ORENCIA Powder for concentrate for solution for infusion Ref.[6045] Active ingredients: Abatacept

Pharmacotherapeutic group: selective immunosuppressants
ATC code: L04AA24

Abatacept is a fusion protein that consists of the extracellular domain of human cytotoxic Tlymphocyte-associated antigen 4 (CTLA-4) linked to a modified Fc portion of human immunoglobulin G1 (IgG1). Abatacept is produced by recombinant DNA technology in Chinese hamster ovary cells.

Mechanism of action

Abatacept selectively modulates a key costimulatory signal required for full activation of T lymphocytes expressing CD28. Full activation of T lymphocytes requires two signals provided by antigen presenting cells: recognition of a specific antigen by a T cell receptor (signal 1) and a second, costimulatory signal. A major costimulatory pathway involves the binding of CD80 and CD86 molecules on the surface of antigen presenting cells to the CD28 receptor on T lymphocytes (signal 2). Abatacept selectively inhibits this costimulatory pathway by specifically binding to CD80 and CD86. Studies indicate that naive T lymphocyte responses are more affected by abatacept than memory T lymphocyte responses.

Studies in vitro and in animal models demonstrate that abatacept modulates T lymphocyte-dependent antibody responses and inflammation. In vitro, abatacept attenuates human T lymphocyte activation as measured by decreased proliferation and cytokine production. Abatacept decreases antigen specific TNFα, interferon-γ, and interleukin-2 production by T lymphocytes.

Pharmacodynamic effects

Dose-dependent reductions were observed with abatacept in serum levels of soluble interleukin2 receptor, a marker of T lymphocyte activation; serum interleukin-6, a product of activated synovial macrophages and fibroblast-like synoviocytes in rheumatoid arthritis; rheumatoid factor, an autoantibody produced by plasma cells; and C-reactive protein, an acute phase reactant of inflammation. In addition, serum levels of matrix metalloproteinase-3, which produces cartilage destruction and tissue remodelling, were decreased. Reductions in serum TNFα were also observed.

Clinical efficacy and safety in adult rheumatoid arthritis

The efficacy and safety of abatacept were assessed in randomised, double-blind, placebo-controlled clinical trials in adult patients with active rheumatoid arthritis diagnosed according to American College of Rheumatology (ACR) criteria. Studies I, II, III, V, and VI required patients to have at least 12 tender and 10 swollen joints at randomization. Study IV did not require any specific number of tender or swollen joints.

In Studies I, II, and V the efficacy and safety of abatacept compared to placebo were assessed in patients with an inadequate response to methotrexate and who continued on their stable dose of methotrexate. In addition, Study V investigated the safety and efficacy of abatacept or infliximab relative to placebo. In Study III the efficacy and safety of abatacept were assessed in patients with an inadequate response to a TNF-inhibitor, with the TNF-inhibitor discontinued prior to randomization; other DMARDs were permitted. Study IV primarily assessed safety in patients with active rheumatoid arthritis requiring additional intervention in spite of current therapy with non-biological and/or biological DMARDs; all DMARDs used at enrollment were continued. In Study VI, the efficacy and safety of abatacept were assessed in methotrexate-naive, Rheumatoid Factor (RF) and/or anti-Cyclic Citrullinated Peptide 2 (Anti-CCP2)-positive patients with early, erosive rheumatoid arthritis (≤2 years disease duration) who were randomized to receive abatacept plus methotrexate or methotrexate plus placebo. Study SC-II investigated the relative efficacy and safety of abatacept and adalimumab, both given subcutaneously without an intravenous loading dose and with background MTX, in patients with moderate to severely active RA and an inadequate response to previous MTX therapy. In study SC-III, abatacept SC was evaluated in combination with methotrexate (MTX), or as abatacept monotherapy, and compared to MTX monotherapy in induction of remission following 12 months of treatment, and the possible maintenance of drug-free remission after complete drug withdrawal, in adult MTX-naive patients with highly active early, rheumatoid arthritis (mean DAS28-CRP of 5.4; mean symptom duration less than 6.7 months) with poor prognostic factors for rapidly progressive disease (e.g., anti-citrullinated protein antibodies [ACPA+], as measured by antiCCP2 assay, and/or RF+, baseline joint erosions).

