- Email: [email protected]
Infliximab Plus Azathioprine for Steroid-Dependent Crohn’s Disease Patients: A Randomized Placebo-Controlled Trial
GASTROENTEROLOGY 2006;130:1054 –1061
Infliximab Plus Azathioprine for Steroid-Dependent Crohn’s Disease Patients: A Randomized Placebo-Controlled Trial MARC LÉMANN,* JEAN–YVES MARY,‡ BERNARD DUCLOS,§ MICHEL VEYRAC,¶ JEAN–LOUIS DUPAS,储 JEAN CHARLES DELCHIER,# DAVID LAHARIE,** JACQUES MOREAU,‡‡ GUILLAUME CADIOT,§§ LAURENCE PICON,¶¶ ARNAUD BOURREILLE,储 储 IRADJ SOBAHNI,## JEAN–FREDERIC COLOMBEL,*** and the Groupe d’Etude Therapeutique des Affections Inflammatoires du Tube Digestif (GETAID)* *Department of Gastroenterology, Hôpital Saint-Louis, Paris, France; ‡INSERM U717, Biostatistics and Clinical Epidemiology, Hôpital SaintLouis, Université Paris, France; §Department of Gastroenterology, Hôpital Hautepierre, Strasbourg, France; ¶Department of Gastroenterology, Hôpital Saint-Eloi, Montpellier, France; 储Department of Gastroenterology, Hôpital Nord, Amiens, France; #Department of Gastroenterology, Hôpital Henri Mondor, Créteil, France; **Department of Gastroenterology, Hôpital Haut-Lévêque, Bordeaux, France; ‡‡Department of Gastroenterology, Hôpital Rangueil, Toulouse, France; §§Department of Gastroenterology, Hôpital Robert Debré, Reims, France; ¶¶Department of Gastroenterology, Hôpital Trousseau, Tours, France; 储 储Department of Gastroenterology, Hôtel Dieu, Nantes, France; ##Department of Gastroenterology, Hôpital Bichât, Paris, France; ***Department of Gastroenterology, Hôpital Claude Huriez, Lille, France
See editorial on page 1354; CME Quiz on page 1348. Background & Aims: The aim of this study was to evaluate the usefulness of short-term infliximab combined with azathioprine (AZA) or 6-mercaptopurine (6-MP) in steroid-dependent Crohn’s disease patients. Methods: Patients with active disease despite prednisone given for more than 6 months were eligible and were stratified as follows: the failure stratum consisted of patients receiving AZA/6-MP at a stable dose for more than 6 months, and the naive stratum consisted of patients not treated previously with AZA/ 6-MP. Patients were randomized to infliximab 5 mg/kg or placebo at weeks 0, 2, and 6. All patients were treated with AZA/6-MP maintained at a stable dose throughout the 52 weeks of the study. The primary end point was remission off steroids at week 24. Results: Among the 113 enrolled patients (55 in the failure stratum), 57 were assigned to infliximab. At week 24, the success rate (intent-to-treat analysis) was higher in the infliximab group than in the placebo group (57% vs 29%; P ⴝ .003); at weeks 12 and 52, the corresponding rates were 75% vs 38% (P < .001) and 40% vs 22% (P ⴝ .04), respectively. In each stratum, the success rate was significantly higher in the infliximab group at weeks 12 and 24, and a trend was found at week 52. In the failure stratum, only 27% of the patients in the infliximab group were still in remission off steroids, compared with 52% in the naive stratum. Steroid resistance was less common and the cumulative dose of prednisone was lower in the infliximab group. Conclusions: Infliximab plus AZA/6-MP is more effective than AZA/6-MP alone in steroid-dependent Crohn’s disease patients.
orticosteroids are the first-line conventional therapy for patients with active Crohn’s disease (CD) of moderate to severe activity. Their efficacy, defined as achievement of a complete clinical remission, ranges from 48% to 92%.1,2 This therapeutic benefit is counterbalanced by side effects attributable to systemic action and to inhibition of endogenous adrenal function. In addition, 20%– 40% of patients develop dependence on corticosteroids and are exposed to prolonged treatment.1,2 Controlled trials,3–5 a meta-analysis,6 and a Cochrane review7 have shown that azathioprine (AZA) and 6-mercaptopurine (6-MP) are effective in CD both for maintaining remission and as steroid-sparing agents in chronic active steroid-dependent patients. A major problem with AZA/6-MP is its delay of action, which has been estimated to be a median of 3 months.4 Consequently, it often is necessary to maintain steroids for several months, exposing patients to their side effects. Infliximab, the chimeric monoclonal IgG1 antibody to tumor necrosis factor (Remicade; Centocor, Malvern, PA) has greatly improved our management of CD. The rapid efficacy of infliximab in inducing remission in patients with active CD was shown in placebo-controlled trials.8,9 We hypothesized that infliximab may be useful for steroid-dependent CD patients in combination with a slowacting immunosuppressant (AZA/6-MP) to reduce steroid exposure and to improve long-term remission rate.
C
Abbreviations used in this paper: AZA, azathioprine; CDEIS, Crohn’s Disease Endoscopic Index of Severity; CI, confidence interval; OR, odds ratio; 6-MP, 6-mercaptopurine. © 2006 by the American Gastroenterological Association Institute 0016-5085/06/$32.00 doi:10.1053/j.gastro.2006.02.014
April 2006
The aim of the present study thus was to evaluate the usefulness of infliximab (3 infusions) combined with AZA/6-MP compared with AZA/6-MP alone for achieving clinical remission off steroids in steroid-dependent CD patients.
Materials and Methods Patients Eligible patients were at least 18 years old and had luminal steroid-dependent CD. The diagnosis of CD was based on established clinical, endoscopic, radiologic, and histologic criteria. Steroid dependency was defined as follows: (1) prednisone had to be given for at least 6 months at a dosage 10 mg/day or more, with no interruption for more than 2 months within the past 6 months; (2) at least 2 clinical luminal relapses when tapering of steroids had been attempted, leading to an increase in the dose to more than 10 mg/day; and (3) the last attempt for steroid tapering had to be within the past 6 months. At baseline, patients had to be treated with prednisone 10 mg/day or more. Regarding AZA/6-MP status at baseline, 2 types of patients could be included: those in the naive stratum who did not receive AZA/6-MP in the past 2 years, and those in the failure stratum who still had clinically active disease (Crohn’s Disease Activity Index [CDAI] ⬎ 150) despite receiving AZA/6-MP for more than 6 months at a stable and appropriate dose (2–3 mg/kg/day for AZA and 1–1.5 mg/kg/day for 6-MP). Men and women with reproductive potential were required to practice birth control throughout the study and for 6 months after the last infliximab infusion. The exclusion criteria were as follows: (1) contraindication to AZA/6-MP or to infliximab according to labeling recommendations; (2) treatment with an immunosuppressive drug other than AZA/6-MP in the past 6 months; (3) previous use of infliximab or other anti–tumor necrosis factor drugs including thalidomide; (4) concomitant treatment with aminosalicylates, budesonide, topical steroids, or artificial nutrition; or (5) presence of at least 1 of the following conditions: symptomatic stricture, intra-abdominal abscess or infection, severe sepsis within the past 3 months, tuberculosis (because the bacillus Calmette-Gue´rin vaccination still is recommended in France, patients with a tuberculin skin test ⬎10 mm and a bacillus Calmette-Gue´rin vaccination performed ⬎10 years before the tuberculin skin test were excluded), history of B or C hepatitis, human immunodeficiency virus infection, liver failure, pregnancy, breast-feeding, or participation in pharmaceutical research within the past 3 months.
Study Design This randomized, multicenter, double-blind, placebocontrolled trial was performed at 22 sites in France; all physicians were members of the Groupe d’Etude Thérapeutique des Affections Inflammatoires du tube Digestif (GETAID). The recruitment of patients took place from June 2000 to May 2002.
INFLIXIMAB AND AZATHIOPRINE
1055
When the study was designed, we hypothesized that results could be different depending on whether or not patients received AZA/6-MP at inclusion. Patients thus were randomized per stratum (AZA/6-MP failure or naive) and per center to receive either infliximab or placebo in a 1:1 ratio. The randomization was performed centrally, using permutation tables of size 2 or 4 according to the number of patients that we anticipated would be enrolled at each center per stratum. The protocol was approved by the ethical committee (Comite´ Consultatif de Protection des Personnes dans La Recherche Biome´dicale Paris Saint-Louis) for each participating center. Written informed consent was obtained from all patients before enrollment into the study.
