GASTROENTEROLOGY 2012;143:62– 69
`
`Abatacept for Crohn’s Disease and Ulcerative Colitis
`WILLIAM J. SANDBORN,*,‡ JEAN–FREDERIC COLOMBEL,§ BRUCE E. SANDS,储,¶ PAUL RUTGEERTS,#
`STEPHAN R. TARGAN,** REMO PANACCIONE,‡‡ BRIAN BRESSLER,§§ KARL GEBOES,储 储 STEFAN SCHREIBER,¶¶
`RICHARD ARANDA,## SHEILA GUJRATHI,## ALLISON LUO,## YUN PENG,## LUISA SALTER–CID,## and
`STEPHEN B. HANAUER***
`
`*University of California San Diego, La Jolla, California; ‡Mayo Clinic, Rochester, Minnesota; §Hôpital Huriez, Lille, France; 储Mount Sinai School of Medicine, New York,
`New York; ¶Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; #University Hospital Gasthuisberg, Leuven, Belgium; **Cedars-
`Sinai Medical Center, Los Angeles, California; ‡‡University of Calgary, Calgary, Alberta, Canada; §§Gastrointestinal Research Institute, Vancouver, Canada;
`储 储Department of Pathology, Katholieke Universiteit, Leuven, Belgium; ¶¶Christian Albrechts University, Kiel, Germany; ##Bristol Myers-Squibb, Princeton, New Jersey;
`***University of Chicago, Chicago, Illinois
`
`CLINICAL AT
`
`See editorial on page 13.
`
`BACKGROUND & AIMS: The efficacy of abatacept, a se-
`lective costimulation modulator, in Crohn’s disease (CD)
`and ulcerative colitis (UC) is unknown. METHODS: Four
`placebo-controlled trials evaluated the efficacy and safety of
`abatacept as induction (IP) and maintenance (MP) therapy
`in adults with active, moderate-to-severe CD (CD-IP; CD-
`MP) and UC (UC-IP1; UC-MP). In CD-IP and UC-IP1, 451
`patients with CD and 490 patients with UC were random-
`ized to abatacept 30, 10, or 3 mg/kg (according to body
`weight) or placebo, and dosed at weeks 0, 2, 4, and 8. In MP,
`90 patients with CD and 131 patients with UC who re-
`sponded to abatacept at week 12 in the induction trials were
`randomized to abatacept 10 mg/kg or placebo every 4 weeks
`through week 52. RESULTS: In CD-IP, 17.2%, 10.2%, and
`15.5% of patients receiving abatacept 30, 10, and 3 mg/kg
`achieved a clinical response at weeks 8 and 12, vs 14.4%
`receiving placebo (P ⫽ .611, P ⫽ .311, and P ⫽ .812, respec-
`tively). In UC-IP1, 21.4%, 19.0%, and 20.3% of patients receiv-
`ing abatacept 30, 10, and 3 mg/kg achieved a clinical re-
`sponse at week 12, vs 29.5% receiving placebo (P ⫽ .124, P ⫽
`.043, and P ⫽ .158, respectively). In CD-MP, 23.8% vs 11.1%
`of abatacept vs placebo patients were in remission at week
`52. In UC-MP, 12.5% vs 14.1% of patients receiving abatacept
`vs placebo were in remission at week 52. Safety generally was
`comparable between groups. CONCLUSIONS: The stud-
`ies showed that abatacept is not efficacious for the treat-
`ment of moderate-to-severe CD or UC. ClinicalTrials.
`gov NCT00406653, NCT00410410.
`
`Keywords: Clinical Trial; Inflammatory Bowel Disease;
`IBD; T-Cell Signaling Inhibitor.
`
`Crohn’s disease (CD) and ulcerative colitis (UC) are
`
`chronic inflammatory bowel diseases.1,2 Treatments ei-
`ther have broad mechanisms of action (mesalamine, corti-
`costeroids, azathioprine, mercaptopurine, methotrexate) or
`target the tumor necrosis factor-␣ cytokine.2 T cells are
`believed to play a role in the pathogenesis of both condi-
`tions; thus, therapies targeting T cells are of interest.3
`
`T-cell activation requires co-stimulatory signaling via T
`cell CD28 and CD80 or CD86 on the antigen-presenting
`cell.4 Naturally occurring inhibitory cytotoxic T-lympho-
`cyte antigen 4 is induced on the T-cell surface 24 – 48
`hours after activation and attenuates CD28-mediated co-
`stimulation. Cytotoxic T-lymphocyte antigen 4 has a
`greater affinity for CD80 or CD86 than CD28, and pre-
`vents CD28 binding to CD80 or CD86.5 Abatacept is a
`recombinant fusion protein comprising a fragment of the
`Fc domain of human IgG1 and the extracellular domain
`of human cytotoxic T-lymphocyte antigen 4.6 Similar to
`cytotoxic T-lymphocyte antigen 4, abatacept competes
`with CD28 for CD80 and CD86 binding to block co-
`stimulatory signaling, thus selectively modulating T-cell
`activation.7 Abatacept is effective for rheumatoid arthri-
`tis8,9 and juvenile idiopathic arthritis.10 In animal models
`of colitis, abatacept reduces inflammation.11,12
`We conducted 12-week induction trials of abatacept in
`patients with moderate-to-severe CD and UC. Responders
`enrolled in 52-week maintenance trials.
