V•37
`
`C . 02 (cid:9)
`TI: GUT
`
`WI GU821
`NO.S (cid:9)
`199a
`SEQ: u14200000
`
`1 1/ 20/95 ut
`
`AN INTERNATIONAL JOURNAL OF GASTROENTEROLOGY AND HEPATOLOGY
`
`Leading articles
`
`593 (cid:9) Metal stents in the oesophagus R P Sturgess, A I Morris
`Peptides and gastrointestinal mucosal integrity R1 Playford
`595 (cid:9)
`
`Oesophagus
`
`Peptic ulcer
`
`598 (cid:9)
`
`603 (cid:9)
`
`Endocytosis of fluorescent microspheres by human oesophageal epithelial cells: comparison
`between normal and inflamed tissue D Hopwood, E M Spiers, P E Ross, J Anderson,
`B McCullough, F E Murray
`Oesophageal sensation assessed by electrical stimuli and brain evoked potentials — a new
`model for visceral nociception 0 Frobert, L Arendt-Nielsen, P Bak, P Funchjensen,
`P Bagger
`
`610 (cid:9)
`
`613 (cid:9)
`
`Basic fibroblast growth factor treatment for non-steroidal anti-inflammatory drug
`associated gastric ulceration M A Hull, D J E Cullen, N Hudson, C J Hawkey
`Corticosteroids reduce regenerative repair of epithelium in experimental gastric ulcers
`M Carpani de Kaski, R Rentsch, S Levi, H J F Hodgson
`617 (cid:9) Mortality after remote surgery for benign gastroduodenal disease
`C C S Stael von Holstein, H Anderson, S B S Eriksson, B Huldt
`
`Small intestine
`
`623 (cid:9)
`
`630 (cid:9)
`
`639 (cid:9)
`
`Intestinal absorptive capacity, intestinal permeability and jejunal histology in HIV and their
`relation to diarrhoea J Keating, I Bjarnason, S Somasundaram, A Macpherson, N Francis,
`A B Price, D Sharpstone, J Smithson, I S Menzies, B G Gazzard
`Administration of insulin-like growth factor-I (IGF-I) peptides for three days stimulates
`proliferation of the small intestinal epithelium in rats C B Steeb, J F Trahair, L C Read
`Coeliac disease and bone mineral density in adult female patients L R Pistorius,
`W H Sweidan, D W Purdie, S A Steel, S Howey, J R Bennett, D R Sutton
`
`Motility
`
`Inflammatory
`bowel disease
`
`643 (cid:9)
`
`Evaluation of antral motility in humans using manometry and scintigraphy K Jones,
`M Edelbroek, M Horowitz, W-M Sun, J Dent, J Roelofs, T Muecke, L Akkermans
`649 (cid:9) Motor activity recorded in the unprepared colon of healthy humans M Lemann,
`B Flourie, L Picon, B Coffin, R Jian, J C Rambaud
`Effects of simvastatin and cholestyramine on bile lipid composition and gall bladder
`motility in patients with hypercholesterolaemia J W A Smit, K J Van Erpecum,
`P Portincasa, W Renooij, I) W Erkelens, G P Van Berge-Henegouwen
`
`654 (cid:9)
`
`660 (cid:9) Mycobacterium paratuberculosis DNA not detected in Crohn's disease tissue by fluorescent
`polymerase chain reaction D S Rowbotham, N P Mapstone, L K Trejdosiewicz,
`P D Howdle, P Quirke
`668 (cid:9) Meta-analysis of the role of oral contraceptive agents in inflammatory bowel disease
`P G Godet, G R May, L R Sutherland
`A controlled double blind study of azathioprine in the management of Crohn's disease
`S Candy, J Wright, M Gerber, G Adams, M Gerig, R Goodman
`
`674 (cid:9)
`
`679 (cid:9)
`
`Quantitative assessment of overall inflammatory bowel disease activity using labelled
`leucocytes: a direct comparison between indium-111 and technetium-99m HMPAO
`methods J C Mansfield, M H Giqffer, W B Tindale, C D Holdsworth
`
`PROPERTY
`llaI6 OFI THE
`LIBRARY OF
`MEDICINE
`
`BMJ
`
`CONTENTS CONTINUED INSIDE
`
`Ex. 1070 - Page 1
`
`

`

`CONTENTS CONTINUED
`
`684 (cid:9)
`
`690 (cid:9)
`
`696 (cid:9)
`
`702 (cid:9)
`
`708 (cid:9)
`
`712 (cid:9)
`
`721 (cid:9)
`
`727 (cid:9)
`
`731 (cid:9)
`
`Kinetic studies on colonocyte metabolism of short chain fatty acids and glucose in
`ulcerative colitis M Rye Clausen, I' B Mortensen
`Transglutaminases in Crohn's disease G D'Argenio, L Biancone, V Cosenza,
`N Della Valle, F 1' D'Artniento, M Boirivant, I; Palime, G Mazzacca
`F Makowiec, E C lehle, M Starlinger
`
`Clinical course of perianal fistulas in Crohn's disease
`Oral efficacy of a leukotriene B , receptor antagonist in colitic cotton-top tamarins