Treatments Infliximab/placebo. According to the randomization, patients received at 0, 2, and 6 weeks either a 2-hour infusion of infliximab (Remicade) administered at a dose of 5 mg/kg or an infusion of an identical placebo. Neither the patients nor the study investigators were aware of the treatment assigned. Azathioprine/6-MP. All patients were treated with AZA (2–3 mg/kg per day) or 6-MP (1–1.5 mg/kg per day). Patients previously treated with AZA or 6-MP (failure stratum) continued their treatment at the same dose; in the naive-stratum patients, AZA (2–2.5 mg/kg per day) was started 1 week after the first infliximab infusion (to differentiate side effects related to infliximab from those related to AZA). The AZA or 6-MP dose had to be maintained at a stable dose throughout the study, except for patients who experienced toxicity related to the drug. If leukopenia (⬍1500 neutrophils/mL) or mild transaminase increases occurred (⬍2 times the upper limit of normal value), the AZA/6-MP dose was decreased by 50%. Steroids. At baseline, all patients were treated with prednisone or prednisolone at a daily dose of 10 mg or more. In patients with active disease (CDAI, ⬎150), the steroid daily dose was increased by 15 mg. After 2 weeks, in patients with a clinical remission (CDAI, ⱕ150), steroids were tapered according to a standardized scheme (10 mg/wk to a daily dose of 20 mg, and then 5 mg/wk). If the CDAI still was more than 150 after 2 weeks, the prednisone daily dose could be increased to 40 mg/day or 1 mg/kg of body weight/day. Throughout the study, in patients who experienced a relapse, steroids were reintroduced or increased until a new remission was achieved, and then tapered according to the same scheme. Patients who did not reach a clinical remission after 2 weeks with 40 mg/day or more were considered to be steroid resistant and were classified as treatment failures; they were offered infliximab infusion in an open-label fashion. Other treatments. Concomitant treatment that has been shown to facilitate steroid withdrawal in steroid-dependent patients such as mesalamine, budesonide, artificial nutrition, or other immunosuppressive drugs were not allowed during the study.
1056
LEMANN ET AL
Data Collection A complete medical history and current medications were recorded at study inclusion. Physical examination and laboratory tests (C-reactive protein, erythrocyte sedimentation rate, blood counts, liver tests) were performed at screening and at each visit (weeks 0, 2, 6, 12, 24, and 52). The CDAI,10 a steroid side-effect score, and the cumulative doses of steroids also were calculated at each visit. A chest radiograph and tuberculin test were performed at inclusion and at week 24. A colonoscopy was performed, if the patient agreed, at baseline and at week 24, and the Crohn’s Disease Endoscopic Index of Severity (CDEIS) was calculated.11 Adverse events were recorded at each visit and were classified as mild, moderate, or severe; the relationship to study medications also was evaluated. The CDAI10 was calculated by using patient reports of symptoms on diary cards within the week preceding each visit. The CDEIS was assessed according to the presence and the extent of lesions and ulcerations (deep or superficial), and according to the presence of ulcerated and nonulcerated strictures on the 5 following segments: rectum, sigmoid and left colon, transverse colon, right colon, and ileum.11 The steroid side-effects score was calculated as follows: 1 point was attributed for each minor to moderate side effect including acne, swelling of the face, buffalo hump, striae, moderate increase in abdominal fat, insomnia or minor psychic disturbances, and other minor side effects; and 2 points for each severe side effect including amyotrophy and/or muscular weakness, pronounced cushingoid obesity, severe acne, diabetes mellitus, cataract, symptomatic bone complications, severe psychic disturbances, high blood pressure, or other severe side effects; the score could vary from 0 to 16.
End Points All primary and secondary end points were prespecified. The primary efficacy end point was the rate of success, defined as a clinical remission (CDAI ⬍150)12 off steroids at week 24. Secondary efficacy end points were as follows: (1) success rate at week 12, (2) rate of steroid resistance, (3) cumulative dose of prednisone at week 24, (4) steroids sideeffect score at weeks 6, 12, and 24, (5) endoscopic improvement between inclusion and week 24, and (6) adverse events. A follow-up evaluation also was planned at week 52, with assessment of the success rate. During the follow-up period (weeks 24 –52), all treatments could be given as necessary; patients who had a relapse or received additional infusions of infliximab were classified as treatment failures.
Statistical Analysis The sample size calculation was based on the assumption that infliximab would be better than placebo (1-sided basis). We estimated that 128 patients were needed to detect a crude difference of 20% for the primary end point between placebo and infliximab, assuming a remission rate of 60% with the placebo, with a power of 80%. The analysis of efficacy was performed according to the intent-to-treat principle. All pa-
GASTROENTEROLOGY Vol. 130, No. 4
tients enrolled and not lost to follow-up evaluation were included in the efficacy analysis, except if the CDAI was missing. The analysis of safety was conducted including all patients who had received at least 1 infusion of study medication. Categoric variables were described using frequencies and percentages and their distributions were compared between treatment groups, globally, and per stratum by using the 2 test or the Fisher exact test. Continuous variables were summarized using the frequency, median, and interquartile range and their distributions were compared between treatment groups using the Mann–Whitney test. Success rates (the primary end point) were compared using the 2 test (globally and per stratum); the multiple logistic regression method also was used to estimate the odds ratio (OR) of success with 95% confidence intervals (CIs) and to test if the efficacy of infliximab compared with placebo was different across strata (interaction between treatment and stratum). Success at 12 or 52 weeks and steroid resistance were analyzed similar to the primary end point. The cumulative dose of prednisone, steroid side-effect score, and decrease in the CDEIS were analyzed using the Mann–Whitney test. Multivariate logistic regression analysis, adjusted on treatment and stratum, with backward selection using the likelihood ratio test, was used to look for factors measured at baseline that were independently predictive of success at week 24. Variables were proposed to this analysis if the P value was less than .30 in the univariate analysis. Continuous factors were categorized as follows: each variable first was divided into 4 categories at approximately the 25th, 50th, and 75th percentiles. If the ORs of adjacent categories were not statistically different, these categories were grouped. If no clear pattern was observed, the median was taken as a cut-off point. Usual limits, such as 150 for the CDAI, also were tested. The treatment effect and treatment-strata interaction were tested after adjustment on independent prognostic factors. Results were expressed as ORs with 95% CIs. Results were considered significant when the P value was less than .05. Data were analyzed with SPSS Software for Windows, release 6.1 (SPSS Inc, Chicago, IL).
Results Baseline Characteristics of Patients The trial profile is shown in Figure 1. A total of 115 patients were randomized (56 in the failure stratum/59 in the naive stratum); 57 patients (26 in the failure stratum) were assigned randomly to receive infliximab and 58 patients (30 in the failure stratum) to receive placebo. Two patients allocated to placebo treatment (1 in each stratum) were not enrolled and did not receive the first infusion, 1 patient because his CDAI was less than 150 at inclusion and the other patient because he withdrew his consent before the inclusion visit.
April 2006
INFLIXIMAB AND AZATHIOPRINE
1057
Figure 1. Flow chart of patients randomized in the trial. The figures in parentheses correspond to the number of patients in the naive stratum and the failure stratum, respectively.