`
`Materials and Methods
`Patients
`The initial CD and UC induction periods (CD-IP and
`UC-IP1) and maintenance periods (CD-MP and UC-MP) were
`randomized, double-blind, placebo-controlled studies
`con-
`ducted at 142 centers between December 2006 and November
`2009. Protocols were approved by institutional review boards.
`Patients provided written informed consent.
`Eligible patients were ⱖ18 years, with CD or UC (ⱖ3 mo).
`Patients with CD had a Crohn’s Disease Activity Index (CDAI)
`score13 of 220 – 450 and a high sensitivity C-reactive protein
`concentration above the upper normal limit. Patients with UC
`had a Mayo Clinic score14 of 6 –12 with moderate-to-severely
`active disease on sigmoidoscopy (endoscopic subscore, ⱖ2). Pa-
`tients had a current/previous inadequate response to, or did not
`
`Abbreviations used in this paper: CD, Crohn’s disease; CDAI, Crohn’s
`Disease Activity Index; CD-IP, Crohn’s disease induction period; CD-MP,
`Crohn’s disease maintenance period; SAE, serious adverse event; Treg,
`regulatory T cells; UC, ulcerative colitis; UC-IP, ulcerative colitis induc-
`tion period; UC-MP, ulcerative colitis maintenance period.
`© 2012 by the AGA Institute
`0016-5085/$36.00
`http://dx.doi.org/10.1053/j.gastro.2012.04.010
`
`

`

`CLINICALAT
`
`July 2012
`
`ABATACEPT FOR CD AND UC 63
`
`tolerate, ⱖ1: oral 5-aminosalicylates at or above the approved
`dose for ⱖ6 weeks (UC) or ⱖ8 weeks (CD); prednisone 40
`mg/day for ⱖ2 weeks or intravenous hydrocortisone ⱖ400 mg/
`day for ⱖ1 week (UC) or ⱖ20 mg/day for ⱖ4 weeks or budes-
`onide 9 mg/day for ⱖ4 weeks (CD); azathioprine ⱖ2 mg/kg
`body weight or 6-mercaptopurine ⱖ1 mg/kg body weight (or
`documented therapeutic concentration of 6-thioguanine nucle-
`otide metabolite concentration) for ⱖ12 weeks; methotrexate
`ⱖ15 mg/week for ⱖ12 weeks (CD); or anti–tumor necrosis
`factor (approved dose) for ⱖ8 weeks. Concurrent therapies,
`including stable doses of oral 5-aminosalicylates, prednisolone
`(ⱕ30 mg/day), budesonide (ⱕ9 mg/day; CD), azathioprine,
`6-mercaptopurine, methotrexate (CD), and antibiotics (CD)
`were permitted.
`Excluded patients were those with previous proctocolectomy
`or subtotal colectomy with ileorectal anastomosis, those who
`needed bowel surgery, or had active tuberculosis within ⱕ3
`years. Patients with symptomatic stricture, abscess, short-bowel
`syndrome, or without colonic or ileal involvement were excluded
`for CD; patients with proctitis were excluded for UC.
`
`Study Design
`The objective of these studies was to evaluate the induc-
`tion and maintenance of response and remission in patients
`with moderate-to-severely active UC and CD.
`In CD-IP and UC-IP1, patients randomly were assigned in a
`1:2:2:2 ratio (CD) or a 2:2:1:2 ratio (UC) to intravenous abata-
`cept (Orencia, Bristol-Myers Squibb, Princeton, NJ) at 30, 10, or
`3 mg/kg, or placebo, dosed at weeks 0, 2, 4, and 8, and followed
`up through week 12. After enrollment of UC-IP1, a second UC
`induction period (UC-IP2) began enrolment, in which patients
`were assigned randomly in a 1:1 ratio to receive intravenous
`abatacept 10 or 30 mg/kg at weeks 0, 2, 4, and 8, and followed
`up through week 12. Patients receiving 30 mg/kg received 2
`doses of 30 mg/kg followed by 2 doses of 10 mg/kg. In CD-MP
`and UC-MP, week 12 responders were assigned randomly (1:1) to
`intravenous abatacept (10 mg/kg) or placebo every 4 weeks
`through week 52. Randomization was performed centrally using
`dynamic treatment allocation.
`In the CD-IP study, randomization was stratified by baseline
`disease (CDAI, ⬍330 or ⱖ330) and concomitant use of azathio-
`prine, 6-mercaptopurine, or methotrexate. In the UC-IP1 study,
`randomization was stratified by history of inadequate response
`or intolerance to infliximab. In the CD-MP study, randomiza-
`tion was stratified by disease activity after induction (CDAI,
`⬍150 or ⱖ150) and concomitant use of azathioprine, 6-mercap-
`topurine, or methotrexate. In the UC-MP study, randomization
`was stratified by concomitant use of corticosteroids, clinical
`remission (Mayo Clinic score ⱕ2 with no individual subscore
`⬎1) at week 12, and history of inadequate response or intoler-
`ance to infliximab.