`I) Fretland, T Sanderson, I' Smith, L Adams, I? Carson, 7 Fuhr, j Tanner, N Glop
`I. Beaugerie, N Patey, N Brousse
`Ranitidine, diarrhoea, and lymphocytic colitis
`
`Overexpression of nn/23-111 and nnt23-.H2 genes in colorectal carcinomas and loss Of
`,
`nm23-H I expression in advanced tumour stages 1 A Martinez, S Prevot, B Nordhnger
`I M A Nguyen, y Lac:at-rim, A Mintier, I Lascu, C Valliant, J capeau, M L Lacombe
`
`Efficacy of longterm interferon treatment in chronic liver disease evaluated by sensitive
`0 Yokosuka, N Kato,
`polymerise chain reaction assay for hepatitis C virus RNA
`K Hosoda, Y Ito, F Imazeki, M Ohto, M Omata
`
`Gangaidzo, V RGordeuk
`Hepatocellular carcinoma and African iron overload I ;/'
`
`A Kapur,
`
`Linear IgA dermatosis, coeliac disease, and extraintestinal B cell lymphoma
`F 7' Isaacs, P R Kelsey
`
`Large intestine
`
`Liver
`
`Hypothesis
`
`Case report
`
`International (cid:9)
`Gastroenterology
`
`734 A D Beattie
`
`736 (cid:9) Book reviews Notes
`
`
`
`Th is m atari a 1 Was copi ed
`
`Ex. 1070 - Page 2
`
`

`

`674
`
`_
`
`Gut 1995; 37: 674-678
`
`A controlled double blind study of azathioprine in
`the management of Crohn's disease
`
`S Candy, J Wright, M Gerber, G Adams, M Gerig, R Goodman
`
`Abstract
`While immunosuppressive agents are
`used widely in the management of
`Crohn's disease, their efficacy has not
`been well established in randomised con-
`trolled trials. This study was designed to
`examine whether azathioprine increases
`remission rate when used in conjunction
`with a diminishing dose regimen of pred-
`nisolone over a period of 12 weeks. It
`further examined whether azathioprine
`offers any therapeutic advantage over
`placebo in the maintenance of remission
`in Crohn's disease over a period of 15
`months. Sixty three patients with active
`Crohn's disease were treated with a 12
`week diminishing dose of prednisolone
`and at the same time entered into a
`randomised, double blind 15 month
`trial of either azathioprine (2.5 mg/kg) or
`placebo. Remission rates between the two
`groups were compared at 12 weeks and at
`15 months. There was no significant
`difference in the proportion of patients
`who had achieved and maintained
`remission by week 12 but at 15 months
`there was a highly significant difference in
`the proportion of patients in remission
`(42% receiving azathioprine v 7% receiv-
`ing placebo), p=0.001. Using life tables
`this beneficial effect was reflected as the
`difference in the median number of days
`on the trial (p =0.02). There were signifi-
`cantly greater decreases over the trial
`period in the median erythrocyte sedi-
`mentation rate, C reactive protein, and
`leucocyte count in the azathioprine group.
`There were no cases of severe bone
`marrow suppression or clinical pancreati-
`tis. In conclusion, azathioprine offers a
`therapeutic advantage over placebo in the
`maintenance of remission in Crohn's
`disease.
`(Gut 1995; 37: 674-678)
`
`Keywords: Crohn's disease, azathioprine,
`immunosuppression.
`
`clinical settings but their efficacy has yet to be
`proved unequivocally in controlled trials. A
`recent meta-analysis of all published single
`drug controlled trials concluded that no single
`drug affords a therapeutic advantage in the
`maintenance of remission in Crohn's disease.'
`This study was designed to examine the com-
`bined effect of azathioprine and prednisolone
`in inducing remission in active Crohn's disease
`and further to test the efficacy of azathioprine
`alone in the maintenance of remission.
`
`Methods
`
`PAT 11'.1511
`The study was approved by the Ethics and
`Research Committee of the Faculty,. of
`Medicine, Groote Schuur Hospital. Each
`patient gave written informed consent before
`entry into the study. Patients between the ages
`of 15 and 65 years with confirmed Crohn's
`disease who were being followed up at the
`Groote Schuur Hospital Inflammatory Bowel
`Disease Clinic provided the source population.