Characteristics of the 113 enrolled patients are shown in Table 1. Comparability of the 2 treatment groups was verified for all clinical characteristics, except for the percentage of patients with active perianal lesions, the CDAI level in failure stratum, and the duration of steroid treatments in the naive stratum. Efficacy The numbers of patients available for efficacy evaluation at weeks 12, 24, and 52 are shown in Figure 1. At week 24, the percentage of success (CDAI ⬍150 and off steroids) was significantly higher in the infliximab group than in the placebo group (57% vs 29%; OR, 3.3; 95% CI, 1.5–7.4; P ⫽ .003); the corresponding rates were 75% vs 38% (OR, 4.9; 95% CI, 2.2–11.0; P ⬍ .001) at week 12, and 40% vs 22% (OR, 2.4; 95% CI, 1.0 –5.7; P ⫽ .04) at week 52 (Figure 2). There was no significant interaction between treatment and stratum (P ⫽ .10; .32, and .82 at weeks 12, 24, and 52, respectively). The comparison of infliximab and placebo in each stratum is indicated in Figure 3. In the naive stratum, at week 24, the percentage of success
was significantly higher in the infliximab group than in the placebo group (63% vs 32%; OR, 3.7; 95% CI, 1.1–11.3; P ⫽ .02); the corresponding rates were 83% vs 41% (OR, 7.1; 95% CI, 2.1–24.9; P ⫽ .009) at week 12, and 52% vs 32%; (OR, 2.3; 95% CI, .7– 6.9; P ⫽ .14) at week 52. In the failure stratum, the success rate also was higher in the infliximab group than in the placebo group at week 24 (50% vs 26%; OR, 2.9; 95% CI, .9 –9.3; P ⫽ .08) and at week 12 (64% vs 34%; OR, 3.4; 95% CI, 1.1–10.3; P ⫽ .03); a trend was found at week 52 (27% vs 12%; OR, 2.8; 95% CI, .6 –12.4; P ⫽ .16). These results must be considered cautiously because of the absence of demonstrable interaction. Steroid resistance was less common in the infliximab group than in the placebo group (5% vs 23%; OR, 5.1; 95% CI, 1.3–19.2; P ⫽ .01). The median cumulative dose (interquartile range) of prednisone was lower in the infliximab group at 1110 mg/24 wk (630 –1720 mg/24 wk) vs 1870 mg/24 wk (1110 – 2710 mg/24 wk) (P ⫽ .002). The median (interquartile range) of the side-effect steroid score, however,
1058
LEMANN ET AL
GASTROENTEROLOGY Vol. 130, No. 4
Table 1. Baseline Characteristics of the Patients Enrolled in the Trial According to Study Treatment and Stratum Failure stratum Placebo n ⫽ 29 Sex (female)a Age, yb Disease duration, yb Disease locationa Small bowel Colon Both Active perianal diseasea CDAIb Duration of steroids, mob Steroid dose, mg/dayb Steroid dose, mg/kg/dayb Duration of AZA/6-MP, mob Dosage of AZA,c mg/kg/dayb Dosage of AZA,c mg/dayb CDEISb (n ⫽ 52) C-reactive protein,b mg/L
69 29 (23–33) 7 (3–11) 11 50 39 14 181 (154–259) 16 (9–36) 30 (20–40) .5 (.3–.7) 21 (10–44) 2.4 (2.1–2.6) 150 (125–150) 9 (6–15)d 8 (4–35)
Naive stratum
Infliximab n ⫽ 26 69 26 (22–37) 5 (4–10) 20 32 48 35 240 (219–281) 25 (13–36) 33 (20–46) .6 (.4–.8) 27 (10–33) 2.3 (2.0–2.5) 125 (100–150) 9 (4–14)e 19 (4–47)h
P
Placebo n ⫽ 27
Infliximab n ⫽ 31
P
.95 .52 .97
43 26 (22–36) 4 (1–8)
39 27 (22–38) 3 (1–6)
.75 .79 .60
.37
.07 .02 .20 .32 .25 .95 .27 .39 .63 .63
15 26 59 7 112 (42–262) 15 (9–52) 25 (20–40) .4 (.3–.6) 0 2.2 (2.0–2.4) 125 (100–150) 6 (3–14)f 17 (7–31)i
35 13 52 32 146 (90–244) 9 (7–22) 25 (20–40) .4 (.3–.6) 0 2.2 (2.0–2.4) 150 (125–150) 11 (5–16)g 20 (10–30)
.74
.02 .48 .03 .85 .77 .94 .20 .50 .36
of patients; 2 test. (interquartile range); Mann–Whitney U test. cAll patients were treated with AZA. dn ⫽ 17. en ⫽ 9. fn ⫽ 14. gn ⫽ 12. hn ⫽ 25. in ⫽ 26. aPercent bMedian
was not different in the 2 treatment groups at 2.0 (.0 – 4.0) vs 2.0 (1.0 – 4.0) at week 6 (P ⫽ .65), 2.0 (.0 –3.0) vs 1.0 (.0 –3.0) at week 12 (P ⫽ .91), and 1.0 (.0 –2.0) vs .5 (.0 –1.8) at week 24 (P ⫽ .42). There was no significant interaction between treatment and stratum. Factors associated with success at week 24 in the multivariate logistic regression analysis were a low CDAI at baseline (OR, 4.5; 95% CI, 1.5–13.7; P ⫽ .005), a
young age (OR, 5.9; 95% CI, 1.8 –19.6; P ⫽ .002), the absence of small-bowel involvement (OR, 4.3; 95% CI, 1.5–12.2; P ⫽ .004), and a long duration of steroids before entry (OR, 5.5; 95% CI, 1.3–22.5; P ⫽ .01). After adjusting for these factors, infliximab still was superior to placebo (OR, 5.7; 95% CI, 2.0 –15.9; P ⬍.001) and no interaction between treatment and strata could be seen (P ⫽ .63).
Figure 2. Percent of patients in clinical remission and off steroids at weeks 12, 24 (primary end point), and 52 (follow-up evaluation). The figures below the bars correspond to the number of patients analyzed at each date. □, AZA/6-MP ⫹ placebo; , AZA/6-MP ⫹ infliximab 5 mg/kg.
April 2006
INFLIXIMAB AND AZATHIOPRINE
1059
Figure 3. Percent of patients in clinical remission and off steroids in the 2 strata at weeks 12, 24 (primary end point), and 52 (follow-up evaluation). The figures below the bars correspond to the number of patients analyzed at each date. □, AZA/ 6-MP ⫹ placebo; , AZA/6-MP ⫹ infliximab 5 mg/kg.
Endoscopy Colonoscopy was performed at baseline in 52 patients (21 in the infliximab group and 31 in the placebo group); 20 of them had a second colonoscopy at week 24 (11 in the infliximab group and 9 in the placebo group). CDEIS decrease (median [interquartile range]) from baseline to week 24 was 6.9 (4.1–9.5) in the infliximab group and 1.2 (⫺1.5 to 4.4) in the placebo group (P ⫽ .05). The number of patients with no ulceration at week 24 was 3 of 11 (27%) and 3 of 9 (33%) in the infliximab and placebo groups, respectively (P ⫽ .77). Among the 16 patients in remission and off steroids at week 24 (9 in the infliximab group and 7 in the placebo group), the CDEIS decrease (median [interquartile range]) was higher in the infliximab group than in the placebo group, 6.9 (1.2–10.2) vs 1.2 (.0 – 4.0) (P ⫽ .05); the CDEIS decrease was more than 5 in 6 of the 9 patients in the infliximab group and in 0 of the 7 patients in the placebo group (P ⫽ .01). The number of patients with no ulceration at week 24 was 3 of 9 (33%) and 3 of 7 (43%) in the infliximab and placebo groups, respectively (P ⫽ .70). Safety The frequency and severity of adverse events were not different between the 2 treatment groups (Table 2). The percent of patients who had at least 1 adverse event was 51% (29 of 57) in the infliximab group and 50% (28 of 56) in the placebo group. The frequency of infection was similar in the 2 treatment groups. Of note, 5 serious adverse events were probably or possibly related to azathioprine (Table 2). One patient in the infliximab group
had a severe reaction after the second and third infusions (2%). No malignancy and no deaths occurred among the study patients.
Discussion Our study was designed to show the superiority of treatment with a combination of AZA/6-MP with infliximab (3 infusions at 0, 2, and 6 weeks), compared with AZA/6-MP alone, in patients with steroid-dependent CD. With the addition of infliximab to AZA/6-MP we showed the following: (1) the rate of success was nearly doubled, defined as complete clinical remission (CDAI, ⬍150) without steroids from 38% to 75% at week 12, from 29% to 57% at week 24, and from 22% to 40% at 52 weeks; the corresponding odds of success were increased by a factor of about 5, 3.5, and 2.5; and (2) it reduces exposure to steroids. Although we found that infliximab was associated with lower steroid consumption, steroid side effects were not decreased significantly by infliximab. Infliximab thus could be used as a bridge, waiting for the delayed effect of AZA/6-MP. An important point to consider in clinical practice, however, is whether or not the bridge effect is holding with time. Two different situations could be seen here. In patients with previous failure to AZA/6-MP, a loss of efficacy was observed gradually, and at week 52 only 27% of the patients were still in remission off steroids in the infliximab group. In such patients, the bridge strategy thus is questionable. The majority of patients probably would benefit from optimization of AZA/6-MP treatment using higher doses adjusted to 6-thioguanine nucleotide levels (which was not made systematically in our study), or from the switch
1060
LEMANN ET AL
GASTROENTEROLOGY Vol. 130, No. 4
Table 2. Adverse Events Occurring During the Study Period: Day 0 to Week 24
Total Mild Moderate Severe Infections Upper respiratory tract Dental Otitis Cutaneous Herpes virus Urinary tract Intestinal Perianal abscess Pelvic abscess Arthralgia, myalgia Abdominal pain, diarrhea Nausea or vomiting Headache Cutaneous rash, pruritus Fatigue Reaction to infusion Fever Mild increase of transaminase levels Pancreatitis Miscellaneous
Number of events in the infliximab group
Number of events in the placebo group
79 39 36 4a 18 8 2 0 3 2 1 1 0 1d 8 9 10e 6 5 5 2f 1 1
66 27 30 9a 16 11b 1 1 1 0 1 0 1c 0 13e 5 3 4 7e 2 0 4 0
0e 14
2e 10
adverse event in 5 cases (infliximab, n ⫽ 3; placebo, n ⫽ 3). patient in the placebo group also had a severe pneumonitis caused by Legionella pneumophila during the follow-up period, and recovered with antibiotics. cPerianal abscess occurring at week 18; surgical treatment. dPelvic abscess discovered at week 24; surgical treatment. eFive patients had adverse events possibly or probably related to azathioprine, 1 in the infliximab group (severe vomiting, week 6), and 4 in the placebo group including 1 case of arthralgia, myalgia, cough, diarrhea, and fever (week 6), 1 case of arthralgia, fever, and cutaneous rash (week 6), and 2 cases of pancreatitis (weeks 4 and 5); all the patients stopped azathioprine and recovered. One additional case of pancreatitis occurred in the infliximab group at week 23 during the follow-up period, azathioprine was stopped but the cause remained unclear because biliary stones also were found. f Two severe reactions in the same patient, 1 after the second infliximab infusion and the other after the third infusion; the patient recovered with steroids and adrenalin. aSerious bOne
to another immunosuppressant such as methotrexate or to maintenance treatment with infliximab. In contrast, among patients naive for AZA/6-MP at inclusion, despite a similar loss of efficacy with time, more than 50% were always in remission and off steroids at 1 year. In such patients the strategy of using infliximab as a bridge between steroids and AZA/6-MP could be considered. Optimization of AZA/6-MP treatment also may have, in this setting, improved long-term results. By comparison, in the ACCENT 1 infliximab maintenance trial, responders to the initial infusion of infliximab (n ⫽ 335)
were randomized to infliximab (5 mg/kg) or placebo at weeks 2 and 6, and then to infusions of infliximab 5 mg/kg, 10 mg/kg, or placebo every 8 weeks. At week 54, the rates of remission and off steroids were 24%, 32%, and 9%, respectively.9 However, in this study only 29% of patients were treated with immunosuppressants at baseline. A potential drawback of the bridge strategy is that the interruption of infliximab after an induction scheme could facilitate the development of anti-infliximab antibody, and thus compromise a re-treatment with infliximab, if necessary.13,14 Unfortunately, we were not able to assess the anti-infliximab antibody in our trial. The risk of immunization must be weighed against the potential for increased toxicity15 and cost for patients receiving regular maintenance treatment. In the present study, the crude number and the rate of adverse events were not different in the 2 treatment groups. However, there is still some uncertainty regarding the long-term safety of maintenance infliximab including infections and the development of lymphoproliferative disorders.16,17 In our study, the rather low effectiveness of AZA/ 6-MP alone in the naive stratum for steroid-dependent patients (32% of success at 24 and 52 weeks) compared with previous studies3,4,6 was probably the effect of the stringent criterion of success that was used (ie, remission off steroids). Moreover, all the patients included in our study were steroid-dependent, although this was not always the case in previous studies. On the other hand, a remarkably high success rate (83%) was observed in the infliximab group at week 12. This may be owing to concurrent AZA/6-MP therapy. Two retrospective and 4 prospective studies that comprised a total of 738 patients with active CD reported a trend toward a higher response to infliximab in patients receiving concurrent immunosuppressants compared with those not receiving any immunosuppressants.18 A subgroup of patients had a colonoscopic evaluation during our study. Despite the small number of patients, endoscopic improvement was more pronounced in the infliximab group than in the placebo group. When the comparison was restricted to patients in clinical remission and off steroids at week 24 (n ⫽ 16), the superiority of infliximab was confirmed. Even if a selection bias cannot be ruled out, this result confirms a previous report of mucosal healing with infliximab.14 In conclusion, our study shows that the combination of infliximab with azathioprine is more effective than azathioprine alone in steroid-dependent patients to achieve remission without steroids, and to reduce exposure to steroids. In AZA/6-MP–naive patients, infliximab could be used as a bridge between steroids and
April 2006
azathioprine. If a relapse occurs, regular maintenance therapy with infliximab should be considered. One cannot exclude, however, that the continuation of regular infusions of infliximab combined with AZA/6-MP could have resulted in a better success rate in our specific study population. This point should be tested in future studies. In AZA/6-MP–treatment failure patients, a loss of effectiveness was observed gradually, and the bridge strategy cannot be recommended. Other options such as infliximab maintenance should be considered.