Study I patients were randomized to receive abatacept 2 or 10 mg/kg or placebo for 12 months. Study II, III, IV, and VI patients were randomized to receive a fixed dose approximating 10 mg/kg of abatacept or placebo for 12 (Studies II, IV, and VI) or 6 months (Study III). The dose of abatacept was 500 mg for patients weighing less than 60 kg, 750 mg for patients weighing 60 to 100 kg, and 1,000 mg for patients weighing greater than 100 kg. Study V patients were randomized to receive this same fixed dose of abatacept or 3 mg/kg infliximab or placebo for 6 months. Study V continued for an additional 6 months with the abatacept and infliximab groups only.

Studies I, II, III, IV, V, VI, SC-II, and SC-III evaluated 339, 638, 389, 1441, 431, 509, 646, and 351 adult patients, respectively.

Clinical response

ACR response

The percent of abatacept-treated patients achieving ACR 20, 50, and 70 responses in Study II (patients with inadequate response to methotrexate), Study III (patients with inadequate response to TNFinhibitor), and Study VI (methotrexate-naive patients) are shown in Table 3.

In abatacept-treated patients in Studies II and III, statistically significant improvement in the ACR 20 response versus placebo was observed after administration of the first dose (day 15), and this improvement remained significant for the duration of the studies. In Study VI, statistically significant improvement in the ACR 20 response in abatacept plus methotrexate-treated patients versus methotrexate plus placebo-treated patients was observed at 29 days, and was maintained through the duration of the study. In Study II, 43% of the patients who had not achieved an ACR 20 response at 6 months developed an ACR 20 response at 12 months.

Table 3. Clinical Responses in Controlled Trials:

* p<0.05, abatacept vs. placebo.
** p<0.01, abatacept vs. placebo.
*** p<0.001, abatacept vs. placebo.
^† p<0.01, abatacept plus MTX vs. MTX plus placebo
^‡ p<0.001, abatacept plus MTX vs. MTX plus placebo
^†† p<0.05, abatacept plus MTX vs. MTX plus placebo
^a Fixed dose approximating 10 mg/kg (see section 4.2).
^b Concurrent DMARDs included one or more of the following: methotrexate, chloroquine/hydroxychloroquine, sulfasalazine, leflunomide, azathioprine, gold, and anakinra.
^c Major clinical response is defined as achieving an ACR 70 response for a continuous 6-month period.
^d After 6 months, patients were given the opportunity to enter an open-label study.
^e DAS28-CRP Remission is defined as a DAS28-CRP score <2.6

Greater improvements were seen with abatacept than with placebo in other measures of rheumatoid arthritis disease activity not included in the ACR response criteria, such as morning stiffness.

DAS28 response

Disease activity was also assessed using the Disease Activity Score 28. There was a significant improvement of DAS in Studies II, III, V, and VI as compared to placebo or comparator.

In study VI, which only included adults, a significantly higher proportion of patients in the abatacept plus methotrexate group (41%) achieved DAS28 (CRP)-defined remission (score <2.6) versus the methotrexate plus placebo group (23%) at year 1. The response at year 1 in the abatacept group was maintained through year 2.

In the substudy of study VI, patients who had achieved remission at 2 years (DAS 28 ESR <2.6) and after at least 1 year of treatment with abatacept in Study VI were eligible to enter a substudy. In the substudy 108 subjects were randomized 1:1 in double blinded fashion to receive abatacept at doses approximating 10 mg/kg (ABA 10) or 5 mg/kg (ABA 5). After 1 year of treatment, the maintenance of remission was assessed by the relapse of the disease. The time to and proportion of patients with the relapse of the disease observed between the two groups were similar.