`The inclusion criteria were a radiological or
`endoscopic diagnosis of Crohn's disease and
`the presence of active disease determined by a
`score of 200 or more on the Crohn's Disease
`Activity Index." Patients excluded were thos,e
`with extensive previous surgery for Crohn s
`disease or those with symptoms suggestive of a
`mechanical obstruction that might require
`imminent surgery. Recent immunosuppres-
`5100, compromised hepatic function, preg-
`nancy, and lactation were further exclusions:
`s
`At entry, age, the time since onset Of symptom
`
`and the time since diagnosis, smoking status,
`extent of
`disease, previous surgery, previous
`medical management (specifically any recent
`corticosteroid treatment), disease activity
`(reflected by the CDAI, erythrocyte sechmen:
`tation rate (ESR), serum C reactive protein
`(CRP) and orosomucoid) and the presence or
`(cid:9) were
`absence of perianal disease and fistulas
`recorded.
`
`The aetiology of Crohn's disease is unknown
`and as a result treatment options are limited.
`Corticosteroids, despite their side effects,
`remain the treatment modalities of choice.
`Intensive research has been directed at finding
`a safe and effective alternative to corticos-
`teroids, particularly for use in resistant Crohn's
`disease and in the maintenance of remission.
`Immunosuppressive drugs such as azathio-
`prine and its active metabolite, 6-mercapto-
`purine, have been used extensively in these
`
`STUDY DRUG
`The azathioprine formulation was in the form
`of film coated tablets, trade name imuran
`(Wellcome Pty, Kempton Park, South Africa).
`
`STUDY DESIGN
`This was a single centre, two phase, douhlde
`blind, placebo controlled trial that compare
`,
`rednisolone and
`the combined effect of p and placebo Ina
`azathioprine v prednisolone
`the treatment of active Crohn's disease over
`
`
`
`This m aterial was copied
`
`Gastrointestinal
`Clinic, Department of
`Medicine, University
`of Cape Town and
`Groote Schuur
`Hospital, South Africa
`S Candy
`J Wright
`M Gerber
`G Adams
`M Gerig
`R Goodman
`
`Correspondence to:
`Dr S Candy, Groote Schuur
`Hospital, Gastro-intestinal
`Clinic, E23, E-Floor,
`NGSH, Observatory, Cape
`Town 7925, South Africa.
`
`Accepted for publication
`16 March 1995
`
`Ex. 1070 - Page 3
`
`

`

`44%041,
`
`A controlled double blind study of azathioprine in the management of Crohn's disease
`
`675
`
`cyte count at week 1 was normal, the patient
`was advised to take the full dose (2.5 mg/kg to
`the nearest 50 mg) as a single daily admini-
`stration. Where the leucocyte count dropped
`to less than 4X 109/1, the dose of trial drug
`was reduced, usually by 50 mg initially. The
`count was then monitored closely. In the
`event of good recovery, the original dose was
`reinstated. Compliance was assessed at each
`subsequent visit by means of a tablet count.
`
`EVALUATION METHODS
`Patients were asked to keep a diary charting
`stool number and consistency, abdominal
`pain, general wellbeing, concomitant medica-
`tion, and any adverse clinical experiences.
`There were two evaluation points; the first
`being at cessation of prednisolone at 12 weeks
`and the second at the end of 15 months. The
`primary measure of therapeutic response was
`in each case the proportion of patients in
`remission. Remission was defined at 12 weeks
`as a CDAI of <150 and at 15 months as a
`CDAI <175. Withdrawal for any reason other
`than clinical relapse was considered thera-
`peutic failure.
`Secondary measures of therapeutic response
`were the median change in CDAI, ESR, serum
`CRP, serum orosomucoid concentration, and
`leucocyte count between the first and last
`visits.
`The difference in leucocyte count between
`baseline and final visit was compared for
`responders v non-responders within the aza-
`thioprine group to ascertain whether induction
`of leucopenia predicts clinical response.
`
`period of 12 weeks (phase 1). The second
`phase (phase 2) compared azathioprine alone
`v placebo in the maintenance of remission
`induced in phase 1 over a further 12 months.
`Patients were recruited as they presented
`over a 36 month period. They were stratified
`according to disease extent into three groups;
`group 1: those with only ileal disease, group 2:
`those with ileocolitis, and group 3: those with
`isolated colonic disease. They were then ran-
`domised to either active medication or placebo
`to ensure equal representation of disease
`extent within the treatment groups. At entry a
`baseline assessment of the previous weeks'
`symptoms was made and the patients were
`examined clinically. Blood was drawn to
`measure ESR, serum CRP and orosomucoid,
`as well as comprehensive haematological and
`biochemical indices.