References 1. Modigliani R, Mary JY, Simon JF, Cortot A, Soulé JC, Gendre JP, Rene E. Clinical, biological, and endoscopic picture of attacks of Crohn’s disease. Evolution on prednisolone. Groupe d’Etude Therapeutique des Affections Inflammatoires Digestives. Gastroenterology 1990;98:811– 818. 2. Munkholm P, Langholz E, Davidsen M, Binder V. Frequency of glucocorticoid resistance and dependency in Crohn’s disease. Gut 1994;35:360 –362. 3. Candy S, Wright J, Gerber M, Adams G, Gerig M, Goodman R. A controlled double blind study of azathioprine in the management of Crohn’s disease. Gut 1995;37:674 – 678. 4. Present DH, Korelitz BI, Wisch N, Glass JL, Sachar DB, Pasternack BS. Treatment of Crohn’s disease with 6-mercaptopurine. A long-term, randomized, double-blind study. N Engl J Med 1980; 302:981–987. 5. O’Donoghue DP, Dawson AM, Powel-Tuck K, Bown RL. Doubleblind withdrawal of azathioprine as maintenance treatment for Crohn’s disease. Lancet 1978;2:944 –946. 6. Pearson DC, May GR, Fick GH, Sutherland LR. Azathioprine and 6-mercaptopurine in Crohn’s disease. A meta-analysis. Ann Intern Med 1995;123:132–142. 7. Pearson DC, May GR, Fick G, Sutherland LR. Azathioprine for maintaining remission of Crohn’s disease. Cochrane Database Syst Rev 2000;(2):CD000067. 8. Targan SR, Hanauer SB, van Deventer SJ, Mayer L, Present DH, Braakman T, DeWoody KL, Schaible TF, Rutgeerts PJ. A shortterm study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohn’s disease. Crohn’s Disease cA2 Study Group. N Engl J Med 1997;337:1029 –1035. 9. Hanauer SB, Feagan BG, Lichtenstein GR, Mayer LF, Schreiber S, Colombel JF, Rachmilewitz D, Wolf DC, Olson A, Bao W, Rutgeerts P. Maintenance infliximab for Crohn’s disease: the ACCENT I randomised trial. Lancet 2002;359:1541–1549. 10. Best WR, Becktel JM, Singleton JW, Kern F. Development of a Crohn’s disease activity index. National Cooperative Crohn’s Disease Study. Gastroenterology 1976;70:439 – 444. 11. Mary JY, Modigliani R. Development and validation of an endoscopic index of the severity for Crohn’s disease: a prospective multicentre study. Groupes d’ Etudes Thérapeutiques des Affections Inflammatoires du Tube Digestif (GETAID). Gut 1989;30:983–989.
INFLIXIMAB AND AZATHIOPRINE
1061
12. Sandborn WJ, Feagan BG, Hanauer SB, Lochs H, Lofberg R, Modigliani R, Present DH, Rutgeerts P, Scholmerich J, Stange EF, Sutherland LR. A review of activity indices and efficacy endpoints for clinical trials of medical therapy in adults with Crohn’s disease. Gastroenterology 2002;122:512–530. 13. Baert F, Noman M, Vermeire S, Van Assche G, D’Haens G, Carbonez A, Rutgeerts P. Influence of immunogenicity on the long-term efficacy of infliximab in Crohn’s disease. N Engl J Med 2003;348:601– 608. 14. Rutgeerts P, Feagan BG, Lichtenstein GR, Mayer LF, Schreiber S, Colombel JF, Rachmilewitz D, Wolf DC, Olson A, Bao W, Hanauer SB. Comparison of scheduled and episodic treatment strategies of infliximab in Crohn’s disease. Gastroenterology 2004;126: 402– 413. 15. Colombel JF, Loftus EV Jr, Tremaine WJ, Egan LJ, Harmsen WS, Schleck CD, Zinsmeister AR, Sandborn WJ. The safety profile of infliximab in patients with Crohn’s disease: the Mayo Clinic Experience in 500 patients. Gastroenterology 2004;126:19 – 31. 16. Ljung T, Karlen P, Schmidt D, Hellstrom PM, Lapidus A, Janczewska I, Sjoqvist U, Lofberg R. Infliximab in inflammatory bowel disease: clinical outcome in a population based cohort from Stockholm County. Gut 2004;53:849 – 853. 17. Wolfe F, Michaud K. Lymphoma in rheumatoid arthritis: the effect of methotrexate and anti-tumor necrosis factor therapy in 18,572 patients. Arthritis Rheum 2004;50:1740 –1751. 18. Su C, Lichtenstein GR. Are there predictors of Remicade treatment success or failure? Adv Drug Deliv Rev 2005;57:237–245.
Received March 21, 2005. Accepted December 28, 2005. Address requests for reprints to: Professor Marc Lémann, MD, PhD, Service de Gastroentérologie, 1 avenue Claude Vellefaux, 75010, Paris, France. e-mail: [email protected]; fax: (33) 1-42-49-91-68. Presented in part at the Digestive Disease Week, Orlando, May 17-22, 2003, and at the United European Gastroenterology Week (1-year follow-up), Madrid, November 1-5, 2003. Supported by Groupe d’Etude Thérapeutique des Affections Inflammatoires du tube Digestif (GETAID), and by grants from Schering Plough, France, with the specific help of Gérard Trape and Yves-Dominique Henry. Drugs were provided by Schering Plough, France. All data analysis and manuscript writing were performed independently by the GETAID, without the involvement of representatives of Schering Plough. In addition to the authors, the following French investigators participated in the study: Robert Modigliani, Hôpital Saint-Louis, Paris, France; Etienne-Henry Metman, Hôpital Trousseau Tours; Jean-Paul Galmiche, Hôtel Dieu, Nantes, France; Jean-Pierre Gendre, Hôpital Saint-Antoine, Paris, France; Marc-André Bigard, Hôpital Brabois, Nancy, France; Stanislas Chaussade and Vered Abitbol, Hôpital Cochin, Paris, France; Jean-Charles Grimaud, Hôpital Nord, Marseille, France; Gilles Bommelaer, Hotel Dieu, Clermont-Ferrand, France; JeanClaude Soulé, Hôpital Bichât, Paris, France; Bruno Bonaz, Hôpital Nord, Grenoble, France; Louis Descos and Bernard Flourié, Hôpital Lyon-Sud, Lyon, France; Eric Lerebours, Hôpital Charles Nicolle, Rouen, France; Benoit Coffin, Hôpital Louis Mourier, Colombes, France. The authors acknowledge Patricia De ´tre ´, Study Coordinator, for her technical help.