Study V: abatacept or infliximab versus placebo

A randomized, double-blind study was conducted to assess the safety and efficacy of abatacept or infliximab versus placebo in patients with an inadequate response to methotrexate (Study V). The primary outcome was the mean change in disease activity in abatacept- treated patients compared to placebo-treated patients at 6 months with a subsequent double-blind assessment of safety and efficacy of abatacept and infliximab at 12 months. Greater improvement (p < 0.001) in DAS28 was observed with abatacept and with infliximab compared to placebo at six months in the placebo-controlled portion of the trial; the results between the abatacept and infliximab groups were similar. The ACR responses in Study V were consistent with the DAS28 score. Further improvement was observed at 12 months with abatacept. At 6 months, the incidence of AE of infections were 48.1% (75), 52.1% (86), and 51.8% (57) and the incidence of serious AE of infections were 1.3% (2), 4.2% (7), and 2.7% (3) for abatacept, infliximab and placebo groups, respectively. At 12 months, the incidence of AE of infections were 59.6% (93), 68.5% (113), and the incidence of serious AE of infections were 1.9% (3) and 8.5% (14) for abatacept and infliximab groups, respectively. The open label period of the study provided an assessment of the ability of abatacept to maintain efficacy for subjects originally randomized to abatacept and the efficacy response of those subjects who were switched to abatacept following treatment with infliximab. The reduction from baseline in mean DAS28 score at day 365 (-3.06) was maintained through day 729 (-3.34) in those patients who continued with abatacept. In those patients who initially received infliximab and then switched to abatacept, the reduction in the mean DAS28 score from baseline were 3.29 at day 729 and 2.48 at day 365.

Study SC-II: abatacept versus adalimumab

A randomized, single(investigator)-blinded, non-inferiority study was conducted to assess the safety and efficacy of weekly subcutaneous (SC) abatacept without an abatacept intravenous (IV) loading dose versus every-other-weekly subcutaneous adalimumab, both with background MTX, in patients with an inadequate response to methotrexate (Study SC-II). The primary endpoint showed noninferiority (predefined margin of 12%) of ACR 20 response after 12 months of treatment, 64.8% (206/318) for the abatacept SC group and 63.4% (208/328) for the adalimumab SC group; treatment difference was 1.8% [95% confidence interval (CI): -5.6, 9.2], with comparable responses throughout the 24-month period. The respective values for ACR 20 at 24 months were 59.7% (190/318) for the abatacept SC group and 60.1% (197/328) for the adalimumab SC group. The respective values for ACR 50 and ACR 70 at 12 months and 24 months were consistent and similar for abatacept and adalimumab. The adjusted mean changes (standard error; SE) from baseline in DAS28-CRP were -2.35 (SE 0.08) [95% CI: -2.51, -2.19] and -2.33 (SE 0.08) [95% CI: -2.50, -2.17] in the SC abatacept group and the adalimumab group, respectively, at 24 months, with similar changes over time. At 24 months, 50.6% (127/251) [95% CI: 44.4, 56.8] of patients in abatacept and 53.3% (130/244) [95% CI: 47.0, 59.5] of patients in adalimumab groups achieved DAS 28 <2.6. Improvement from baseline as measured by HAQ-DI at 24 months and over time was also similar between abatacept SC and adalimumab SC.

Safety and structural damage assessments were conducted at one and two years. The overall safety profile with respect to adverse events was similar between the two groups over the 24-month period. After 24 months, adverse reactions were reported in 41.5% (132/318) and 50% (164/328) of abatacept and adalimumab-treated patients. Serious adverse reactions were reported in 3.5% (11/318) and 6.1% (20/328) of the respective group. At 24 months, 20.8% (66/318) of patients on abatacept and 25.3% (83/328) on adalimumab had discontinued.

In SC-II, serious infections were reported in 3.8% (12/318) of patients treated with abatacept SC weekly, none of which led to discontinuation and in 5.8% (19/328) of patients treated with adalimumab SC every-other-week, leading to 9 discontinuations in the 24-month period. The frequency of local injection site reactions was 3.8% (12/318) and 9.1% (30/328) at 12 months (p=0.006) and 4.1% (13/318) and 10.4% (34/328) at 24 months for abatacept SC and adalimumab SC, respectively. Over the 2 year study period, 3.8% (12/318) and 1.5% (5/328) patients treated with abatacept SC and adalimumab SC respectively reported autoimmune disorders mild to moderate in severity (e.g., psoriasis, Raynaud’s phenomenon, erythema nodosum).