`Visits were scheduled weekly for the first
`month, twice weekly for the second month, and
`monthly thereafter for the remainder of the trial
`or until withdrawal. Predetermined withdrawal
`criteria were: failure to achieve remission within
`the first 12 weeks (defined by a CDAI of
`--ISO), relapse (defined as a CDAI >175) at
`any stage thereafter, patient request, discretion
`of the investigators and bone marrow suppres-
`sion, defined as a haemoglobin of less than
`7.0 g/dl, a white cell count of less than 2 x 109/1
`or a platelet count lower than 100X109/1.
`Haematological and biochemical indices were
`monitored independently by a physician not
`involved in the clinical assessment. Decisions
`regarding drug dose adjustment were relayed to
`the patient by the nurse practitioner who was
`also responsible for symptom assessment,
`drawing blood, counting remaining tablets, and
`issuing the trial medications.
`The physician responsible for the clinical
`examination recorded any reported side effect
`or adverse event experienced by the trialist
`since the last visit and noted any concomitant
`medication taken.
`All medications for Crohn's disease were
`stopped at entry except for antidiarrhoeal
`drugs, which too were recorded at each visit.
`
`STATISTICAL METHODS
`All patients randomised into the trial were
`analysed — that is, analysis was performed on
`an intention to treat basis. No data necessary
`for the calculation of the CDAI were missing.
`A two-tailed X2 test was used to analyse the
`difference in the proportion of patients in
`remission at week 12 and at trial end. A
`Wilcoxon test was used to compare the median
`change from baseline to last visit in CDAI,
`ESR, CRP, orosomucoid concentration, and
`leucocyte count between the two groups.
`The calculated 'relative risk' — that is, the rela-
`tive protective effect proferred by azathioprine —
`was adjusted for key potential confounders
`using the Mantel-Haenszel calculation. The
`simultaneous influence of a number of covari-
`ates on the association of treatment status and
`outcome was examined by means of the logistic
`regression procedure of SAS (Ver 6.08). A
`Kaplan-Meier life table analysis was used to
`compute the median time to relapse. Analysis of
`variance was used to compare the median time
`to relapse by disease extent.
`
`Results
`
`PATIENTS
`A total of 63 patients were randomised to
`receive either placebo or azathioprine in
`
`slimy DRUG ADMINISTRATION
`At day 1, all randomised patients were given
`prednisolone at a starting dose of 1 mg/kg/day.
`They were requested to decrease this dose by
`5 mg per week to zero at the end of week 12.
`Where a patient weighed less than 60 kg, the
`starting dose was maintained for a longer
`period of time before decreasing to ensure that
`prednisolone was continued throughout phase
`I. Similarly, if the weight was greater than 60
`kg, the dose was reduced more rapidly over the
`first two weeks to ensure completion of
`prednisolone by the end of week 12.
`Also on day 1, patients were presented with
`tablets of identical appearance (scored '50 mg'
`on one side and 'Imuran' on the other). The
`dose was calculated at 2.5 mg/kg but each
`patient was asked to take only one tablet (50
`mg) per day lbr the first week as a preventative
`measure against idiosyncratic leucopenia in
`those randomised to azathioprine. If the leuco-
`
`This material was copied
`
`Ex. 1070 - Page 4
`
`

`

`676
`
`TABLE 1 Baseline characteristics
`
`Baseline characteristic
`
`Male (%)
`Female (%)
`Age (y) (median and range)
`Weight (kg) (median and range)
`Duration of symptoms (y) (median and range)
`Time since diagnosis (y) (median and range)
`Extent (I:IC:C)
`Previous surgery (%)
`Episode in previous 6-12 months (%)
`Corticosteroids in previous 6-12 months (%)
`No previous corticosteroids (%)
`Smoking status (%)
`Severe disease (CDAI >300) (%)
`CDAI*
`ESR (min/h)*
`CRP (mg%)*
`Orosomucoid (g/1)*
`Leucocyte count (x 109/1)*
`
`Azathioprine (n-33)
`
`Placebo (n-30)
`
`7 (21)
`26 (79)
`33.9 (15-60)
`53.1 (46.7-62)
`3.7 (0.6-19.3)
`2.6 (0'1 19.0)
`8:20:5
`6 (18)
`28 (85)
`22 (67)
`5 (15)
`22 (67)
`17 (52)
`301 (264-358)
`.45 (20-59)
`5.4 (2.9-7.3)
`1.85 (1.27-2.52)
`9.0 (7.0-10.9)
`
`11 (37)
`19 (63)
`31.8 (21-62)
`57.5 (48-64)
`4.7 (0.2-19•4)
`3.7 (0.1-18.7)
`6:19:5
`8 (27)
`21 (70)
`19 (63)
`
`2 7) (
`20 (67)
`13 (43)
`282 (240-356)
`26 (15-42)
`3.9 (2.8-51)
`1.79 (1.47-2.52)
`10.0 (5.6-12.6)
`
`*Median and interquartile range. (cid:9)
`
`C-colonic disease.