Infliximab Plus Azathioprine for Steroid-Dependent Crohn’s Disease Patients: A Randomized Placebo-Controlled Trial MARC LÉMANN,* JEAN–YVES MARY,‡ BERNARD DUCLOS,§ MICHEL VEYRAC,¶ JEAN–LOUIS DUPAS,储 JEAN CHARLES DELCHIER,# DAVID LAHARIE,** JACQUES MOREAU,‡‡ GUILLAUME CADIOT,§§ LAURENCE PICON,¶¶ ARNAUD BOURREILLE,储 储 IRADJ SOBAHNI,## JEAN–FREDERIC COLOMBEL,*** and the Groupe d’Etude Therapeutique des Affections Inflammatoires du Tube Digestif (GETAID)* *Department of Gastroenterology, Hôpital Saint-Louis, Paris, France; ‡INSERM U717, Biostatistics and Clinical Epidemiology, Hôpital SaintLouis, Université Paris, France; §Department of Gastroenterology, Hôpital Hautepierre, Strasbourg, France; ¶Department of Gastroenterology, Hôpital Saint-Eloi, Montpellier, France; 储Department of Gastroenterology, Hôpital Nord, Amiens, France; #Department of Gastroenterology, Hôpital Henri Mondor, Créteil, France; **Department of Gastroenterology, Hôpital Haut-Lévêque, Bordeaux, France; ‡‡Department of Gastroenterology, Hôpital Rangueil, Toulouse, France; §§Department of Gastroenterology, Hôpital Robert Debré, Reims, France; ¶¶Department of Gastroenterology, Hôpital Trousseau, Tours, France; 储 储Department of Gastroenterology, Hôtel Dieu, Nantes, France; ##Department of Gastroenterology, Hôpital Bichât, Paris, France; ***Department of Gastroenterology, Hôpital Claude Huriez, Lille, France
See editorial on page 1354; CME Quiz on page 1348. Background & Aims: The aim of this study was to evaluate the usefulness of short-term infliximab combined with azathioprine (AZA) or 6-mercaptopurine (6-MP) in steroid-dependent Crohn’s disease patients. Methods: Patients with active disease despite prednisone given for more than 6 months were eligible and were stratified as follows: the failure stratum consisted of patients receiving AZA/6-MP at a stable dose for more than 6 months, and the naive stratum consisted of patients not treated previously with AZA/ 6-MP. Patients were randomized to infliximab 5 mg/kg or placebo at weeks 0, 2, and 6. All patients were treated with AZA/6-MP maintained at a stable dose throughout the 52 weeks of the study. The primary end point was remission off steroids at week 24. Results: Among the 113 enrolled patients (55 in the failure stratum), 57 were assigned to infliximab. At week 24, the success rate (intent-to-treat analysis) was higher in the infliximab group than in the placebo group (57% vs 29%; P ⴝ .003); at weeks 12 and 52, the corresponding rates were 75% vs 38% (P < .001) and 40% vs 22% (P ⴝ .04), respectively. In each stratum, the success rate was significantly higher in the infliximab group at weeks 12 and 24, and a trend was found at week 52. In the failure stratum, only 27% of the patients in the infliximab group were still in remission off steroids, compared with 52% in the naive stratum. Steroid resistance was less common and the cumulative dose of prednisone was lower in the infliximab group. Conclusions: Infliximab plus AZA/6-MP is more effective than AZA/6-MP alone in steroid-dependent Crohn’s disease patients.
orticosteroids are the first-line conventional therapy for patients with active Crohn’s disease (CD) of moderate to severe activity. Their efficacy, defined as achievement of a complete clinical remission, ranges from 48% to 92%.1,2 This therapeutic benefit is counterbalanced by side effects attributable to systemic action and to inhibition of endogenous adrenal function. In addition, 20%– 40% of patients develop dependence on corticosteroids and are exposed to prolonged treatment.1,2 Controlled trials,3–5 a meta-analysis,6 and a Cochrane review7 have shown that azathioprine (AZA) and 6-mercaptopurine (6-MP) are effective in CD both for maintaining remission and as steroid-sparing agents in chronic active steroid-dependent patients. A major problem with AZA/6-MP is its delay of action, which has been estimated to be a median of 3 months.4 Consequently, it often is necessary to maintain steroids for several months, exposing patients to their side effects. Infliximab, the chimeric monoclonal IgG1 antibody to tumor necrosis factor (Remicade; Centocor, Malvern, PA) has greatly improved our management of CD. The rapid efficacy of infliximab in inducing remission in patients with active CD was shown in placebo-controlled trials.8,9 We hypothesized that infliximab may be useful for steroid-dependent CD patients in combination with a slowacting immunosuppressant (AZA/6-MP) to reduce steroid exposure and to improve long-term remission rate.
C
Abbreviations used in this paper: AZA, azathioprine; CDEIS, Crohn’s Disease Endoscopic Index of Severity; CI, confidence interval; OR, odds ratio; 6-MP, 6-mercaptopurine. © 2006 by the American Gastroenterological Association Institute 0016-5085/06/$32.00 doi:10.1053/j.gastro.2006.02.014
April 2006
The aim of the present study thus was to evaluate the usefulness of infliximab (3 infusions) combined with AZA/6-MP compared with AZA/6-MP alone for achieving clinical remission off steroids in steroid-dependent CD patients.
Materials and Methods Patients Eligible patients were at least 18 years old and had luminal steroid-dependent CD. The diagnosis of CD was based on established clinical, endoscopic, radiologic, and histologic criteria. Steroid dependency was defined as follows: (1) prednisone had to be given for at least 6 months at a dosage 10 mg/day or more, with no interruption for more than 2 months within the past 6 months; (2) at least 2 clinical luminal relapses when tapering of steroids had been attempted, leading to an increase in the dose to more than 10 mg/day; and (3) the last attempt for steroid tapering had to be within the past 6 months. At baseline, patients had to be treated with prednisone 10 mg/day or more. Regarding AZA/6-MP status at baseline, 2 types of patients could be included: those in the naive stratum who did not receive AZA/6-MP in the past 2 years, and those in the failure stratum who still had clinically active disease (Crohn’s Disease Activity Index [CDAI] ⬎ 150) despite receiving AZA/6-MP for more than 6 months at a stable and appropriate dose (2–3 mg/kg/day for AZA and 1–1.5 mg/kg/day for 6-MP). Men and women with reproductive potential were required to practice birth control throughout the study and for 6 months after the last infliximab infusion. The exclusion criteria were as follows: (1) contraindication to AZA/6-MP or to infliximab according to labeling recommendations; (2) treatment with an immunosuppressive drug other than AZA/6-MP in the past 6 months; (3) previous use of infliximab or other anti–tumor necrosis factor drugs including thalidomide; (4) concomitant treatment with aminosalicylates, budesonide, topical steroids, or artificial nutrition; or (5) presence of at least 1 of the following conditions: symptomatic stricture, intra-abdominal abscess or infection, severe sepsis within the past 3 months, tuberculosis (because the bacillus Calmette-Gue´rin vaccination still is recommended in France, patients with a tuberculin skin test ⬎10 mm and a bacillus Calmette-Gue´rin vaccination performed ⬎10 years before the tuberculin skin test were excluded), history of B or C hepatitis, human immunodeficiency virus infection, liver failure, pregnancy, breast-feeding, or participation in pharmaceutical research within the past 3 months.
Study Design This randomized, multicenter, double-blind, placebocontrolled trial was performed at 22 sites in France; all physicians were members of the Groupe d’Etude Thérapeutique des Affections Inflammatoires du tube Digestif (GETAID). The recruitment of patients took place from June 2000 to May 2002.
INFLIXIMAB AND AZATHIOPRINE
1055
When the study was designed, we hypothesized that results could be different depending on whether or not patients received AZA/6-MP at inclusion. Patients thus were randomized per stratum (AZA/6-MP failure or naive) and per center to receive either infliximab or placebo in a 1:1 ratio. The randomization was performed centrally, using permutation tables of size 2 or 4 according to the number of patients that we anticipated would be enrolled at each center per stratum. The protocol was approved by the ethical committee (Comite´ Consultatif de Protection des Personnes dans La Recherche Biome´dicale Paris Saint-Louis) for each participating center. Written informed consent was obtained from all patients before enrollment into the study.
