Oral Fluticasone Propionate in Active Distal Ulcerative Colitis

Gut, 1992, 33, 711-714 71

CLINICAL TRIAL Gut: first published as 10.1136/gut.33.5.711 on 1 May 1992. Downloaded from

Oral fluticasone propionate in active distal ulcerative colitis

P Angus, J A Snook, M Reid, D P Jewell

Abstract Patients and methods Fluticasone propionate is a new corticosteroid As the study was placebo-controlled, entry was with low systemic bioavailability. This study restricted to patients (aged 18-65 years) with reports the outcome of a double blind clinical mild or moderately active ulcerative colitis that trial comparing oral fluticasone propionate (5 did not extend beyond the splenic flexure. mg four times daily) with placebo for the Patients had to be systemically well and to be treatment of active distal ulcerative colitis. passing more than three but less than seven Sixty patients were treated for four weeks, with motions daily. Before inclusion, the presence of assessments at two and four weeks. One active inflammation was confirmed by rigid patient was withdrawn when she was found to sigmoidoscopy. In all patients, the extent of the have amoebiasis. Thus, results are presented disease had been established within the previous for 29 patients who received placebo and 30 five years by either colonoscopy or barium who received fluticasone propionate. The two radiology. groups were weli matched for age, sex, length Pregnant or breastfeeding women and patients of history, and extent of disease. After four with significant cardiovascular, renal, hepatic, or weeks of therapy the clinical, sigmoidoscopic, metabolic diseases were ineligible for inclusion.

and histological responses were similar in the The study was approved by the Central http://gut.bmj.com/ two groups. It is concluded that fluticasone Oxfordshire Research and Ethics Committee. propionate (5 mg four times daily) is not Written consent was obtained from each patient effective treatment for active distal ulcerative after a full explanation of the trial. colitis.

DRUGS AND DOSAGES

Corticosteroids have proved highly effective in Fluticasone propionate (5 mg tablets) and match- on September 30, 2021 by guest. Protected copyright. the treatment of acute attacks ofulcerative colitis ing placebo tablets were provided by Glaxo but their use is mitigated by side effects. Thus, Group Research Ltd. The active drug was used at the development of a corticoid with low systemic a dose of 20 mg daily. bioavailability would be a considerable advance. Several steroid compounds that have been shown to give very low blood concentrations and hence STUDY DESIGN have no effect on the hypothalamic-pituitary- Patients were randomly allocated to receive adrenal axis are available for topical use. These either fluticasone propionate (5 mg four times include betamethasone,' beclomethasone- 17 val- daily) or matching placebo tablets (one, four erate,2 and prednisolone metasulphobenzoate,3 times daily) for four weeks. The randomisation all of which are effective treatments for active code was generated using an in-house Glaxo distal ulcerative colitis. More recently, tixocortol program, PACT (Patient Allocation for Clinical pivalate4 and budesonide' have also proved as Trials). Patients were randomised in blocks of effective as a steroid enema. six. Fluticasone propionate is a new corticosteroid Concomitant treatment with systemic or topic- with low systemic bioavailability because of poor ally active steroids ofany kind was not permitted absorption from the gastrointestinal tract and during the study and patients who had received considerable first pass metabolism Group such medication within the 14 Gastroenterology Unit, (Glaxo previous days Radcliffe Infirmary, Research - personal communication). When were excluded. Patients were allowed to continue Oxford given in a single oral dose to patients with a well longterm treatment with sulphasalazine, P Angus established ileostomy, 72.7% of the drug was mesalazine, or olsalazine at the same dose pro- J A Snook M Reid recovered in the ileostomy effluent (J A Snook vided that therapy had already begun before the D P Jewell and D P Jewell; unpublished observations). relapse that brought the patient into the trial. Correspondence to: The purpose ofthis study was to evaluate, over At the beginning of the study, frequency of Dr D P Jewell, Gastroenterology Unit, a four week period, the efficacy and safety of 20 bowel actions, stool consistency, and severity of Radcliffe Infirmary, mg of fluticasone propionate given orally in rectal bleeding were noted. The macroscopic Woodstock Road, Oxford OX2 6HE. divided doses (5 mg four times daily) compared appearance of the rectal mucosa was graded 0-3 Accepted for publication with placebo for the treatment of active ulcera- according to the criteria of Baron et al.6 Based on 5 August 1991 tive colitis. the symptoms, clinical signs, and endoscopic 712 Oralfluticasone propionate in active distal ulcerative colitis