`Oral mesalamine, azathioprine, 6-mercaptopurine, metho-
`trexate, and antibiotics were continued at stable doses during all
`studies. Prednisone and budesonide were continued at a stable
`dose during IP. During MP, tapering of corticosteroids was
`recommended, but not mandated, for patients in remission or
`with satisfactory improvement (at the investigator’s based on
`clinical assessment).
`
`Follow-Up Evaluation, Efficacy, and Safety
`For CD, disease activity was assessed by CDAI, deter-
`mined at weeks 0, 2, 4, and every 4 weeks thereafter.15 Efficacy
`variables were as follows: response, reduction from baseline in
`
`CDAI score of ⱖ100 or absolute CDAI score of ⬍150; remission,
`absolute CDAI of ⬍150; and relapse: CDAI of ⱖ220 and an
`increase from week 12 of ⱖ100 for 2 consecutive visits. For UC,
`disease activity was assessed by the Mayo Clinic score, deter-
`mined at weeks 0, 8, 12, 36, and 64.16 Efficacy variables were as
`follows: response (reduction from baseline in Mayo Clinic score
`by ⱖ3 and ⱖ30%, with decrease in rectal bleeding subscore of
`ⱖ1, or absolute rectal bleeding subscore of 0 or 1); remission
`(Mayo Clinic score ⱕ2 with no individual subscore ⬎1); and
`mucosal healing (absolute endoscopy score of 0 or 1). The
`partial Mayo Clinic score was determined at weeks 2, 4, 16, 20,
`24, 28, 32, 40, 44, 48, and 52. Efficacy variables were as follows:
`response (reduction from baseline in partial Mayo Clinic score
`ⱖ2 and ⱖ30%, with decrease in rectal bleeding subscore of ⱖ1,
`or absolute rectal bleeding subscore of 0 or 1) and relapse
`(partial Mayo Clinic score of ⱖ4 and increase from week 12 of
`ⱖ2, and endoscopy subscore of 2 or 3).
`Colon biopsy specimens were obtained in UC-IP1 and UC-MP
`at weeks 0, 8, 12, 36, and 52. Histologic disease activity was
`assessed by the Geboes17 index. Immunohistochemistry was per-
`formed using a Dako automatic stainer with antibodies to
`CD68, CD20, CD86, CD4, FOXp3, caspase 3, and tenascin.
`Adverse events and concomitant medications were followed
`up through 8 weeks after the last dose of study drug. Blood
`samples were collected for laboratory evaluations.
`
`Statistical Analysis
`The primary end point for CD-IP was CDAI response at
`both weeks 8 and 12; remission at these time points was a
`secondary end point. The primary end point for CD-MP was
`remission at week 52 of the MP. Secondary end points included
`response at week 52, remission at both weeks 24 and 52, and
`corticosteroid-free remission at week 52. The primary end point
`for UC-IP1 was the Mayo clinic score response at week 12.
`Secondary assessments included remission and mucosal healing
`at week 12. Efficacy results for UC-IP2 are not presented. The
`primary end point for UC-MP was response at week 52. Second-
`ary end points included remission, mucosal healing, and corti-
`costeroid-free remission at week 52, and remission at both weeks
`24 and 52. Response, remission, and mucosal rates are presented
`with 95% confidence intervals and were compared for each
`abatacept treatment group vs placebo using the Cochran–Man-
`tel–Haenszel chi-square tests, accounting for randomization
`stratification factors.
`To control the overall type I error rate, all primary and key
`secondary end points in the CD-IP, UC-IP1, and UC-MP trials
`were tested in a prespecified sequential manner; accordingly, if
`prior comparisons were not significant (0.05 [2-sided] signifi-
`cance level for abatacept 30 mg/kg vs placebo), no subsequent
`comparisons were conducted. Because of the early termination
`of CD-MP and UC-MP, only descriptive statistics are presented.
`To evaluate the consistency of treatment effect on response,
`subgroup analyses based on demographic or baseline disease
`characteristics were performed, as prespecified in the statistical
`analysis plan.
`Efficacy and safety analyses included all randomized patients
`who received ⱖ1 dose of study medication. Patients who discon-
`tinued were considered not to have a response/remission. For the
`primary and key secondary end points of response and remission at
`both weeks 8 and 12, if a value was missing for reasons other than
`early discontinuation, the patient was excluded.
`For the primary end point of response at weeks 8 and 12 in
`CD-IP, it was estimated that the abatacept 30-mg/kg and pla-
`
`

`

`64 SANDBORN ET AL
`
`GASTROENTEROLOGY Vol. 143, No. 1
`
`60.7
`39.3
`
`50.0
`48.6
`
`58.3
`41.0
`
`8.8(1.6)
`
`8.6(1.8)
`
`8.8(1.7)
`
`—
`
`44.3
`34.3
`
`—
`
`38.6
`35.7
`
`—
`
`41.7
`41.0
`
`51.1
`48.2
`
`8.9(1.7)
`
`—
`
`41.8
`27.0
`
`15.5(37.1)
`
`16.7(36.7)
`
`12.6(24.3)
`
`18.1(40.2)
`
`32.1
`59.3
`70.7
`89.3
`
`34.3
`62.9
`58.6
`94.3
`
`6.6(6.2)
`
`74.4(16.8)
`
`47.1
`
`5.4(4.7)
`
`73.6(19.6)
`
`37.1
`
`40.9(13.1)
`
`40.4(13.4)
`
`(n⫽140)
`Placebo
`
`(n⫽70)
`mg/kg
`
`33.1
`60.4
`64.7
`87.8
`
`7.0(7.3)
`
`74.4(17.6)
`
`37.4
`
`42.1(13.5)
`
`(n⫽139)
`
`mg/kg
`
`31.9
`53.9
`60.3
`89.4
`
`6.3(6.7)
`
`73.2(18.9)
`
`40.4
`
`43.4(14.4)
`
`(n⫽141)
`
`mg/kg
`
`Abatacept3
`
`Abatacept⬃10
`
`Abatacept30/⬃10
`
`Ulcerativecolitis
`
`—
`
`—
`
`—
`
`64.8
`
`CLINICAL AT
`
`aOralcorticosteroidsforCrohn’sdiseaseandoralorintravenouscorticosteroidsforulcerativecolitis.
