Immunogenicity, Safety, and Efficacy of Abatacept Administered Subcutaneously With or Without Background Methotrexate in Patients With Rheumatoid Arthritis: Results From a Phase III, International, Multicenter, Parallel-Arm, Open-Label Study

Authors

  • Peter Nash,

    Corresponding author
    1. University of Queensland, Brisbane, Queensland, Australia
    • Department of Medicine, University of Queensland, 288 Herston Road, Brisbane, Queensland 4006, Australia

    Search for more papers by this author
    • Dr. Nash has received consultant fees, speaking fees, and/or honoraria (less than $10,000) from Bristol-Myers Squibb.

  • Sauithree Nayiager,

    1. St. Augustine's Hospital, Durban, South Africa
    Search for more papers by this author
  • Mark C. Genovese,

    1. Stanford University, Palo Alto, California
    Search for more papers by this author
    • Dr. Genovese has received consultant fees, speaking fees, and/or honoraria (less than $10,000) from Bristol-Myers Squibb.

  • Alan J. Kivitz,

    1. Altoona Center for Clinical Research, Duncansville, Pennsylvania
    Search for more papers by this author
    • Dr. Kivitz has received consultant fees, speaking fees, and/or honoraria (less than $10,000) from and owns stock or stock options in Bristol-Myers Squibb.

  • Kurt Oelke,

    1. Rheumatic Disease Center, Milwaukee, Wisconsin
    Search for more papers by this author
    • Dr. Oelke has received consultant fees, speaking fees, and/or honoraria (less than $10,000 each) from Pfizer, Amgen, Sanofi-Aventis, URL Pharma, Abbott, and UCB.

  • Charles Ludivico,

    1. East Penn Rheumatology Associates, Bethlehem, Pennsylvania
    Search for more papers by this author
    • Dr. Ludivico has received consultant fees and/or speaking fees (less than $10,000 each) from Bristol-Myers Squibb, Roche, and UCB.

  • William Palmer,

    1. Westroads Medical Group, Omaha, Nebraska
    Search for more papers by this author
    • Dr. Palmer has received consultant fees, speaking fees, and/or honoraria (less than $10,000 each) from Bristol-Myers Squibb and Amgen/Pfizer.

  • Cristian Rodriguez,

    1. Bristol-Myers Squibb, Princeton, New Jersey
    Search for more papers by this author
    • Drs. Rodriguez, Delaet, and Elegbe own stock or stock options in Bristol-Myers Squibb.

  • Ingrid Delaet,

    1. Bristol-Myers Squibb, Princeton, New Jersey
    Search for more papers by this author
    • Drs. Rodriguez, Delaet, and Elegbe own stock or stock options in Bristol-Myers Squibb.

  • Ayanbola Elegbe,

    1. Bristol-Myers Squibb, Princeton, New Jersey
    Search for more papers by this author
    • Drs. Rodriguez, Delaet, and Elegbe own stock or stock options in Bristol-Myers Squibb.

  • Michael Corbo

    1. Bristol-Myers Squibb, Princeton, New Jersey
    Search for more papers by this author
    • Dr. Corbo owns stock or stock options in Bristol-Myers Squibb and has received fees from Bristol-Myers Squibb for patents pending approval.


  • ClinicalTrials.gov identifier: NCT00547521.

Abstract

Objective

To evaluate the impact of concomitant methotrexate (MTX) on subcutaneous (SC) abatacept immunogenicity, and to assess safety and efficacy.

Methods

This phase III, open-label study had a 4-month short-term (ST) period and an ongoing long-term extension (LTE) period. Rheumatoid arthritis patients were stratified to receive SC abatacept (125 mg/week) with (combination) or without MTX (monotherapy), with no intravenous loading dose; patients receiving monotherapy could add MTX in the LTE period. Immunogenicity (percentage of anti-abatacept antibody–positive patients) was assessed. ST and LTE period data are reported, including efficacy through LTE month 14 and safety through LTE month 20.