Study SC-III: Induction of remission in methotrexate-naive RA patients

A randomized and double-blinded study evaluated abatacept SC in combination with methotrexate (abatacept + MTX), abatacept SC monotherapy, or methotrexate monotherapy (MTX group) in induction of remission following 12 months of treatment, and maintenance of drug-free remission after complete drug withdrawal in MTX-naive adult patients with highly active early rheumatoid arthritis with poor prognostic factors. Complete drug withdrawal led to loss of remission (return to disease activity) in all three treatment arms (abatacept with methotrexate, abatacept or methotrexate alone) in a majority of patients (Table 4).

Table 4. Remission Rates at End of Drug Treatment and Drug Withdrawal Phases in Study SC-III:

^a DAS28-defined remission (DAS28-CRP <2.6)
^b SDAI criterion (SDAI ≤3.3)

In SC-III the safety profiles of the three treatment groups (abatacept + MTX, abatacept monotherapy, MTX group) were overall similar. During the 12-month treatment period, adverse reactions were reported in 44.5% (53/119), 41.4% (48/116), and 44.0% (51/116) and serious adverse reactions were reported in 2.5% (3/119), 2.6% (3/116) and 0.9% (1/116) of patients treated in the three treatment groups, respectively. Serious infections were reported in 0.8% (1/119), 3.4% (4/116) and 0% (0/116) patients.

Radiographic response

Structural joint damage was assessed radiographically over a two-year period in Studies II, and VI. The results were measured using the Genant-modified total Sharp score (TSS) and its components, the erosion score and joint space narrowing (JSN) score.

In Study II, the baseline median TSS was 31.7 in abatacept-treated patients and 33.4 in placebo-treated patients. Abatacept/methotrexate reduced the rate of progression of structural damage compared to placebo/methotrexate after 12 months of treatment as shown in Table 5. The rate of progression of structural damage in year 2 was significantly lower than that in year 1 for patients randomized to abatacept (p<0.0001). Subjects entering the long term extension after 1 year of double blind treatment all received abatacept treatment and radiographic progression was investigated through year 5. Data were analyzed in an as-observed analysis using mean change in total score from the previous annual visit. The mean change was, 0.41 and 0.74 from year 1 to year 2 (n=290, 130), 0.37 and 0.68 from year 2 to year 3 (n=293, 130), 0.34 and 0.43 year from 3 to year 4 (n=290, 128) and the change was 0.26 and 0.29 (n=233, 114) from year 4 to year 5 for patients originally randomized to abatacept plus MTX and placebo plus MTX respectively.

Table 5. Mean Radiographic Changes Over 12 Months in Study II:

^a Based on non-parametric analysis.

In Study VI, the mean change in TSS at 12 months was significantly lower in patients treated with abatacept plus methotrexate compared to those treated with methotrexate plus placebo. At 12 months 61% (148/242) of the patients treated with abatacept plus methotrexate and 53% (128/242) of the patients treated with methotrexate plus placebo had no progression (TSS ≤0). The progression of structural damage was lower in patients receiving continuous abatacept plus methotrexate treatment (for 24 months) compared to patients who initially received methotrexate plus placebo (for 12 months) and were switched to abatacept plus methotrexate for the next 12 months. Among the patients who entered the open-label 12 month period, 59% (125/213) of patients receiving continuous abatacept plus methotrexate treatment and 48% (92/192) of patients who initially received methotrexate and switched to combination with abatacept had no progression.

In Study SC-III, structural joint damage was assessed by MRI. The abatacept + MTX group had less progression in structural damage compared with MTX group as reflected by mean treatment difference of the abatacept + MTX group versus MTX group (Table 6).

Table 6. Structural and Inflammatory MRI Assessment in Study SC-III:

* n=119 for Abatacept SC + MTX; n=116 for MTX

Physical function response

Improvement in physical function was measured by the Health Assessment Questionnaire Disability Index (HAQ-DI) in Studies II, III, IV, V, and VI and the modified HAQ-DI in Study I. The results from Studies II, III, and VI are shown in Table 7.

Table 7. Improvement in Physical Function in Controlled Trials:

*** p<0.001, abatacept vs. placebo.
^† p<0.05, abatacept plus MTX vs MTX plus placebo
^a Fixed dose approximating 10 mg/kg (see section 4.2).
^b Concurrent DMARDs included one or more of the following: methotrexate, chloroquine/hydroxychloroquine, sulfasalazine, leflunomide, azathioprine, gold, and anakinra.
^c Health Assessment Questionnaire; 0 = best, 3 = worst; 20 questions; 8 categories: dressing and grooming, arising, eating, walking, hygiene, reach, grip, and activities.
^d Reduction in HAQ-DI of ≥ 0.3 units from baseline.
^e After 6 months, patients were given the opportunity to enter into an open-label study.