`hsCRP,highsensitivityC-reactiveprotein;SD,standarddeviation;TNF,tumornecrosisfactor.
`
`—
`
`—
`
`—
`
`—
`
`—
`
`—
`
`—
`
`—
`
`—
`
`320.7(72.1)
`
`317.9(59.9)
`
`318.9(65.1)
`
`320.6(61.6)
`
`34.4
`42.2
`
`33.1
`36.9
`
`38.3
`41.4
`
`36.9
`36.9
`
`27.9(33.0)
`
`23.5(23.3)
`
`24.2(28.8)
`
`29.3(40.4)
`
`60.2
`81.3
`71.9
`71.9
`
`59.2
`83.1
`76.2
`73.1
`
`9.8(8.3)
`
`72.2(24.4)
`
`9.2(8.0)
`
`74.5(20.3)
`
`60.0
`
`38.0(13.0)
`
`36.9(13.4)
`
`(n⫽128)
`Placebo
`
`(n⫽130)
`
`mg/kg
`
`Abatacept3
`
`67.2
`75.8
`75.0
`78.9
`
`9.9(8.7)
`
`72.0(18.7)
`
`60.9
`
`38.6(12.9)
`
`(n⫽128)
`
`mg/kg
`
`64.6
`83.1
`75.4
`75.4
`
`8.4(7.5)
`
`72.2(24.4)
`
`58.5
`
`36.0(11.1)
`
`(n⫽65)
`mg/kg
`
`Abatacept⬃10
`
`Abatacept30/⬃10
`
`Crohn’sdisease
`
`Extensivediseaseother
`Left-sideddiseaseonly
`
`Extentofdisease,%
`MeanMayoscore(SD)
`MeanCDAIscore(SD)
`Prednisoneuse,%
`Immunosuppressantuse,%
`MeanhsCRP,mg/L(SD)
`
`Anti-TNFagent(s)
`Immunosuppressants
`Corticosteroidsa
`Oralaminosalicylates
`topriortherapy,%
`
`Inadequateresponse/intolerance
`Meandiseaseduration,y(SD)
`Meanweight,kg(SD)
`Sex,%female
`Meanage,y(SD)
`
`Characteristic
`
`Table1.BaselineDemographicsandDiseaseCharacteristicsforInductionPeriodStudies
`
`

`

`CLINICALAT
`
`July 2012
`
`ABATACEPT FOR CD AND UC 65
`
`cebo groups would require 134 patients each to provide 99%
`power to detect a difference of 30%, assuming a 55% response
`rate to abatacept 30 mg/kg and a 25% response rate to placebo.
`For the primary end point of response at week 12 in UC-IP1, it
`was estimated that the abatacept 30 mg/kg and placebo groups
`would require 140 patients each to provide 98% power to detect
`a difference of 25%, assuming a 65% response rate to abatacept
`30 mg/kg and a 40% response rate to placebo.
`
`Results
`Patient Characteristics
`For CD, 451 patients were randomized in CD-IP,
`with 90 randomized in CD-MP. For UC, 490 and 101 pa-
`tients were randomized in UC-IP1 and UC-IP2, respectively,
`with 131 randomized in UC-MP. Patient disposition is
`shown (Supplementary Figure 1A and B). Because of early
`study termination, subsequent to analyses indicating lack of
`efficacy in CD-IP and UC-IP1, only 24 and 26 patients
`completed CD-MP and UC-MP, respectively. Therefore, sec-
`ondary end points at week 52 were not analyzed. Baseline
`disease characteristics for CD-IP and UC-IP1 were similar
`between treatment groups (Tables 1 and 2).
`
`Efficacy
`Crohn’s disease. Response at weeks 8 and 12 oc-
`curred in 17.2% of patients receiving abatacept 30 mg/kg (11
`of 64), 10.2% receiving abatacept 10 mg/kg (13 of 128), and
`15.5% receiving abatacept 3 mg/kg (20 of 129), vs 14.4%
`receiving placebo (18 of 125; P ⫽ .611, P ⫽ .311, and P ⫽
`.812 for abatacept 30, 10, and 3 mg/kg vs placebo; Figure
`1A). Remission rates at weeks 8 and 12 are shown in Figure
`1A. Over the 12-week IP, there was no difference between
`groups in mean CDAI scores (Supplementary Figure 2).