Treatments Infliximab/placebo. According to the randomization, patients received at 0, 2, and 6 weeks either a 2-hour infusion of infliximab (Remicade) administered at a dose of 5 mg/kg or an infusion of an identical placebo. Neither the patients nor the study investigators were aware of the treatment assigned. Azathioprine/6-MP. All patients were treated with AZA (2–3 mg/kg per day) or 6-MP (1–1.5 mg/kg per day). Patients previously treated with AZA or 6-MP (failure stratum) continued their treatment at the same dose; in the naive-stratum patients, AZA (2–2.5 mg/kg per day) was started 1 week after the first infliximab infusion (to differentiate side effects related to infliximab from those related to AZA). The AZA or 6-MP dose had to be maintained at a stable dose throughout the study, except for patients who experienced toxicity related to the drug. If leukopenia (⬍1500 neutrophils/mL) or mild transaminase increases occurred (⬍2 times the upper limit of normal value), the AZA/6-MP dose was decreased by 50%. Steroids. At baseline, all patients were treated with prednisone or prednisolone at a daily dose of 10 mg or more. In patients with active disease (CDAI, ⬎150), the steroid daily dose was increased by 15 mg. After 2 weeks, in patients with a clinical remission (CDAI, ⱕ150), steroids were tapered according to a standardized scheme (10 mg/wk to a daily dose of 20 mg, and then 5 mg/wk). If the CDAI still was more than 150 after 2 weeks, the prednisone daily dose could be increased to 40 mg/day or 1 mg/kg of body weight/day. Throughout the study, in patients who experienced a relapse, steroids were reintroduced or increased until a new remission was achieved, and then tapered according to the same scheme. Patients who did not reach a clinical remission after 2 weeks with 40 mg/day or more were considered to be steroid resistant and were classified as treatment failures; they were offered infliximab infusion in an open-label fashion. Other treatments. Concomitant treatment that has been shown to facilitate steroid withdrawal in steroid-dependent patients such as mesalamine, budesonide, artificial nutrition, or other immunosuppressive drugs were not allowed during the study.
1056
LEMANN ET AL
Data Collection A complete medical history and current medications were recorded at study inclusion. Physical examination and laboratory tests (C-reactive protein, erythrocyte sedimentation rate, blood counts, liver tests) were performed at screening and at each visit (weeks 0, 2, 6, 12, 24, and 52). The CDAI,10 a steroid side-effect score, and the cumulative doses of steroids also were calculated at each visit. A chest radiograph and tuberculin test were performed at inclusion and at week 24. A colonoscopy was performed, if the patient agreed, at baseline and at week 24, and the Crohn’s Disease Endoscopic Index of Severity (CDEIS) was calculated.11 Adverse events were recorded at each visit and were classified as mild, moderate, or severe; the relationship to study medications also was evaluated. The CDAI10 was calculated by using patient reports of symptoms on diary cards within the week preceding each visit. The CDEIS was assessed according to the presence and the extent of lesions and ulcerations (deep or superficial), and according to the presence of ulcerated and nonulcerated strictures on the 5 following segments: rectum, sigmoid and left colon, transverse colon, right colon, and ileum.11 The steroid side-effects score was calculated as follows: 1 point was attributed for each minor to moderate side effect including acne, swelling of the face, buffalo hump, striae, moderate increase in abdominal fat, insomnia or minor psychic disturbances, and other minor side effects; and 2 points for each severe side effect including amyotrophy and/or muscular weakness, pronounced cushingoid obesity, severe acne, diabetes mellitus, cataract, symptomatic bone complications, severe psychic disturbances, high blood pressure, or other severe side effects; the score could vary from 0 to 16.
End Points All primary and secondary end points were prespecified. The primary efficacy end point was the rate of success, defined as a clinical remission (CDAI ⬍150)12 off steroids at week 24. Secondary efficacy end points were as follows: (1) success rate at week 12, (2) rate of steroid resistance, (3) cumulative dose of prednisone at week 24, (4) steroids sideeffect score at weeks 6, 12, and 24, (5) endoscopic improvement between inclusion and week 24, and (6) adverse events. A follow-up evaluation also was planned at week 52, with assessment of the success rate. During the follow-up period (weeks 24 –52), all treatments could be given as necessary; patients who had a relapse or received additional infusions of infliximab were classified as treatment failures.
Statistical Analysis The sample size calculation was based on the assumption that infliximab would be better than placebo (1-sided basis). We estimated that 128 patients were needed to detect a crude difference of 20% for the primary end point between placebo and infliximab, assuming a remission rate of 60% with the placebo, with a power of 80%. The analysis of efficacy was performed according to the intent-to-treat principle. All pa-
GASTROENTEROLOGY Vol. 130, No. 4
tients enrolled and not lost to follow-up evaluation were included in the efficacy analysis, except if the CDAI was missing. The analysis of safety was conducted including all patients who had received at least 1 infusion of study medication. Categoric variables were described using frequencies and percentages and their distributions were compared between treatment groups, globally, and per stratum by using the 2 test or the Fisher exact test. Continuous variables were summarized using the frequency, median, and interquartile range and their distributions were compared between treatment groups using the Mann–Whitney test. Success rates (the primary end point) were compared using the 2 test (globally and per stratum); the multiple logistic regression method also was used to estimate the odds ratio (OR) of success with 95% confidence intervals (CIs) and to test if the efficacy of infliximab compared with placebo was different across strata (interaction between treatment and stratum). Success at 12 or 52 weeks and steroid resistance were analyzed similar to the primary end point. The cumulative dose of prednisone, steroid side-effect score, and decrease in the CDEIS were analyzed using the Mann–Whitney test. Multivariate logistic regression analysis, adjusted on treatment and stratum, with backward selection using the likelihood ratio test, was used to look for factors measured at baseline that were independently predictive of success at week 24. Variables were proposed to this analysis if the P value was less than .30 in the univariate analysis. Continuous factors were categorized as follows: each variable first was divided into 4 categories at approximately the 25th, 50th, and 75th percentiles. If the ORs of adjacent categories were not statistically different, these categories were grouped. If no clear pattern was observed, the median was taken as a cut-off point. Usual limits, such as 150 for the CDAI, also were tested. The treatment effect and treatment-strata interaction were tested after adjustment on independent prognostic factors. Results were expressed as ORs with 95% CIs. Results were considered significant when the P value was less than .05. Data were analyzed with SPSS Software for Windows, release 6.1 (SPSS Inc, Chicago, IL).
Results Baseline Characteristics of Patients The trial profile is shown in Figure 1. A total of 115 patients were randomized (56 in the failure stratum/59 in the naive stratum); 57 patients (26 in the failure stratum) were assigned randomly to receive infliximab and 58 patients (30 in the failure stratum) to receive placebo. Two patients allocated to placebo treatment (1 in each stratum) were not enrolled and did not receive the first infusion, 1 patient because his CDAI was less than 150 at inclusion and the other patient because he withdrew his consent before the inclusion visit.
April 2006
INFLIXIMAB AND AZATHIOPRINE
1057
Figure 1. Flow chart of patients randomized in the trial. The figures in parentheses correspond to the number of patients in the naive stratum and the failure stratum, respectively.
Characteristics of the 113 enrolled patients are shown in Table 1. Comparability of the 2 treatment groups was verified for all clinical characteristics, except for the percentage of patients with active perianal lesions, the CDAI level in failure stratum, and the duration of steroid treatments in the naive stratum. Efficacy The numbers of patients available for efficacy evaluation at weeks 12, 24, and 52 are shown in Figure 1. At week 24, the percentage of success (CDAI ⬍150 and off steroids) was significantly higher in the infliximab group than in the placebo group (57% vs 29%; OR, 3.3; 95% CI, 1.5–7.4; P ⫽ .003); the corresponding rates were 75% vs 38% (OR, 4.9; 95% CI, 2.2–11.0; P ⬍ .001) at week 12, and 40% vs 22% (OR, 2.4; 95% CI, 1.0 –5.7; P ⫽ .04) at week 52 (Figure 2). There was no significant interaction between treatment and stratum (P ⫽ .10; .32, and .82 at weeks 12, 24, and 52, respectively). The comparison of infliximab and placebo in each stratum is indicated in Figure 3. In the naive stratum, at week 24, the percentage of success
was significantly higher in the infliximab group than in the placebo group (63% vs 32%; OR, 3.7; 95% CI, 1.1–11.3; P ⫽ .02); the corresponding rates were 83% vs 41% (OR, 7.1; 95% CI, 2.1–24.9; P ⫽ .009) at week 12, and 52% vs 32%; (OR, 2.3; 95% CI, .7– 6.9; P ⫽ .14) at week 52. In the failure stratum, the success rate also was higher in the infliximab group than in the placebo group at week 24 (50% vs 26%; OR, 2.9; 95% CI, .9 –9.3; P ⫽ .08) and at week 12 (64% vs 34%; OR, 3.4; 95% CI, 1.1–10.3; P ⫽ .03); a trend was found at week 52 (27% vs 12%; OR, 2.8; 95% CI, .6 –12.4; P ⫽ .16). These results must be considered cautiously because of the absence of demonstrable interaction. Steroid resistance was less common in the infliximab group than in the placebo group (5% vs 23%; OR, 5.1; 95% CI, 1.3–19.2; P ⫽ .01). The median cumulative dose (interquartile range) of prednisone was lower in the infliximab group at 1110 mg/24 wk (630 –1720 mg/24 wk) vs 1870 mg/24 wk (1110 – 2710 mg/24 wk) (P ⫽ .002). The median (interquartile range) of the side-effect steroid score, however,
1058
LEMANN ET AL
GASTROENTEROLOGY Vol. 130, No. 4
Table 1. Baseline Characteristics of the Patients Enrolled in the Trial According to Study Treatment and Stratum Failure stratum Placebo n ⫽ 29 Sex (female)a Age, yb Disease duration, yb Disease locationa Small bowel Colon Both Active perianal diseasea CDAIb Duration of steroids, mob Steroid dose, mg/dayb Steroid dose, mg/kg/dayb Duration of AZA/6-MP, mob Dosage of AZA,c mg/kg/dayb Dosage of AZA,c mg/dayb CDEISb (n ⫽ 52) C-reactive protein,b mg/L
69 29 (23–33) 7 (3–11) 11 50 39 14 181 (154–259) 16 (9–36) 30 (20–40) .5 (.3–.7) 21 (10–44) 2.4 (2.1–2.6) 150 (125–150) 9 (6–15)d 8 (4–35)