In Study II, among patients with clinically meaningful improvement at month 12, 88% retained the response at month 18, and 85% retained the response at month 24. During the open-label periods of Studies I, II, III, and VI the improvement in physical function has been maintained through 7 years, 5 years, 5 years, and 2 years, respectively. In Study SC-III, the proportion of subjects with a HAQ response as a measure of clinically meaningful improvement in physical function (reduction from baseline in HAQ-D1 score of ≥0.3) was greater for the abatacept+ MTX group vs. the MTX group at Month 12 (65.5% vs 44.0%, respectively; treatment difference vs. MTX group of 21.6% [95% CI: 8.3, 34.9]).

Health-related outcomes and quality of life

Health-related quality of life was assessed by the SF-36 questionnaire at 6 months in Studies I, II, and III and at 12 months in Studies I and II. In these studies, clinically and statistically significant improvement was observed in the abatacept group as compared with the placebo group in all 8 domains of the SF-36 (4 physical domains: physical function, role physical, bodily pain, general health; and 4 mental domains: vitality, social function, role emotional, mental health), as well as the Physical Component Summary (PCS) and the Mental Component Summary (MCS). In Study VI, improvement was observed at 12 months in abatacept plus methotrexate group as compared with the methotrexate plus placebo group in both PCS and MCS, and was maintained through 2 years.

Study VII: Safety of abatacept in patients with or without washout of previous TNF-inhibitor therapy

A study of open-label abatacept on a background of nonbiologic DMARDs was conducted in patients with active RA who had an inadequate response to previous (washout for at least 2 months; n=449) or current (no washout period; n=597) TNF-inhibitor therapy (Study VII). The primary outcome, incidence of AEs, SAEs, and discontinuations due to AEs during 6 months of treatment, was similar between those who were previous and current TNF-inhibitor users at enrollment, as was the frequency of serious infections.

Clinical efficacy and safety in adult psoriatic arthritis

The efficacy and safety of abatacept were assessed in two randomized, double-blind, placebocontrolled trials (Studies PsA-I and PsA-II) in adult patients, age 18 years and older. Patients had active PsA (≥3 swollen joints and ≥ 3 tender joints) despite prior treatment with DMARD therapy and had one qualifying psoriatic skin lesion of at least 2 cm in diameter.

In study PsA-I, 170 patients received placebo or abatacept intravenously (IV) on Day 1, 15, 29, and then every 28 days thereafter in a double blind manner for 24 weeks, followed by open-label abatacept 10 mg/kg IV every 28 days. Patients were randomized to receive placebo or abatacept 3 mg/kg, 10 mg/kg, or two doses of 30 mg/kg followed by 10 mg/kg, without escape for 24 weeks, followed by open label abatacept 10 mg/kg monthly IV every month. Patients were allowed to receive stable doses of concomitant methotrexate, low dose corticosteroids (equivalent to ≤10 mg of prednisone) and/or NSAIDs during the trial.

In study PsA-II, 424 patients were randomized 1:1 to receive in a double-blind manner weekly doses of subcutaneous (SC) placebo or abatacept 125 mg without a loading dose for 24 weeks, followed by open-label abatacept 125 mg SC weekly. Patients were allowed to receive stable doses of concomitant methotrexate, sulfasalazine, leflunomide, hydroxychloroquine, low dose corticosteroids (equivalent to ≤10 mg of prednisone) and/or NSAIDs during the trial. Patients who had not achieved at least a 20% improvement from baseline in their swollen and tender joint counts by Week 16 escaped to open-label abatacept 125 mg SC weekly.

The primary endpoint for both PsA-I and PsA-II was the proportion of patients achieving ACR 20 response at Week 24 (Day 169).

Clinical Response

Signs and symptoms

The percent of patients achieving ACR 20, 50, or 70 responses at the recommended abatacept dose in Studies PsA-I (10 mg/kg IV) and PsA-II (125 mg SC) are presented in Table 8 below.