`Remission at week 52 of the MP occurred in 23.8% (10 of
`42) of patients receiving abatacept vs 11.1% (5 of 45) of
`
`patients receiving placebo (P ⫽ .082). Response at week 52
`of the MP occurred in 26.2% and 15.6% of patients receiving
`abatacept (11 of 42) and placebo (7 of 45), respectively (P ⫽
`.175).
`
`Ulcerative colitis. Response at week 12 occurred
`in 21.4% of patients receiving abatacept 30 mg/kg (30 of
`140), 19.0% receiving 10 mg/kg (26 of 137), and 20.3%
`receiving 3 mg/kg (14 of 69) vs 29.5% receiving placebo
`(41 of 139; P ⫽ .124, P ⫽ .043, and P ⫽ .158 for abatacept
`30, 10, and 3 mg, respectively, vs placebo). Remission rates
`at week 12 are shown in Figure 1B. Mucosal healing at
`week 12 occurred in 17.1% (24 of 140), 14.6% (20 of 137),
`and 15.9% (11 of 69) of patients receiving 30, 10, and 3 mg
`abatacept, respectively, vs 25.9% (36 of 139) receiving
`placebo. There was no difference between treatment
`groups in change from baseline in partial Mayo score over
`time (Supplementary Table 1).
`Response at week 52 of the MP occurred in 17.2% of
`patients receiving abatacept (11 of 64) compared with
`17.2% receiving placebo (11 of 64) (P ⫽ .999); clinical
`remission occurred in 12.5% (8 of 64) of patients receiving
`abatacept vs 14.1% (9 of 64) receiving placebo (P ⫽ .740).
`Additional analyses. No consistent pattern was
`identified when assessing response rates during the IPs by
`concomitant immunomodulator use, prior therapy fail-
`ure, disease duration, or baseline high sensitivity C-reac-
`tive protein in the subgroup analyses (Supplementary
`Figure 3). Changes from baseline to week 12 in high
`sensitivity C-reactive protein were comparable between
`groups (Supplementary Figure 4).
`
`Safety
`Crohn’s disease. Safety generally was comparable
`across groups during the IP and MP (Table 2). During IP,
`frequencies of serious adverse events (SAEs) generally were
`
`Table 2. Safety Summary From Crohn’s Disease Studies
`
`n (%)
`
`AEs
`Related AEs
`AEs leading to
`discontinuation
`Deaths
`SAEs
`Related SAEs
`Infections
`Serious infections
`Opportunistic infections
`Malignancies
`Possible autoimmune events
`Acute infusional AEs
`
`Induction period
`
`Maintenance period
`
`Abatacept 30/⬃10
`mg/kg
`(n ⫽ 65)
`
`Abatacept ⬃10
`mg/kg
`(n ⫽ 128)
`
`Abatacept 3
`mg/kg
`(n ⫽ 130)
`
`49 (75.4)
`24 (36.9)
`1 (1.5)
`
`0
`11 (16.9)
`2 (3.1)
`13 (20.0)
`2 (3.1)
`0
`0
`2 (3.1)
`3 (4.6)
`
`97 (75.8)
`47 (36.7)
`11 (8.6)
`
`0
`22 (17.2)
`7 (5.5)
`33 (25.8)
`9 (7.0)
`0
`0
`1 (0.8)
`3 (2.3)
`
`96 (73.8)
`46 (35.4)
`5 (3.8)
`
`0
`20 (15.4)
`9 (6.9)
`31 (23.8)
`4 (3.1)
`0
`3 (2.3)
`2 (1.5)
`4 (3.1)
`
`Placebo
`(n ⫽ 128)
`
`95 (74.2)
`40 (31.3)
`12 (9.4)
`
`0
`20 (15.6)
`6 (4.7)
`42 (32.8)
`3 (2.3)
`0
`0
`2 (1.6)
`5 (3.9)
`
`Abatacept ⬃10
`mg/kg
`(n ⫽ 44)
`
`31 (70.5)
`12 (27.3)
`1 (2.3)
`
`0
`5 (11.4)
`0
`16 (36.4)
`1 (2.3)
`0
`0
`2 (4.5)
`1 (2.3)
`
`Placebo
`(n ⫽ 46)
`
`32 (69.6)
`15 (32.6)
`0
`
`0
`9 (19.6)
`1 (2.2)
`18 (39.1)
`1 (2.2)
`0
`0
`1 (2.2)
`0
`
`NOTE. The most common serious infection in the abatacept groups during the IP was anal abscess (n ⫽ 7); other serious infections were each
`reported in only 1 or 2 patients. No opportunistic infections were reported. Three malignancies occurred during the IP and MP periods (squamous
`cell carcinoma, 2; breast cancer, 1). The most common autoimmune event was erythema nodosum (6 events during the IP [abatacept 30/⬃10
`mg/kg, 2; abatacept ⬃10 mg/kg, 1; abatacept 3 mg/kg, 2; placebo, 1] and one in the MP [placebo]).
