clofazimine combination: interim report. Gastroenterol 1998;114: A938.
12. Best WR, Bccktel JM, Singleton JW, Kern F. Development of a Crohn's disease activity index. Gastroenterol 1976; 70:439-444.
13. VanHees PAM, van Elteren PH, van Lier HJJ, van Tongeren JHN. An index of inflammatory activity in patients with Crohn's disease. Gut 1980; : 279-286.
14. Guyatt G, Mitchell A . A new measure of health status for clinical trials in inflammatory bowel disease. Gastroenterol 1989;96:804-810
15. Irvine EJ, Feagan BG, Wong CJ. Does self administration of a quality of life index for inflammatory bowel disease change the results? J Clin Epid 1996;49:1177-1185.
16. Fleiss JL. Statistical methods for rates and proportions. John Wiley and Sons 1973
17. Matthews JN, Altman DG, Campbell MJ, Royston P. Analysis of serial measurements in medical research. BMJ 1990;300:230-5.