findings, an overall physician's impression was noted improvement in their symptoms of colitis given as follows: 0-remission, 1-mild, 2-mode- (Table II). The physician's overall assessments rate, 3-severe, and 4-fulminant disease. A rectal showed that 14 of 30 patients (47%) treated with biopsy specimen was taken and the severity ofthe fluticasone propionate had improved or gone into Gut: first published as 10.1136/gut.33.5.711 on 1 May 1992. Downloaded from inflammation seen on histological sections remission by the end of the trial compared with stained with haematoxylin and eosin was graded 13 of 29 (45%) receiving placebo (p=0.888). according to the criteria of Powell-Tuck et al.7 Only four patients receiving fluticasone propio- The histopathologist (MR) was unaware of the nate and five on placebo were considered to be in patients' clinical findings. A full physical exami- remission as defined by the absence of symptoms nation was performed at entry and blood samples and quiescent disease on sigmoidoscopy. were taken for routine haematology, biochemis- The sigmoidoscopic gradings at four weeks (or try, and measurement of erythrocyte sedimenta- at the time of withdrawal) improved by at least tion rate, C-reactive protein, and orosomucoids. one grade in 13 in the fluticasone propionate Urine analysis was performed at each visit. group (43%) compared with 14 in the placebo Patients were seen in the outpatient clinic at group (48%) (p=0706). In one patient on two and four weeks after starting treatment. At fluticasone propionate and in four receiving both visits, symptom assessment, sigmoido- placebo the sigmoidoscopic appearances scopic and histological grading of disease returned to normal (grade 0). activity, general examination, and blood tests Forty eight of the 49 patients who completed were repeated. the study had adequate rectal biopsy specimens The effect of treatment on endogenous corti- taken at their last visit. The improvement in costeroid production was monitored by compar- histological scores was similar in the two groups - ing cortisol concentrations in saliva at 9.00 am on 11 showing improvement in the fluticasone pro- day 1 of the study (before entry) with the pionate group (48%) compared with nine in the concentrations on day 28 of treatment or at the placebo group (36%), p=0-411. Complete reso- time of withdrawal. The cortisol was measured lution of inflammation was noted in the biopsy by direct radioimmunoassay by BioClinical specimens of only one fluticasone propionate Services Ltd, Cardiff. patient and two placebo patients.

STATISTICAL ANALYSIS TABLE I Patient characteristics The number of patients to be recruited was Group calculated assuming a remission rate of 80% with

fluticasone propionate and 35% with placebo. Fluticasone http://gut.bmj.com/ Type 1 error, 2-tail, was set at 0.05 and type 2 propionate Placebo error, 1-tail, set at 0.1 to give a power of 90%. Sex: Male 18 12 Female 12 17 This gave a total sample size of 60 patients Age (years): 41 36 allowing for a proportion ofnon-assessable cases. Median (range) (25-64) (18-68) Weight (kg): 73-3 (13.9) 68-7 (11.8) Comparison of the changes in symptoms, Mean (SD) signs, and histological scores after four weeks Temperature ('C): 36-5 (0.2) 36-6 (0.2) was X2 test. t Mean (SD) performed using the The Student's Pulse (beats/min): 73 (6) 75 (8) on September 30, 2021 by guest. Protected copyright. test was used to compare mean salivary cortisol Mean (SD) Length ofhistory (months): 72 48 concentrations. Results were considered signifi- Median (range) (2-240) (2-240) cant if p was <0 05. Duration ofcurrent symptoms (weeks): 4 8 Median (range) (2-32) (1-24) Extent ofdisease: Rectum 14 12 Rectosigmoid 6 6 Results To splenic flexure 10 11 Sixty patients were recruited into the study. However, one patient was withdrawn after the biopsy report at 14 days which described amoe- TABLE II Percentage ofpatients completing the trial who biasis. She had been successfully treated for reported an improvement in symptoms ulcerative colitis during the previous two years Group but she was excluded from all subsequent analysis. Therefore, data from 59 patients have Fluticasone propionate Placebo been analysed on an intention to treat basis; 29 (n=23) (n =26) Significance received placebo and 30 received fluticasone Bowel frequency 54 58 p=0 804 propionate. The demographic and clinical Stool consistency 40 52 p=0.430 characteristics ofthe two groups were similar and Blood in stools* 36 39 p=0.850 are com- shown in Table I. Forty nine patients * Only 22 in the fluticasone propionate group and 23 in the placebo pleted the study. Nine were withdrawn (6 flut- group reported blood on entry. icasone propionate, 3 placebo) during the course of the trial due to lack of response (day of TABLE III Salivary cortisol concentrations (nmolll, mean withdrawal ranged from 5-18). One additional (SD)) patient on fluticasone propionate was withdrawn Group after an adverse event (see later). Fluticasone No. propionate Placebo Significance ASSESSMENT OF DISEASE ACTIVITY At entry 25 110 (16.0) 8.4(8.1) After28days 21 6-4 (4.8) 9-1(6.1) p=0-0849 For those who completed four weeks of treatment, similar numbers of patients in each group Normal range 5-25 nmol/1 Angus, Snook, Reid,Jewell 713