`
`

`

`66 SANDBORN ET AL
`
`GASTROENTEROLOGY Vol. 143, No. 1
`
`(nosocomial pneumonia, septic shock, and acute respira-
`tory failure) and 1 death occurred in the abatacept group
`during the MP (varicella, varicella pneumonia, and septi-
`cemia). In the IP, a numerically higher proportion of SAEs
`occurred with the abatacept groups vs placebo, driven by
`a higher rate of UC exacerbations in patients with severe
`UC (Mayo score, ⱖ10) at baseline (Supplementary Table
`2). Of these patients with severe UC at baseline, the
`proportions requiring surgery during the IP for SAEs of
`UC exacerbation were 3.5%, 3.6%, 0%, and 0% for patients
`receiving 30, 10, or 3 mg/kg abatacept, or placebo, respec-
`tively. A numerically higher proportion of SAEs occurred
`with abatacept vs placebo during the MP, driven primarily
`by a higher frequency of serious infections (abatacept,
`7.7%; placebo, 3.0%). Two opportunistic infections were
`reported during the IP, and 1 was reported in the MP.
`Immunohistochemistry. Among the 52 patients in
`the biopsy substudy there was no significant difference in
`change in histologic grade over time across treatment
`groups (data not shown). The level of FOXp3 (marker for
`regulatory T cell) expression over time did not differ
`among biopsy specimens from abatacept-treated vs placebo-
`treated patients, or across abatacept treatment groups
`(Supplementary Figure 5). There was no correlation be-
`tween the level of expression of CD86 (marker for anti-
`gen-presenting cell) at baseline, and duration of disease
`(data not shown).
`
`Discussion
`These results show that abatacept is not effective
`in patients who have moderate-to-severe CD or UC. In
`patients with severe UC (Mayo score, ⱖ10), there were
`numerically higher rates of disease exacerbation and co-
`lectomy among patients treated with abatacept. There was
`no evidence of efficacy in any patient subgroups including
`concomitant immunomodulator use, disease duration,
`high sensitivity C-reactive protein and type of, or reason
`for, prior treatment failure. A potential limitation of these
`studies was that colonoscopy/sigmoidoscopy was not per-
`formed to confirm active disease at enrollment in the CD
`study, and data were not collected during treatment in the
`study. Such assessments were not common practice when
`these trials were designed.
`A hallmark of the immunopathology of active UC and
`CD is infiltration of innate and adaptive immune cells
`into the lamina propria. Increased numbers and activa-
`tion of these cells increase concentrations of proinflam-
`matory cytokines in the mucosa, including tumor necrosis
`factor-␣, interleukin-1␤, interleukin-6, and interferon-␥.
`Tumor necrosis factor antagonists have shown efficacy in
`both CD and UC, and other therapies that modulate
`cytokines currently are being actively tested. Over the past
`decade, therapies targeting T-cell activation have been
`investigated. Cyclosporine and tacrolimus are effective for
`induction of response in hospitalized patients with intra-
`venous steroid-refractory UC.18,19 However, it is question-
`able whether the biologic effects of these calcineurin in-
`
`Figure 1. Response rates for the primary end points during the induc-
`tion period. (A) CD-IP. Clinical response defined as a reduction from
`baseline in CDAI score by ⱖ100 points, or an absolute CDAI score ⬍150
`points. Remission was defined as an absolute CDAI score ⬍150. (B)
`UC-IP1. Clinical response was defined as a reduction from baseline in
`Mayo score of ⱖ3 points and ⱖ30%, with decrease from baseline in
`rectal bleeding subscore of ⱖ1 point or absolute rectal bleeding sub-
`score of ⱕ1. Remission was defined as a Mayo score ⱕ2 and no indi-
`vidual subscore ⬎1 point. *Comparison with placebo using a Cochran–
`Mantel–Haenszel chi-square test controlling for randomization stratification
`factors. Based on a prespecified hierarchical comparison procedure, no
`formal testing of the primary end point for other treatment groups was
`conducted because abatacept 30/⬃10 mg/kg vs placebo was not statis-
`tically significant. Error bars, 95% CI.
`
`comparable; during MP, a higher proportion of patients
`in the placebo group reported an SAE (19.6% vs 11.4%).
`During IP, serious infections were reported in 3.1%–7.0%
`of patients across the abatacept groups, and in 2.3% in the
`placebo group. During the MP, the frequencies of serious
`infections were comparable between treatment groups.
`No opportunistic infections were reported.