`
`addition to a 12 week course of prednisolone.
`The treatment and placebo groups were
`similar with respect to the basic demographic
`data (Table I).
`
`EARLY DISCONTINUATION OF THERAPY
`At 12 weeks, 20 of 63 randomised patients
`(32%) had been withdrawn: nine of 33 (27%)
`in the azathioprine group and 11 of 30 (37%)
`in the placebo group.
`Overall, 47 (74.6%) of those randomised
`withdrew before the end of the trial: 28 of 30
`(93%) in the placebo group and 19 of 33
`(58%) in the azathioprine group. Table II gives
`the reasons for withdrawal.
`
`PRIMARY EFFICACY PARAMETERS
`
`Remission at 12 weeks
`There was no significant difference between
`the proportions of patients who had achieved
`and maintained remission at 12 weeks: 24 of
`33 (73%) in the azathioprine group v 19 of 30
`(63%) receiving placebo (p
`
`Remission at 15 months
`There was a highly significant difference
`between the proportions of patients who were
`still in remission at the end of the trial period: 14
`of 33 (42%) in the azathioprine group v two of
`30 (7%) receiving placebo, (1)=0.001).
``Relative risk' (relative protective effect of aza-
`thioprine v placebo was 6.36 (95% confidence
`intervals (CI) 1.6 to 25.7). This ratio remained
`clinically and statistically significant after adjust-
`ing for disease severity (RR-=6.84 (1.68 to
`
`TABLE 11 Reasons for withdrawal
`
`Reason
`
`No response phase I
`Default
`In remission
`Not in remission
`Clinical side effect
`Haematological side effect
`Patient request
`Clinical relapse (phase 2)
`Other, for example, pregnancy
`
`Treatment group
`
`Placebo (cid:9)
`(n-28) (cid:9)
`
`Azathioprine
`(n=19)
`
`9
`
`0
`2
`0
`
`13
`
`7
`
`2
`
`7
`
`This material was copi ed
`fibs hLLhd and may be
`
`Candy, Wright, Gerber, Adams, Gerig, Goodman
`
`24.93), time since diagnosis (RR=6.48 (1.68 to
`24.93)), and sex (RR= 6.9 (1.65 to 28.8)).
`The influence of a number of variables was
`further examined with logistic regression using
`a stepwise procedure, without changing the
`result. The analysis was repeated after exclud-
`ing defaulters and those withdrawn for reasons
`other than relapse; the result remained
`clinically and statistically highly significant.
`The question of effect modification by
`disease extent was considered by stratifying
`into three groups; ileitis (n-14), ileocolitis
`(n=39), and colitis (n=10). Table III gives
`a summary of this analysis. Although the
`category `ileocolitis' seems to predispose to a
`CI
`more favourable outcome (OR= 14.7 (95%
`2.2 to 96•5), the small numbers in the other
`groups and wide confidence intervals make
`analysis meaningless.
`
`Life table analysis
`Overall - the median number of days on the
`trial was 263 (range 4-540) with azathioprine v
`139 (range 7-446) with placebo (v-0.02)
`(Figure).
`Analysed by disease extent -
`number of days on the trial for those ram
`domised to azathioprine was 345 where disease
`was restricted to the ileum, 249 for those with
`ileocolitis, and 233 for those with colitis alone
`(p=0.57). Similarly, for those receiving
`placebo, the number of days on the trial were
`96, 147, and 141 respectively (p= 0'74).
`
`the median
`
`SECONDARY EFFICACY PARAMETERS
`
`At 12 weeks
`There were no significant differences from
`baseline between the two groups with respect
`0•06) or ()roo
`to
`=0•9), CRP (cid:9)
`CDA1 (cid:9)
`mucoid (p0•8). 'The median LSR increased
`,
`by 1.5 nun/hour among those receiving
`17,
`prednisolone alone while it dropped by
`,..
`nun/hour in thosepatients receiving azatilt°
` p
`1 (cid:9),). rine. This represents a highly significant 04)
`
`ference in favour of azathioprine 09 '4.0
`
`At 15 months
`Fable IV lists the median differences in
`for trial corn
`secondary efficacy parameters
`mencement to trial end.