Changes in haemoglobin, platelet count, inflammation while avoiding the systemic side erythrocyte sedimentation rate, C-reactive pro- effects ofother corticosteroid preparations. tein, orosomucoids, and serum albumin were This trial indicates that, in a dose of 20 mg weeks. daily, fluticasone propionate is no more effective similar for both groups at two and four Gut: first published as 10.1136/gut.33.5.711 on 1 May 1992. Downloaded from than placebo in the treatment ofactive ulcerative colitis. It is possible that the results were com- SALIVARY CORTISOL pounded by the high placebo response but this Eighteen patients in the fluticasone propionate degree of response has been noted in previous group and 20 in the placebo group returned placebo controlled studies.89 However, there was adequate salivary samples on entry and at day 28. not even a trend in favour of fluticasone pro- There was no significant change in the mean pionate and, although many patients showed salivary cortisol concentrations in either group at improvement, few in either group (13% fluti- the end ofthe study (Table III). casone propionate, 17% placebo) actually went into clinical remission. Furthermore, the power of the study was such that even with a placebo ADVERSE EVENTS response of 40%, the probability of failing to Two patients who received fluticasone propion- detect a true benefit of fluticasone propionate ate (rash on arms (1), tension headache (1), and over placebo was less than 5%. three who received placebo (herpes labialis (2), Since topical corticosteroids are known to be light-headedness (1)) reported trivial adverse effective in the treatment of active ulcerative events that resolved or improved with continuing colitis, and fluticasone propionate has a topical treatment. One further patient in the fluticasone action in the bronchial tree,'" why was it not propionate group developed an acute oligo- effective in this trial? Perhaps the most likely arthritis two days before completion of the trial. explanation is that the drug did not reach the This patient had not improved during the course site of disease in sufficient concentration or of the trial and the arthritis was considered to be in the appropriate form. None of the patients in related to active ulcerative colitis rather than the the study had colitis that extended beyond the effect of fluticasone propionate treatment. Both splenic flexure and in many (approximately two the colitis and the joints responded rapidly to thirds) the disease was confined to the rectosig- subsequent treatment with prednisolone. moid. After defecation, even in patients with One other patient in the fluticasone propionate diarrhoea, the distal colon is likely to be largely group developed a more serious adverse reaction. empty of faeces for some time. Thus, the most A 43 year old woman with no history of diabetes actively inflamed part of the colon might only be

and a normal pretreatment plasma glucose con- exposed to the drug intermittently. Further- http://gut.bmj.com/ centration, was noted to have glycosuria and a more, in patients with diarrhoea, the concentra- raised plasma glucose value (20-9 mmol/l) after tions of fluticasone propionate achieved in the two weeks of treatment and was subsequently colon may be considerably lower than those withdrawn from the trial. At the time of with- achieved in controls. Studies of the faecal excre- drawal, the patient's symptoms of colitis had tion of fluticasone propionate in patients will be resolved, the sigmoidoscopic appearance of the required to answer these questions.