Naive stratum
Infliximab n ⫽ 26 69 26 (22–37) 5 (4–10) 20 32 48 35 240 (219–281) 25 (13–36) 33 (20–46) .6 (.4–.8) 27 (10–33) 2.3 (2.0–2.5) 125 (100–150) 9 (4–14)e 19 (4–47)h
P
Placebo n ⫽ 27
Infliximab n ⫽ 31
P
.95 .52 .97
43 26 (22–36) 4 (1–8)
39 27 (22–38) 3 (1–6)
.75 .79 .60
.37
.07 .02 .20 .32 .25 .95 .27 .39 .63 .63
15 26 59 7 112 (42–262) 15 (9–52) 25 (20–40) .4 (.3–.6) 0 2.2 (2.0–2.4) 125 (100–150) 6 (3–14)f 17 (7–31)i
35 13 52 32 146 (90–244) 9 (7–22) 25 (20–40) .4 (.3–.6) 0 2.2 (2.0–2.4) 150 (125–150) 11 (5–16)g 20 (10–30)
.74
.02 .48 .03 .85 .77 .94 .20 .50 .36
of patients; 2 test. (interquartile range); Mann–Whitney U test. cAll patients were treated with AZA. dn ⫽ 17. en ⫽ 9. fn ⫽ 14. gn ⫽ 12. hn ⫽ 25. in ⫽ 26. aPercent bMedian
was not different in the 2 treatment groups at 2.0 (.0 – 4.0) vs 2.0 (1.0 – 4.0) at week 6 (P ⫽ .65), 2.0 (.0 –3.0) vs 1.0 (.0 –3.0) at week 12 (P ⫽ .91), and 1.0 (.0 –2.0) vs .5 (.0 –1.8) at week 24 (P ⫽ .42). There was no significant interaction between treatment and stratum. Factors associated with success at week 24 in the multivariate logistic regression analysis were a low CDAI at baseline (OR, 4.5; 95% CI, 1.5–13.7; P ⫽ .005), a
young age (OR, 5.9; 95% CI, 1.8 –19.6; P ⫽ .002), the absence of small-bowel involvement (OR, 4.3; 95% CI, 1.5–12.2; P ⫽ .004), and a long duration of steroids before entry (OR, 5.5; 95% CI, 1.3–22.5; P ⫽ .01). After adjusting for these factors, infliximab still was superior to placebo (OR, 5.7; 95% CI, 2.0 –15.9; P ⬍.001) and no interaction between treatment and strata could be seen (P ⫽ .63).
Figure 2. Percent of patients in clinical remission and off steroids at weeks 12, 24 (primary end point), and 52 (follow-up evaluation). The figures below the bars correspond to the number of patients analyzed at each date. □, AZA/6-MP ⫹ placebo; , AZA/6-MP ⫹ infliximab 5 mg/kg.
April 2006
INFLIXIMAB AND AZATHIOPRINE
1059
Figure 3. Percent of patients in clinical remission and off steroids in the 2 strata at weeks 12, 24 (primary end point), and 52 (follow-up evaluation). The figures below the bars correspond to the number of patients analyzed at each date. □, AZA/ 6-MP ⫹ placebo; , AZA/6-MP ⫹ infliximab 5 mg/kg.
Endoscopy Colonoscopy was performed at baseline in 52 patients (21 in the infliximab group and 31 in the placebo group); 20 of them had a second colonoscopy at week 24 (11 in the infliximab group and 9 in the placebo group). CDEIS decrease (median [interquartile range]) from baseline to week 24 was 6.9 (4.1–9.5) in the infliximab group and 1.2 (⫺1.5 to 4.4) in the placebo group (P ⫽ .05). The number of patients with no ulceration at week 24 was 3 of 11 (27%) and 3 of 9 (33%) in the infliximab and placebo groups, respectively (P ⫽ .77). Among the 16 patients in remission and off steroids at week 24 (9 in the infliximab group and 7 in the placebo group), the CDEIS decrease (median [interquartile range]) was higher in the infliximab group than in the placebo group, 6.9 (1.2–10.2) vs 1.2 (.0 – 4.0) (P ⫽ .05); the CDEIS decrease was more than 5 in 6 of the 9 patients in the infliximab group and in 0 of the 7 patients in the placebo group (P ⫽ .01). The number of patients with no ulceration at week 24 was 3 of 9 (33%) and 3 of 7 (43%) in the infliximab and placebo groups, respectively (P ⫽ .70). Safety The frequency and severity of adverse events were not different between the 2 treatment groups (Table 2). The percent of patients who had at least 1 adverse event was 51% (29 of 57) in the infliximab group and 50% (28 of 56) in the placebo group. The frequency of infection was similar in the 2 treatment groups. Of note, 5 serious adverse events were probably or possibly related to azathioprine (Table 2). One patient in the infliximab group
had a severe reaction after the second and third infusions (2%). No malignancy and no deaths occurred among the study patients.
Discussion Our study was designed to show the superiority of treatment with a combination of AZA/6-MP with infliximab (3 infusions at 0, 2, and 6 weeks), compared with AZA/6-MP alone, in patients with steroid-dependent CD. With the addition of infliximab to AZA/6-MP we showed the following: (1) the rate of success was nearly doubled, defined as complete clinical remission (CDAI, ⬍150) without steroids from 38% to 75% at week 12, from 29% to 57% at week 24, and from 22% to 40% at 52 weeks; the corresponding odds of success were increased by a factor of about 5, 3.5, and 2.5; and (2) it reduces exposure to steroids. Although we found that infliximab was associated with lower steroid consumption, steroid side effects were not decreased significantly by infliximab. Infliximab thus could be used as a bridge, waiting for the delayed effect of AZA/6-MP. An important point to consider in clinical practice, however, is whether or not the bridge effect is holding with time. Two different situations could be seen here. In patients with previous failure to AZA/6-MP, a loss of efficacy was observed gradually, and at week 52 only 27% of the patients were still in remission off steroids in the infliximab group. In such patients, the bridge strategy thus is questionable. The majority of patients probably would benefit from optimization of AZA/6-MP treatment using higher doses adjusted to 6-thioguanine nucleotide levels (which was not made systematically in our study), or from the switch
1060
LEMANN ET AL
GASTROENTEROLOGY Vol. 130, No. 4
Table 2. Adverse Events Occurring During the Study Period: Day 0 to Week 24
Total Mild Moderate Severe Infections Upper respiratory tract Dental Otitis Cutaneous Herpes virus Urinary tract Intestinal Perianal abscess Pelvic abscess Arthralgia, myalgia Abdominal pain, diarrhea Nausea or vomiting Headache Cutaneous rash, pruritus Fatigue Reaction to infusion Fever Mild increase of transaminase levels Pancreatitis Miscellaneous
Number of events in the infliximab group
Number of events in the placebo group
79 39 36 4a 18 8 2 0 3 2 1 1 0 1d 8 9 10e 6 5 5 2f 1 1
66 27 30 9a 16 11b 1 1 1 0 1 0 1c 0 13e 5 3 4 7e 2 0 4 0
0e 14
2e 10
adverse event in 5 cases (infliximab, n ⫽ 3; placebo, n ⫽ 3). patient in the placebo group also had a severe pneumonitis caused by Legionella pneumophila during the follow-up period, and recovered with antibiotics. cPerianal abscess occurring at week 18; surgical treatment. dPelvic abscess discovered at week 24; surgical treatment. eFive patients had adverse events possibly or probably related to azathioprine, 1 in the infliximab group (severe vomiting, week 6), and 4 in the placebo group including 1 case of arthralgia, myalgia, cough, diarrhea, and fever (week 6), 1 case of arthralgia, fever, and cutaneous rash (week 6), and 2 cases of pancreatitis (weeks 4 and 5); all the patients stopped azathioprine and recovered. One additional case of pancreatitis occurred in the infliximab group at week 23 during the follow-up period, azathioprine was stopped but the cause remained unclear because biliary stones also were found. f Two severe reactions in the same patient, 1 after the second infliximab infusion and the other after the third infusion; the patient recovered with steroids and adrenalin. aSerious bOne
to another immunosuppressant such as methotrexate or to maintenance treatment with infliximab. In contrast, among patients naive for AZA/6-MP at inclusion, despite a similar loss of efficacy with time, more than 50% were always in remission and off steroids at 1 year. In such patients the strategy of using infliximab as a bridge between steroids and AZA/6-MP could be considered. Optimization of AZA/6-MP treatment also may have, in this setting, improved long-term results. By comparison, in the ACCENT 1 infliximab maintenance trial, responders to the initial infusion of infliximab (n ⫽ 335)
were randomized to infliximab (5 mg/kg) or placebo at weeks 2 and 6, and then to infusions of infliximab 5 mg/kg, 10 mg/kg, or placebo every 8 weeks. At week 54, the rates of remission and off steroids were 24%, 32%, and 9%, respectively.9 However, in this study only 29% of patients were treated with immunosuppressants at baseline. A potential drawback of the bridge strategy is that the interruption of infliximab after an induction scheme could facilitate the development of anti-infliximab antibody, and thus compromise a re-treatment with infliximab, if necessary.13,14 Unfortunately, we were not able to assess the anti-infliximab antibody in our trial. The risk of immunization must be weighed against the potential for increased toxicity15 and cost for patients receiving regular maintenance treatment. In the present study, the crude number and the rate of adverse events were not different in the 2 treatment groups. However, there is still some uncertainty regarding the long-term safety of maintenance infliximab including infections and the development of lymphoproliferative disorders.16,17 In our study, the rather low effectiveness of AZA/ 6-MP alone in the naive stratum for steroid-dependent patients (32% of success at 24 and 52 weeks) compared with previous studies3,4,6 was probably the effect of the stringent criterion of success that was used (ie, remission off steroids). Moreover, all the patients included in our study were steroid-dependent, although this was not always the case in previous studies. On the other hand, a remarkably high success rate (83%) was observed in the infliximab group at week 12. This may be owing to concurrent AZA/6-MP therapy. Two retrospective and 4 prospective studies that comprised a total of 738 patients with active CD reported a trend toward a higher response to infliximab in patients receiving concurrent immunosuppressants compared with those not receiving any immunosuppressants.18 A subgroup of patients had a colonoscopic evaluation during our study. Despite the small number of patients, endoscopic improvement was more pronounced in the infliximab group than in the placebo group. When the comparison was restricted to patients in clinical remission and off steroids at week 24 (n ⫽ 16), the superiority of infliximab was confirmed. Even if a selection bias cannot be ruled out, this result confirms a previous report of mucosal healing with infliximab.14 In conclusion, our study shows that the combination of infliximab with azathioprine is more effective than azathioprine alone in steroid-dependent patients to achieve remission without steroids, and to reduce exposure to steroids. In AZA/6-MP–naive patients, infliximab could be used as a bridge between steroids and
April 2006
azathioprine. If a relapse occurs, regular maintenance therapy with infliximab should be considered. One cannot exclude, however, that the continuation of regular infusions of infliximab combined with AZA/6-MP could have resulted in a better success rate in our specific study population. This point should be tested in future studies. In AZA/6-MP–treatment failure patients, a loss of effectiveness was observed gradually, and the bridge strategy cannot be recommended. Other options such as infliximab maintenance should be considered.