Table 8. Proportion of Patients With ACR Responses at Week 24 in Studies PsA-I and PsA-II:

* p<0.05 vs placebo, p values not assessed for ACR 50 and ACR 70.
^a 37% of patients were previously treated with TNF inhibitor.
^b 61% of patients were previously treated with TNF inhibitor.
^c Patients who had less than 20% improvement in tender or swollen joint counts at Week 16 met escape criteria and were
considered non-responders.

A significantly higher proportion of patients achieved an ACR 20 response after treatment with abatacept 10 mg/kg IV in PsA-I or 125 mg SC in PsA-II compared to placebo at Week 24 in the overall study populations. Higher ACR 20 responses were observed with abatacept vs placebo regardless of prior TNF-inhibitor treatment in both studies. In the smaller study PsA-I, the ACR 20 responses with abatacept 10 mg/kg IV vs placebo in patients who were TNF inhibitor-naive were 55.6% vs 20.0%, respectively, and in patients who were TNF inhibitor-experienced were 30.8% vs 16.7%, respectively. In study PsA-II, the ACR 20 responses with abatacept 125 mg SC vs placebo in patients who were TNF inhibitor-naive were 44.0% vs 22.2%, respectively (21.9 [8.3, 35.6], estimate of difference [95% CI]), and in patients who were TNF inhibitor-experienced were 36.4% vs 22.3%, respectively (14.0 [3.3, 24.8], estimate of difference [95% CI]).

Higher ACR 20 responses in study PsA-II were seen with abatacept 125 mg SC vs. placebo irrespective of concomitant nonbiological DMARD treatment. The ACR 20 responses with abatacept 125 mg SC vs placebo in patients who did not use nonbiological DMARDs were 27.3% vs 12.1%, respectively, (15.15 [1.83, 28.47], estimate of difference [95% CI]), and in patients who had used nonbiological DMARDs were 44.9% vs 26.9%, respectively, (18.00 [7.20, 28.81], estimate of difference [95% CI]). Clinical responses were maintained or continued to improve up to one year in studies PsA-I and PsA-II.

Structural response

In study PsA-II, the proportion of radiographic non-progressors (≤0 change from baseline) in total PsA-modified SHS on x-rays at Week 24 was greater with abatacept 125 mg SC (42.7%) than placebo (32.7%) (10.0 [1.0, 19.1] estimate of difference [95% CI]).

Physical Function Response

In study PsA-I, the proportion of patients with ≥0.30 decrease from baseline in HAQ-DI score was 45.0% with IV abatacept vs 19.0% with placebo (26.1 [6.8, 45.5], estimate of difference [95% CI]) at Week 24. In study PsA-II, the proportion of patients with at least ≥ 0.35 decrease from baseline in HAQ-DI was 31.0% with abatacept vs. 23.7% with placebo (7.2 [-1.1, 15.6], estimate of difference [95% CI]). Improvement in HAQ-DI scores was maintained or improved for up to 1 year with continuing abatacept treatment in both PsA-I and PsA-II studies.

No significant changes in PASI scores with abatacept treatment were seen over the 24-week doubleblind period. Patients entering the two PsA studies had mild to moderate psoriasis with median PASI scores of 8.6 in PsA-I and 4.5 in PsA-II. In study PsA-I, the proportions of patients achieving PASI 50 response was 28.6% with abatacept vs. 14.3% with placebo (14.3 [-15.3, 43.9], estimate of difference [95% CI]), and the proportion of patients who achieved PASI 75 response was 14.3% with abatacept vs. 4.8% with placebo (9.5 [-13.0, 32.0], estimate of difference [95% CI]). In study PsA-II, the proportion of patients who achieved PASI 50 response was 26.7% with abatacept vs. 19.6% with placebo (7.3 [-2.2, 16.7], estimate of difference [95% CI]), and the proportion of patients who achieved PASI 75 response was 16.4% with abatacept vs. 10.1% with placebo (6.4 [-1.3, 14.1], estimate of difference [95% CI]).