`
`end
`TABLE III Proportion of patients in remission at trial
`by disease extent (cid:9)
`
`Responders (n•=16)
`
`Non-responders 01'47)
`
`14
`2
`
`4
`
`Total (n=63)
`Azathioprine
`Placebo
`Ileitis (n=14)
`Azathioprine
`Placebo
`Ileocolitis (n-39)
`Azathioprine
`Placebo
`Colitis (n-10)
`Azathioprine
`Placebo
`
`19
`18.
`
`it
`
`11
`
`1A)1( (cid:9)
`I l
`
`I./
`
`1 h
`
`SE (cid:9)
`SE
`
`I
`
`Ex. 1070 - Page 5
`
`(cid:9)
`(cid:9)
`

`

`A controlled double blind study of azathioprine in the management of Crohn's disease
`
`677
`
`100
`
`80
`
`60
`
`0
`W..?, 40
`
`20
`
`N
`
`• Azathioprine
`+ Placebo
`
`in the general population; 70% v 30% for
`white women and 68% v 33% for women of
`mixed race. A less striking trend was found
`among male patients with prevalences of 57%
`v 41% for white, and 64% v 50% for men of
`mixed race.
`
`_ _L (cid:9)
`I (cid:9)
`I (cid:9)
`1 (cid:9)
`(cid:9)1 (cid:9) IIIII
`I
`I (cid:9)
`l (cid:9)
`1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
`Duration of trial (months)
`
`Life table representation.
`
`Leucocyte count
`The change in median white cell count was
`significantly different between the two groups.
`Median leucocyte count actually rose in the
`placebo group.
`In those patients randomised to azathioprine,
`the white cell count at the final visit was signifi-
`cantly lower in those who responded — that is,
`those in remission at trial end — than those who
`did not (p=0.005) (Table V). Nine of 13
`patients who developed leucopenia sufficient to
`warrant a decrease in drug dose v five of 20 who
`did not, were still in remission at the end of the
`trial (OR=6.75 (95% CI 1.15 to 42.7)).
`
`Acute phase reactants
`The reduction in median values for ESR and
`CRP were significantly different between the
`two groups. For technical reasons final visit
`orosomucoid concentrations were performed
`in only 45 patients. The median drop in this
`index of disease activity was 0.58 in the
`azathioprine group v 0.28 in those receiving
`placebo. This difference did not reach statisti-
`cal significance.
`
`CDAI
`As a result of the
`median differences
`cant between the
`groups (p=0.06).
`
`wide range in values, the
`in CDAI were not signifi-
`azathioprine and placebo
`
`Smoking
`To ascertain whether the rates of smoking were
`consistent with the published data for South
`Africa,3 the group was analysed by sex, popula-
`tion group, and smoking status. The preva-
`lence of smoking among female patients in this
`trial was approximately double that expected
`
`TABLF, iv
`subjects)
`
`Change from baseline to trial end (median, interquartile range, and number of
`
`Treatment gmttp
`
`Subject (n) AzoilOoprinc
`
`Subject (n) Placebo
`
`30 (cid:9)
`30 (cid:9)
`30 (cid:9)
`22 (cid:9)
`30 (cid:9)
`30 (cid:9)
`
`50 (-8-222)
`-6.5 (-17-9)
`0(-I0-17)
`0.28 (0.14-0.70)
`(-2.8-3.2)
`(-
`
`v0.06
`p=0.0005
`p=0.003
`p=0.15
`p=0.02
`p-0.9
`
`Cl)AI
`Rsik
`CRp
`()1,e rtiocmomucoid
`. ncYte
`"'eight
`
`32
`191.5 (15.5 256.5)
`32
`1551 II)
`31
`30 (0 5 3)
`21
`058 (0.18- 1.06)
`12
`2.8 (045 4.85)
`31
`( 10.2 0)
`negative value ,:hows an increase over vial period.
`
`A
`
`Discussion
`In this placebo controlled study of 63 patients
`with active Crohn's disease, azathioprine
`offered a clinically significant advantage over
`placebo in the maintenance of remission
`induced by a three month course of diminish-
`ing dose prednisolone. No significant adjunc-
`tive role could be shown for this drug when
`used with prednisolone in the induction of
`remission over a period of 12 weeks. A signifi-
`cantly greater decrease in the ESR was noted
`in those receiving combination therapy, how-
`ever.