rectal mucosa was normal, and histological Although, in general, the drug was well toler- on September 30, 2021 by guest. Protected copyright. examination of the rectal biopsy specimen ated, one patient developed hyperglycaemia with showed resolution of the active inflammation. evidence of suppression of endogenous corti- Her salivary cortisol concentration had fallen costeroid production. There was no reason to from 15 nmol/l on day 1 to unmeasurable levels suspect drug overdosage or liver dysfunction in on the day of withdrawal. It was felt that the this patient and her colitis did not seem to be hyperglycaemia was almost certainly due to a more active or more extensive than in many other systemic effect of the drug. Blood glucose con- patients in the trial. Thus, it must be assumed centrations returned to normal within two days that this patient's hyperglycaemia was due to the ofstopping therapy and the patient subsequently systemic effect of the drug. It is of interest that remained normoglycaemic. the colitis went into complete remission at two weeks, which might also suggest an abnormal systemic bioavailability of the drug. Discussion In conclusion, oral fluticasone propionate, 5 Preliminary studies in animals and in human mg four times daily is not effective treatment for volunteers indicated that oral fluticasone pro- active distal ulcerative colitis. More knowledge pionate has poor systemic bioavailability. As a of the pharmacodynamics of the drug may be preliminary study to the therapeutic trial, eight helpful to design studies using higher concentra- patients with a well established ileostomy (all tions. It is possible that the drug will prove more patients having had ulcerative colitis) were given effective in the treatment of inflammation in the a single oral dose of 5 mg fluticasone propionate. small bowel or in the proximal colon. Ileostomy effluent was collected over the subse- We are grateful for generous support and advice given by Glaxo quent 24 hour period in three aliquots (0 6, Group Research Ltd. 6-12, 12-24 hours). Fluticasone propionate was 1 Multicentre Trial. Betamethasone 17-valerate and predni- subsequently assayed and the mean recovery rate solone 21-phosphate retention enema in proctocolitis. BMJ was 72*7% (range 45 7-97 6%), mostly within 12 1971; ii: 84-6. hours. (J A Snook and D P Jewell; unpublished 2 Kumana CR, Seaton T, Meghji M, Castelli M, Benson R, Sivtakumaran T. Beclomethasone dipropionate enemas for observations). It was therefore hoped that, in treating inflammatory bowel disease without producing patients with colitis, the drug would reach the Gushing's syndrome or hypothalmic pituitary suppression. colon in sufficient quantities to suppress the Lancet 1982; i: 579-82. 714 Oralfluticasone propionate in active distal ulcerative colitis

3 McIntyre PB, Macrae F, Berghouse L, English J, Lennard- 7 Powell-Tuck J, Day DW, Bucknell NA, Wadsworth J, Jones JE. Therapeutic benefits from a poorly absorbed Lennard-Jones JE. Correlations between defined sig- prednisolone enema in distal colitis. Gut 1985; 26: 822-4. moidoscopic appearances and other measures of disease 4 Hanauer SB. Clinical experience with tixocortol pivalate. Can activity in ulcerative colitis. Dig Dis Sci 1982; 27: 533-7. J Gastroenterol 1988; 2: 156-8. 8 Selby WS, Barr GD, Ireland A, Mason CH, Jewell DP. 5 Danielsson A, Hellers G, Lyrenas E, Lofberg R, Nilsson A, Olsalazine for the treatment of active ulcerative colitis. BMJ Gut: first published as 10.1136/gut.33.5.711 on 1 May 1992. Downloaded from Olsson 0, et al. A controlled randomised trial of budesonide 1985; 291: 1373-5. versus prednisolone retention enemas in active distal colitis. 9 Burke DA, Axon ATR, Clayden SA, Dixon MF, Johnston D, ScandJ Gastroenterol 1987; 22: 987-92. Lacey RW. The efficacy of Tobramycin in the treatment of 6 Baron JH, Connell AM, Lennard-Jones JE. Variation between ulcerative colitis. Aliment Pharmacol Therap 1990; 4: 123-9. observers in describing mucosal appearances in procto- 10 Bauer K, Bamtje TA, Sips AR, Bogaerts YJM, Gillard C, colitis. BMJ 1964; i: 89-92. Kardos P, etal. EurRespir] 1988; 1 (suppl 2): 201. http://gut.bmj.com/ on September 30, 2021 by guest. Protected copyright.