Results

Ninety-six of 100 enrolled patients completed the ST period; 3.9% (combination) and 4.1% of patients (monotherapy) developed transient immunogenicity, and no patients were antibody positive at month 4. Serious adverse events (SAEs) were reported in 3.9% (combination) and 6.1% of patients (monotherapy); 5.9% (combination) and 8.2% of patients (monotherapy) experienced SC injection reactions, and all were mild in intensity. Mean 28-joint Disease Activity Score (DAS28) changes were −1.67 (95% confidence interval [95% CI] −2.06, −1.28; combination) and −1.94 (95% CI −2.46, −1.42; monotherapy) at month 4. Ninety patients entered and were treated in the LTE period; 83.3% (75 of 90) remained ongoing at month 24. One LTE-treated patient (1.1%) developed immunogenicity, 14.4% of patients experienced SAEs, and no SC injection reactions were reported. For patients entering the LTE period, mean DAS28 changes from baseline were −1.84 (95% CI −2.23, −1.34; combination) and −2.86 (95% CI −3.46, −2.27; monotherapy) at month 18.

Conclusion

SC abatacept did not elicit immunogenicity associated with loss of safety or efficacy, either with or without MTX.

INTRODUCTION

Biologic therapies, administered intravenously (IV) or subcutaneously (SC), can elicit immunogenic responses, resulting in changes in pharmacokinetics, efficacy, and/or safety of the drug in some patients (1–3). These changes occur through reduction in drug level via antibody-mediated clearance and prevention of drug binding to its target by antibody interference (2, 4, 5). Immunogenicity involving IgE antidrug antibodies has been described for tocilizumab and rituximab in patients with rheumatoid arthritis (RA) and correlates with hypersensitivity reactions (6). Concomitant administration of immunosuppressive drugs, such as methotrexate (MTX), reduces immunogenicity and is required for some agents to achieve or maintain clinical efficacy; for example, infliximab and golimumab are indicated for the treatment of RA only in combination with MTX (7, 8). However, in up to 25% of patients, MTX is not tolerated or is contraindicated (9).

Abatacept is approved for use with or without MTX. During clinical trials, IV abatacept administered with or without MTX demonstrated transient, low-titer immunogenicity without clinical consequence or influence on efficacy or safety (2, 10, 11). Significant and sustained efficacy, combined with favorable tolerability, has been demonstrated with long-term IV abatacept across a range of populations (12–15).

SC abatacept was approved for RA treatment in the US in September 2011 (16). Studies have shown that weekly fixed dosing of 125 mg SC abatacept plus MTX provides comparable serum trough concentrations to the approved monthly, weight-tiered (∼10 mg/kg) IV abatacept dose, with patients achieving comparable efficacy and safety measures (17, 18). The ACCOMPANY (Abatacept in Subjects with Rheumatoid Arthritis Administered Plus or Minus Background Methotrexate Subcutaneously) trial assessed the immunogenicity, safety, and efficacy of SC abatacept, both with and without concomitant MTX and in the absence of an initial IV loading dose of abatacept.

Significance & Innovations

  • In clinical practice, a significant number of patients with rheumatoid arthritis (RA) will be intolerant to or opt not to take methotrexate (MTX), and will require treatments that can be administered as monotherapy.

  • The ACCOMPANY (Abatacept in Subjects with Rheumatoid Arthritis Administered Plus or Minus Background Methotrexate Subcutaneously) study of subcutaneously (SC) administered abatacept is the first investigation into the immunogenicity, safety, and efficacy of SC abatacept administered with or without concomitant MTX and in the absence of an initial intravenous loading dose of abatacept.

  • The data included here show that SC abatacept has low immunogenicity and favorable safety and clinical efficacy in patients with RA when administered in the presence or absence of background MTX.

PATIENTS AND METHODS

Patients.

Eligible patients were ages ≥18 years, had RA (by 1987 American College of Rheumatology [ACR] and 1991 ACR [functional class I, II, or III] criteria) (19, 20), and scored >20 on the global assessment of disease activity 100-mm visual analog scale. Exclusion criteria included previous treatment with rituximab, abatacept, live vaccines, or leflunomide nonapproved biologic therapy, or evidence of viral infection or latent bacterial infections.