`Previous controlled trials of immuno-
`suppression in Crohn's disease have shown
`conflicting results.4-11 Small numbers and dif-
`fering methodologies make comparison of
`these studies difficult, though Pearson, in a
`recent meta-analysis, has shown a significant
`role for azathioprine and 6-mercaptopurine in
`both active and maintenance therapy.12 In a
`comparatively large randomised withdrawal
`study, O'Donoghue and colleagues found
`azathioprine at a dose of 2 mg/kg/day to be
`effective in the maintenance of remission.8
`This early study has been criticised for having
`examined a subgroup of patients who had
`responded to the drug before withdrawal. The
`largest and perhaps most vigorous of these
`studies, the National Cooperative Crohn's
`Disease Study failed to show a significant role
`for azathioprine in either acute management
`(part 1) or as maintenance therapy (part 2).9
`Their negative findings may be explained by
`the fact that during both parts of this trial,
`azathioprine was used as single drug therapy
`and mostly at the low dose of 1 mg/kg.
`Interestingly, multivariate analysis of the
`maintenance trial, despite the overall negative
`findings, established that patients treated with
`corticosteroids before entry into the study,
`maintained remission for significantly longer
`than those who were not.
`The largest controlled trial to find a positive
`role for immunosuppression in Crohn's disease
`was that of Present et al, which was published
`shortly after the findings of the NCCDS in
`1980.10 In this two year double blind crossover
`study of 83 patients, the authors showed
`6-mercaptopurine (the active metabolite of
`azathioprine) to be significantly superior to
`placebo. At a mean starting dose of 1 mg/kg,
`patients receiving 6-mercaptopurine had a 67%
`overall improvement in symptoms compared
`with 8% receiving placebo. Their study showed
`convincing support for 6-mercaptopurine in
`the treatment of fistulas and as a corticosteroid
`sparing agent.
`Ewe and colleagues who have published
`data to support the use of a combination
`regimen of azathioprine at a dose of 2.5 mg/kg
`and prednisolone in active Crohn's disease,
`
`Ilimmumummimmounimmiimiimil
`
`rhismaterialwascopie-d
`
`Ex. 1070 - Page 6
`
`

`

`678
`
`Candy, Wright, Gerber, Adams, Gerig, Goodman
`
`TABLE v Median white cell count and interquartile range in subjects randomised to
`azathioprine. A comparison of those who completed the trial in remission (responders) with
`those who did not (non-responders)
`
`Subjects (n) (cid:9)
`
`Baseline
`
`Final visit (cid:9)
`
`Difference
`
`Responders
`Non-responders
`
`14
`18
`
`9.3 (7.0-13.0)
`8.5 (5-6-10.7)
`
`4.9 (3.9-5,7)
`6.8 (5.1-9.0)
`
`4.3 (2.0 (cid:9) 5.9)
`1•5 ( -0.9-3.0)*
`
`*p=0.005. The number of patients analysed was 32 not 33 as one was withdrawn on day 5
`without final visit data.
`
`concluded that patients went into remission
`more often and earlier and stayed in remission
`longer on the combined regimen.' I We were
`unable to show this adjunctive role in phase 1
`of our study. There are two possible explana-
`tions. Firstly, the protocol in Ewe's study
`allowed for the dose of prednisolone to be
`increased in the event of deterioration in symp-
`toms on two occasions. In our study a CDAI
`score of more than 150 in phase 1 resulted in
`immediate withdrawal. Secondly, their trial
`period was four months in total compared with
`the three months of our phase 1. Assuming
`that azathioprine may take up to four months
`to become effective,13 a possible subgroup of
`late responders would have been lost in our
`study. Examination of the life table graph
`shows divergence of the group's relapse rates at
`approximately four months. It is interesting to
`note, however, that the median ESR was
`significantly lower at 12 weeks in those
`randomised to azathioprine. Though not
`statistically significant, there was a similar
`trend for CRP.
`The rationale for not using azathioprine and
`6-mercaptopurine as first line therapy in
`Crohn's disease has been the potentially
`serious side effects. Clinical pancreatitis,
`reportedly the commonest short-term compli-
`cation (3-7%)I 4 was not encountered in our
`study, though routine amylase and lipase tests
`were not performed. A single patient devel-
`oped epigastric pain sufficient to warrant with-
`drawal from the study. Serum and urinary
`amylase values taken at the time of withdrawal
`were normal. The pain resolved spontaneously
`and at completion of trial analysis, the patient
`was found to have been randomised to
`placebo. Bone marrow suppression sufficient
`to warrant withdrawal from the study (see
`criteria in methods) was not encountered, but
`13 of 33 patients in the azathioprine group did
`require a decrease in their drug dose as a result
`of a fall in leucocyte count below 4 X109/1. (All
`counts increased after diminution of dose.)
`Leucopenia did not occur during the combina-
`tion prednisolone/azathioprine period and,
`where it did occur thereafter, there was no
`consistent time interval in which it developed.