References 1. Modigliani R, Mary JY, Simon JF, Cortot A, Soulé JC, Gendre JP, Rene E. Clinical, biological, and endoscopic picture of attacks of Crohn’s disease. Evolution on prednisolone. Groupe d’Etude Therapeutique des Affections Inflammatoires Digestives. Gastroenterology 1990;98:811– 818. 2. Munkholm P, Langholz E, Davidsen M, Binder V. Frequency of glucocorticoid resistance and dependency in Crohn’s disease. Gut 1994;35:360 –362. 3. Candy S, Wright J, Gerber M, Adams G, Gerig M, Goodman R. A controlled double blind study of azathioprine in the management of Crohn’s disease. Gut 1995;37:674 – 678. 4. Present DH, Korelitz BI, Wisch N, Glass JL, Sachar DB, Pasternack BS. Treatment of Crohn’s disease with 6-mercaptopurine. A long-term, randomized, double-blind study. N Engl J Med 1980; 302:981–987. 5. O’Donoghue DP, Dawson AM, Powel-Tuck K, Bown RL. Doubleblind withdrawal of azathioprine as maintenance treatment for Crohn’s disease. Lancet 1978;2:944 –946. 6. Pearson DC, May GR, Fick GH, Sutherland LR. Azathioprine and 6-mercaptopurine in Crohn’s disease. A meta-analysis. Ann Intern Med 1995;123:132–142. 7. Pearson DC, May GR, Fick G, Sutherland LR. Azathioprine for maintaining remission of Crohn’s disease. Cochrane Database Syst Rev 2000;(2):CD000067. 8. Targan SR, Hanauer SB, van Deventer SJ, Mayer L, Present DH, Braakman T, DeWoody KL, Schaible TF, Rutgeerts PJ. A shortterm study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohn’s disease. Crohn’s Disease cA2 Study Group. N Engl J Med 1997;337:1029 –1035. 9. Hanauer SB, Feagan BG, Lichtenstein GR, Mayer LF, Schreiber S, Colombel JF, Rachmilewitz D, Wolf DC, Olson A, Bao W, Rutgeerts P. Maintenance infliximab for Crohn’s disease: the ACCENT I randomised trial. Lancet 2002;359:1541–1549. 10. Best WR, Becktel JM, Singleton JW, Kern F. Development of a Crohn’s disease activity index. National Cooperative Crohn’s Disease Study. Gastroenterology 1976;70:439 – 444. 11. Mary JY, Modigliani R. Development and validation of an endoscopic index of the severity for Crohn’s disease: a prospective multicentre study. Groupes d’ Etudes Thérapeutiques des Affections Inflammatoires du Tube Digestif (GETAID). Gut 1989;30:983–989.
INFLIXIMAB AND AZATHIOPRINE
1061
12. Sandborn WJ, Feagan BG, Hanauer SB, Lochs H, Lofberg R, Modigliani R, Present DH, Rutgeerts P, Scholmerich J, Stange EF, Sutherland LR. A review of activity indices and efficacy endpoints for clinical trials of medical therapy in adults with Crohn’s disease. Gastroenterology 2002;122:512–530. 13. Baert F, Noman M, Vermeire S, Van Assche G, D’Haens G, Carbonez A, Rutgeerts P. Influence of immunogenicity on the long-term efficacy of infliximab in Crohn’s disease. N Engl J Med 2003;348:601– 608. 14. Rutgeerts P, Feagan BG, Lichtenstein GR, Mayer LF, Schreiber S, Colombel JF, Rachmilewitz D, Wolf DC, Olson A, Bao W, Hanauer SB. Comparison of scheduled and episodic treatment strategies of infliximab in Crohn’s disease. Gastroenterology 2004;126: 402– 413. 15. Colombel JF, Loftus EV Jr, Tremaine WJ, Egan LJ, Harmsen WS, Schleck CD, Zinsmeister AR, Sandborn WJ. The safety profile of infliximab in patients with Crohn’s disease: the Mayo Clinic Experience in 500 patients. Gastroenterology 2004;126:19 – 31. 16. Ljung T, Karlen P, Schmidt D, Hellstrom PM, Lapidus A, Janczewska I, Sjoqvist U, Lofberg R. Infliximab in inflammatory bowel disease: clinical outcome in a population based cohort from Stockholm County. Gut 2004;53:849 – 853. 17. Wolfe F, Michaud K. Lymphoma in rheumatoid arthritis: the effect of methotrexate and anti-tumor necrosis factor therapy in 18,572 patients. Arthritis Rheum 2004;50:1740 –1751. 18. Su C, Lichtenstein GR. Are there predictors of Remicade treatment success or failure? Adv Drug Deliv Rev 2005;57:237–245.
Received March 21, 2005. Accepted December 28, 2005. Address requests for reprints to: Professor Marc Lémann, MD, PhD, Service de Gastroentérologie, 1 avenue Claude Vellefaux, 75010, Paris, France. e-mail: [email protected]; fax: (33) 1-42-49-91-68. Presented in part at the Digestive Disease Week, Orlando, May 17-22, 2003, and at the United European Gastroenterology Week (1-year follow-up), Madrid, November 1-5, 2003. Supported by Groupe d’Etude Thérapeutique des Affections Inflammatoires du tube Digestif (GETAID), and by grants from Schering Plough, France, with the specific help of Gérard Trape and Yves-Dominique Henry. Drugs were provided by Schering Plough, France. All data analysis and manuscript writing were performed independently by the GETAID, without the involvement of representatives of Schering Plough. In addition to the authors, the following French investigators participated in the study: Robert Modigliani, Hôpital Saint-Louis, Paris, France; Etienne-Henry Metman, Hôpital Trousseau Tours; Jean-Paul Galmiche, Hôtel Dieu, Nantes, France; Jean-Pierre Gendre, Hôpital Saint-Antoine, Paris, France; Marc-André Bigard, Hôpital Brabois, Nancy, France; Stanislas Chaussade and Vered Abitbol, Hôpital Cochin, Paris, France; Jean-Charles Grimaud, Hôpital Nord, Marseille, France; Gilles Bommelaer, Hotel Dieu, Clermont-Ferrand, France; JeanClaude Soulé, Hôpital Bichât, Paris, France; Bruno Bonaz, Hôpital Nord, Grenoble, France; Louis Descos and Bernard Flourié, Hôpital Lyon-Sud, Lyon, France; Eric Lerebours, Hôpital Charles Nicolle, Rouen, France; Benoit Coffin, Hôpital Louis Mourier, Colombes, France. The authors acknowledge Patricia De ´tre ´, Study Coordinator, for her technical help.