`Ulcerative colitis. Safety generally was comparable
`across groups during the IP and MP (Table 3). One death
`occurred in the 30-mg/kg abatacept group during the IP
`
`CLINICAL AT
`
`

`

`July 2012
`
`ABATACEPT FOR CD AND UC 67
`
`Table 3. Safety Summary From Ulcerative Colitis Studies
`
`CLINICALAT
`
`Induction period 1
`
`Induction period 2
`
`Maintenance period
`
`Abatacept
`30/⬃10
`mg/kg
`(n ⫽ 141)
`
`85 (60.3)
`48 (34.0)
`1 (0.7)
`4 (2.8)
`
`22 (15.6)
`4 (2.8)
`13 (9.2)
`23 (16.3)
`5 (3.5)
`0
`1 (0.7)
`2 (1.4)
`
`Abatacept
`⬃10 mg/
`kg
`(n ⫽ 139)
`
`92 (66.2)
`46 (33.1)
`0
`6 (4.3)
`
`20 (14.4)
`1 (0.7)
`16 (11.5)
`29 (20.9)
`0
`1 (0.7)
`0
`1 (0.7)
`
`Abatacept 3
`mg/kg
`(n ⫽ 70)
`
`39 (55.7)
`23 (32.9)
`0
`2 (2.9)
`
`8 (11.4)
`1 (1.4)
`5 (7.1)
`8 (11.4)
`1 (1.4)
`1 (1.4)
`0
`0
`
`Placebo
`(n ⫽ 140)
`
`86 (61.4)
`37 (26.4)
`0
`5 (3.6)
`
`7 (5.0)
`3 (2.1)
`4 (2.9)
`25 (17.9)
`0
`0
`1 (0.7)
`4 (2.9)
`
`Abatacept
`30/⬃10
`mg/kg
`(n ⫽ 51)
`
`26 (51.0)
`10 (19.6)
`0
`1 (2.0)
`
`6 (11.8)
`1 (2.0)
`5 (9.8)
`12 (23.5)
`0
`0
`0
`0
`
`Abatacept
`⬃10 mg/kg
`(n ⫽ 50)
`
`Abatacept
`⬃10 mg/kg
`(n ⫽ 65)
`
`27 (54.0)
`11 (22.0)
`0
`0
`
`4 (8.0)
`0
`4 (8.0)
`7 (14.0)
`0
`0
`0
`0
`
`39 (60.0)
`12 (18.5)
`1 (1.5)
`1 (1.5)
`
`7 (10.8)
`2 (3.1)
`3 (4.6)
`39 (60.0)
`5 (7.7)
`1 (1.5)
`0
`2 (3.1)
`
`Placebo
`(n ⫽ 66)
`
`36 (54.5)
`13 (19.7)
`0
`3 (4.5)
`
`4 (6.1)
`2 (3.0)
`1 (1.5)
`36 (54.4)
`2 (3.0)
`0
`1 (1.5)
`0
`
`4 (2.8)
`
`4 (2.9)
`
`0
`
`2 (1.4)
`
`0
`
`0
`
`2 (3.1)
`
`1 (1.5)
`
`n (%)
`
`AEs
`Related AEs
`Deaths
`AEs leading to
`discontinuation
`SAEs
`Related SAEs
`SAE⫺UC exacerbations
`Infections
`Serious infections
`Opportunistic infections
`Malignancies
`Possible autoimmune
`events
`Acute infusional AEs
`
`NOTE. A higher rate of infection was observed in both groups during the MP compared with the IP. Two opportunistic infections were reported
`during the IP (candidiasis [⬃10 mg/day], esophageal candidiasis [3 mg/day]), and 1 infection in the MP (varicella pneumonia [⬃10 mg/day]).
`Three malignancies were reported during the IP and MP periods (abatacept 30/⬃10 mg/day, basal cell carcinoma; placebo, malignant
`melanoma and breast cancer); the case of basal cell carcinoma was considered unrelated to the study drug. The frequency of autoimmune events
`and acute infusional AEs was low during both the IP and MP, and rates were similar across groups.
`
`hibitors are limited to T-cell activation. Conversely, more
`specific T-cell therapies such as visilizumab (anti-CD3),20
`daclizumab,21 and basiliximab (anti-CD25)22 failed to
`show efficacy in UC. Abatacept modulates T-cell function
`by inhibiting the co-stimulatory signal required for full
`naive T-cell activation. In aggregate, these findings sug-
`gest that T-cell modulation alone may not be effective for
`treatment of UC, and likely CD as well.
`The balance between effector and CD4⫹ Foxp3⫹ regu-
`latory T cells (Tregs) is critical for maintaining immune
`homeostasis in the colon. Although a decrease of Treg
`numbers and increase of proinflammatory Th17 cells has
`been observed in the peripheral blood of inflammatory
`bowel disease patients,23 an increase in numbers of Tregs
`has been seen in the lamina propria and mesenteric lymph
`nodes.24 –26 A recent study found that Foxp3 was highly
`expressed in the intestinal mucosa of both UC and CD
`patients, suggesting that Tregs actively were being re-
`cruited to suppress ongoing active inflammation.23
`Whether these Tregs have a suppressor activity and/or
`they are on their way to be converted to Th17 cells by
`prolonged exposure to inflammatory cytokines, such as
`interleukin-1␤ and interleukin-6, is unknown.27 We as-
`sessed whether decreased Tregs in colonic tissue poten-
`tially could explain the lack of efficacy observed in these
`studies by performing immunohistochemistry staining
`for Foxp3 expression in colonic biopsy specimens. Treat-
`ment with abatacept did not change the number of Foxp3
`Tregs in colonic tissue. Whether or not these cells still had
`suppressive activity was not determined, but in vitro stud-
`ies have suggested that abatacept does not impact Treg
`function (data not shown).
`
`It has been suggested that dysregulation of T-cell acti-
`vation may play a significant role early in the disease
`process in which an innate immune response to antigens
`from commensal flora leads to an increase in cytokines
`produced by CD4 T cells. Our data suggest that in more
`established disease, additional factors (such as disruption
`of intestinal epithelia barrier function and inflammatory
`mediators leading to recruitment of leukocytes) likely
`create a hurdle that cannot be overcome by blocking
`T-cell activation alone.