Paediatric population in polyarticular juvenile idiopathic arthritis

Children and adolescents with moderate to severe active JIA, ages 6 to 17 years with an inadequate response or intolerance to at least one DMARD, which may have included biologic agents, were enrolled. The safety and efficacy of abatacept were assessed in a three-part study. Period A was a 4- month open-label lead-in designed to induce an ACR Pedi 30 response. Patients achieving at least a ACR Pedi 30 response at the end of Period A were randomized into a double-blind, withdrawal phase (Period B), and received either abatacept or placebo for 6 months or until JIA disease flare as defined in the study. Unless they had discontinued due to safety reasons, all patients who completed, or had a flare during Period B or were non-responders in Period A were offered entry into Period C, the openlabel extension, which assessed long-term safety and efficacy.

In Period A all patients received 10 mg/kg of abatacept on days 1, 15, 29, 57 and 85 and were assessed on day 113. During period A, 74% were taking methotrexate (mean dose at study entry, 13.2 mg/m²/week) thus, 26% of patients received abatacept monotherapy in Period A. Of the 190 patients entering the study, 57 (30%) had previously been treated with TNF-inhibitor therapy.

ACR Pedi 30 responders at the end of Period A were randomized into Period B, the double-blind, withdrawal phase, to receive either abatacept or placebo for 6 months or until JIA flare.

Flare was defined as:

• ≥30% worsening in at least 3 of the 6 polyarticular JIA core set variables
• ≥30% improvement in not more than 1 of the 6 polyarticular JIA core set variables
• ≥2 cm (possible up to 10 cm) of worsening must have been present if the Physician or Parent Global Assessment was used to define flare
• worsening in ≥2 joints must have been present if the number of active joints or joints with limited range of motion was used to define flare

The patients entered in the trial were a mean of 12.4 years of age with mean disease duration of 4.4 years. They had active disease, with baseline mean active joint count of 16 and a mean number of joints with loss of motion of 16; and elevated C-reactive protein (CRP) levels (mean, 3.2 mg/dl) and ESRs (mean, 32 mm/h). Their JIA subtypes at disease onset were: Oligoarticular (16%), Polyarticular (64%; 20% of the total were rheumatoid factor positive), and Systemic (20%).

Of the 190 patients enrolled, 170 completed Period A, 65% (123/190) achieved an ACR Pedi 30 response, and 122 were randomized to Period B. Responses were similar in all subtypes of JIA studied and for patients with or without methotrexate use. Of the 133 (70%) patients with no prior TNFinhibitor therapy, 101 (76%) achieved at least an ACR Pedi 30 response; of the 57 patients who had received prior TNF-inhibitor therapy, 22 (39%) achieved at least an ACR Pedi 30 response.

During Period B, the time to disease flare for the patients randomized to placebo was significantly shorter than for those randomized to abatacept (primary endpoint, p=0.0002; log-rank test). Significantly more placebo recipients flared during Period B (33/62; 53%) than those maintained on abatacept (12/60; 20%; chi-square p<0.001). The risk of disease flare for patients continuing on abatacept was less than one third that for placebo-treated patients (hazard ratio estimate=0.31; 95% CI 0.16, 0.59).

Most randomized Period B patients entered Period C (58/60 Period B abatacept recipients; 59/62 Period B placebo recipients), as did 36 of the 47 Period A non-responders (n=153 total patients).

Response rates at the end of Period A, at the end of Period B and after 5 years exposure in Period C are summarized in Table 9:

Table 9. Proportion (%) of Polyarticular JIA Patients with ACR Responses or Inactive Disease:

^a Day 169 Last Observation Carried Forward (LOCF) for patients treated in Period C
^b As observed

Participants in Period C at day 1765 included 33 of the 58 Period B abatacept recipients, 30 of the 59 Period B placebo recipients, and 13 of the 36 Period A non-responders. The median duration of abatacept treatment in Period C was 1815 days (range 57–2,415 days; nearly 61 months). One hundred and two (67%) of the subjects had received at least 1,080 days (~36 months) of abatacept therapy in Period C. All patients had at least 4 months of prior, open-label abatacept treatment in Period A.

The European Medicines Agency has waived the obligation to submit the results of studies with ORENCIA in all subsets of the paediatric population from birth to less than 18 years of age with Rheumatoid arthritis (see section 4.2 for information on paediatric use).