`Despite small numbers, there was a highly
`significant negative correlation between leuco-
`cyte count and therapeutic outcome within
`the azathioprine group. Colonna
`et al in a
`retrospective study found a strong positive
`correlation between extent of drug induced
`leucopenia and clinical success of immuno-
`suppressive therapy,15 Our study confirms this
`finding.
`
`Under the influence of the beneficial find-
`ings of Korelitz et (4416 im munosuppression has
`
`Th is material was copied
`at the NLM and may be
`ht Laws
`
`been used extensively in fistulous disease. As
`radiological studies were not performed at
`entry and completion in this trial, success of
`therapy could not be ascertained for entero-
`enteral fistulas. In the two patients with recto-
`vaginal fistulas receiving active medication,
`drainage diminished entirely during the trial
`but this may reflect firming of the stool rather
`than fistulous healing. The only patient with
`extensive perianal Crohn's randomised . to
`
`
`azathioprinedrawnbet.( )re d1i2d wneoeticsr.espond and was with-
`
`The association between smoking and
`Crohn's disease has been established epidemi-
`-
`ologically." When analysed by sex and popula
`tion group, the prevalence of smoking in our
`overall sample was higher than in the general
`populational Multivariate analysis, however,
`prine. to show smoking status as an effect
`-
`azathio-
`modifier in determining response to azathio
`
`The results of this trial add weight to the
`-
`accumulating evidence tor the use of a combi
`nation of azathioprine and short-term corticos-
`teroids in active Crohn's disease. Further, our
`data suggest that azathioprine at a starting dose
`5 mg/kg/day is safe and effective in the
`of 2.
`longterm maintenance of remission induced by
`such a combination regimen.
`illienitri)rwaiiitial nitillepslitti.citeit:tailibalreltaslywsiesr.e kindly supplied bY .Well• °919s
`SA I,ty I Ad. We are grateful to Doctor Rodney Ehrlich for hi-
`
`1 Salomon I', Kornbluth A, Ai ,:•nberg J, janowitz HD. flow
`, (cid:9) 1,,"1",,l'', , I . 1( /i:ii t ri r,..iivi,il 1992; 14: 211 5.
`I''''Illi:11..1t,'1•,:-1.1:ii. oi l' i'rjW(;r,flKIlleln'sidlif.e'll)seev.ell\opliMicetna-1
`:::/.'x:;i::'.7:0;;;',';,,i;,-
`. (cid:9)
`Gitstrocnterub0
`;1 1 ,i , ,liii'. I )1,-t - Activity Index.
`•,1 (cid:9)
`litiiiii,yi r:,..yl?,:,S,,aloojee Y. Si,noking (in Soudthi,A912.ri;e10::
`1) wi:,iii:::::ilit:i.li.;;..1,1ii,111:.1,,..,r.,,,1111:,,,,,i,i .: ,11.,,;i;;..,,, i,1 11,:iiii,cti .: rek,eitileoij, .,i:)narwn son Am.
`
`I Y,2;,-11111,12.111),i.,(1,N1,111
`
`I../ (cid:9)
`
`:0,.1:;iiiii , ,t, I ), 1,,,.cl, 1), f
`
`.,::::::11111:',%(dIti5metarsocIleldatfricia411
`
`i(p)7iri ;o fi,,Ii•:
`i tr'hj''' . is,t111,cr'‘';'r: i nis:RI , S
`(..1 :i.,1/1,1',,)1,,,,,,),I,./i):,,/, ,,III,•,,,,/,),!,,,i!.18, ii;1,, if )ii i(,::(i.or:)15(iatiwi,.;is,,,,,,,r,t,nInts:e‘tici,:r(ted,ow:viiiwii,c).7iili...1a:t11.66unco:kp(,r)iiinloco:vin,...clui(11.:i, 6 (cid:9) Kl..'slin ''/ 111, i'l
`l):11,1::: I Ii j", M itchell
`
`
`8., (cid:9)
`
`
`I
`tsrleeZ /)\Ocr9117t5“;) 7'6:
`.'ncI lilji::11 ;:./iii iii)),I):).1-') i)i)_. li''); (IX:1;41)11111A1'sj:dKiseiri.s
`
`);aix:ralhtnri,:l d(i)sf ci::se-.,
`
`N LIII,:_i': :: :1; ii),,,i,.)(1,-,1 p:,
`ljj,f)out )Ict-rbelaitnnlewntitlfl(d
`maintenance
`
`. i Hest
`,,.,
`) met) RW, Switi IM4, Sessions j" I' Jr, 13ecktelj.Y ,
`National Cooperative Crdr.,,-
`9 (cid:9)
`II I
`. _ (cid:9) W1