`In trinitrobenzene sulfonic acid and oxazalone murine
`colitis models, prophylactic treatment with abatacept
`showed efficacy as measured by prevention of weight loss,
`reduction in proinflammatory cytokine production, and
`reduction in mucosal damage.28,29 However, in the same
`murine model, abatacept failed to induce remission in
`established colitis, once again suggesting that targeting
`T-cell activation alone may not be efficacious in estab-
`lished disease.
`In conclusion, these studies showed that abatacept is
`not efficacious for the treatment of moderate-to-severe
`CD or UC.
`
`Supplementary Material
`Note: To access the supplementary material
`accompanying this article, visit the online version of
`Gastroenterology at www.gastrojournal.org, and at http://
`dx.doi.org/10.1053/j.gastro.
`
`References
`1. Baumgart DC, Carding SR. Inflammatory bowel disease: cause
`and immunobiology. Lancet 2007;369:1627–1640.
`
`

`

`68 SANDBORN ET AL
`
`GASTROENTEROLOGY Vol. 143, No. 1
`
`2. Baumgart DC, Sandborn WJ. Inflammatory bowel disease: clinical
`aspects and established and evolving therapies. Lancet 2007;
`369:1641–1657.
`3. Peyrin-Biroulet L, Desreumaux P, Sandborn WJ, et al. Crohn’s
`disease: beyond antagonists of tumour necrosis factor. Lancet
`2008;372:67– 81.
`4. Lenschow DJ, Walunas TL, Bluestone JA. CD28/B7 system of T
`cell costimulation. Ann Rev Immunol 1996;14:233–258.
`5. Collins AV, Brodie DW, Gilbert RJ, et al. The interaction properties
`of costimulatory molecules revisited. Immunity 2002;17:201–
`210.
`6. Linsley PS, Brady W, Urnes M, et al. CTLA-4 is a second receptor
`for the B cell activation antigen B7. J Exp Med 1991;174:561–
`569.
`7. Emery P. The therapeutic potential of costimulatory blockade with
`CTLA4Ig in rheumatoid arthritis. Exp Opin Investig Drugs 2003;12:
`673– 681.
`8. Kremer JM, Dougados M, Emery P, et al. Treatment of rheumatoid
`arthritis with the selective costimulation modulator abatacept:
`twelve-month results of a phase iib, double-blind, randomized,
`placebo-controlled trial. Arthritis Rheum 2005;52:2263–2271.
`9. Kremer JM, Genant HK, Moreland LW, et al. Effects of abatacept
`in patients with methotrexate-resistant active rheumatoid arthritis:
`a randomized trial. Ann Intern Med 2006;144:865– 876.
`10. Ruperto N, Lovell DJ, Quartier P, et al. Abatacept in children with
`juvenile idiopathic arthritis: a randomised, double-blind, placebo-
`controlled withdrawal trial. Lancet 2008;372:383–391.
`11. Davenport CM, McAdams HA, Kou J, et al. Inhibition of pro-
`inflammatory cytokine generation by CTLA4-Ig in the skin and
`colon of mice adoptively transplanted with CD45RBhi CD4⫹ T
`cells correlates with suppression of psoriasis and colitis. Int
`Immunopharmacol 2002;2:653– 672.
`12. Read S, Malmstrom V, Powrie F. Cytotoxic T lymphocyte-associ-
`ated antigen 4 plays an essential
`role in the function of
`CD25(⫹)CD4(⫹) regulatory cells that control intestinal inflamma-
`tion. J Exp Med 2000;192:295–302.
`13. Best WR, Becktel JM, Singleton JW, et al. Development of a
`Crohn’s disease activity index. National Cooperative Crohn’s Dis-
`ease Study. Gastroenterology 1976;70:439 – 444.
`14. Schroeder KW, Tremaine WJ, Ilstrup DM. Coated oral 5-aminosal-
`icylic acid therapy for mildly to moderately active ulcerative colitis.
`A randomized study. N Engl J Med 1987;317:1625–1629.
`15. Sandborn WJ, Feagan BG, Hanauer SB, et al. 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.
`16. D’Haens G, Sandborn WJ, Feagan BG, et al. A review of activity
`indices and efficacy end points for clinical trials of medical therapy
`in adults with ulcerative colitis. Gastroenterology 2007;132:763–
`786.
`17. Geboes K, Riddell R, Ost A, et al. A reproducible grading scale for
`histological assessment of inflammation in ulcerative colitis. Gut
`2000;47:404 – 409.
`18. Loftus CG, Loftus EV Jr, Sandborn WJ. Cyclosporin for refractory
`ulcerative colitis. Gut 2003;52:172–173.
`19. Ogata H, Matsui T, Nakamura M, et al. A randomised dose finding
`study of oral tacrolimus (FK506) therapy in refractory ulcerative
`colitis. Gut 2006;55:1255–1262.
`20. Sandborn WJ, Colombel JF, Frankel M, et al. Anti-CD3 antibody
`visilizumab is not effective in patients with intravenous corticoste-
`roid-refractory ulcerative colitis. Gut 2010;59:1485–1492.
`21. Van Assche G, Sandborn WJ, Feagan BG, et al. Daclizumab, a
`humanised monoclonal antibody to the interleukin 2 recept