Study design.

ACCOMPANY was a phase III, international, multicenter, parallel-arm, open-label study (21) consisting of a 4-month short-term (ST) treatment period (to day 113) and a long-term extension (LTE) period. Twenty-two sites participated, including 15 in the US, 3 in Australia, 2 in South Africa, and 2 in Mexico; patients were enrolled by site investigators. Efficacy data are included up to month 18 (4-month ST period plus 14-month LTE period). Safety data are provided up to database lock (4-month ST period plus up to 20-month LTE period). The protocol and patients' informed consent received institutional review board/independent ethics committee approval. The study was conducted in accordance with the Declaration of Helsinki (22) and was consistent with the International Conference on Harmonisation Guidelines for Good Clinical Practice.

The primary objective was to evaluate the immunogenicity to abatacept, with or without MTX and in the absence of an initial IV loading dose of abatacept, over 4 months. Immunogenicity was assessed as the proportion of patients with a positive antibody response to abatacept. Secondary objectives included safety and efficacy. The LTE objectives were to assess the long-term safety of SC abatacept and the maintenance of efficacy in patients remaining on treatment.

Patients were randomized (central randomization system) and stratified 1:1 based on MTX use prior to the study start: those not currently receiving MTX (SC abatacept monotherapy group) or those receiving stable MTX (SC abatacept plus MTX [combination] group). Stable MTX was defined as ≥10 mg for ≥4 weeks. Patients stratified to monotherapy included patients who had never taken MTX who were considered nonresponders to ≥1 disease-modifying antirheumatic drug (DMARD), anti–tumor necrosis factor (anti-TNF) therapy, or anakinra, and patients who discontinued MTX due to a lack of efficacy or tolerability ≥4 weeks prior to the first study dose. Patients stratified to combination therapy continued their current MTX dose. During the ST period, patients self-administered, or received from their caregiver, SC abatacept 125 mg/week, provided as a ready-to-use, 1.0-ml prefilled syringe. Patients did not receive an initial IV loading dose of abatacept. Other DMARDs, biologic or nonbiologic, were discontinued. Per the protocol, patients who missed ≥2 consecutive SC abatacept doses during the ST period were discontinued.

At month 4, patients who completed the ST period could enter the LTE period and continue to receive SC abatacept 125 mg/week. Patients receiving monotherapy were permitted to receive MTX at the investigator's discretion (dose adjustment permitted). The addition of sulfasalazine, chloroquine, hydroxychloroquine, or azathioprine was allowed.

Throughout, patients could receive stable-dose nonsteroidal antiinflammatory drugs (NSAIDs). Low-dose oral corticosteroids (dosage of ≤10 mg/day of prednisone or equivalent) were permitted, except within ≤28 days of day 113, when joint counts were performed. Patients were permitted to receive treatment with 1 course of a high-dose prespecified corticosteroid (per the investigator's discretion). Treatment with analgesics from a predefined list was permitted as needed, except within ≤12 hours of a scheduled joint assessment.

Immunogenicity assessments.

Blood samples for immunogenicity were collected prior to SC abatacept administration on scheduled assessment days and analyzed centrally (Prevalere Life Sciences, Whitesboro, New York, and Tandem Labs, Trenton, New Jersey). Assessment of immunogenicity during the ST period was performed by enzyme-linked immunosorbent assay (ELISA; Prevalere Life Sciences), which detected antibodies to the whole abatacept molecule (both CTLA-4 and Ig) or the CTLA-4 portion alone (CTLA-4T or “tip”). The ELISA included a positive control (affinity purified anti–CTLA-4Ig) to validate the assay.

During the LTE period, immunogenicity was measured using an electrochemiluminescence (ECL) immunoassay (Meso-Scale Discovery instrumentation), which detected antibodies to CTLA-4 and possibly Ig or the IgG and/or junction region, owing to the potential for a polyclonal response against CTLA-4. The ECL was also used to assess immunogenicity during the ST period and was validated as more sensitive and tolerant for abatacept than the ELISA.

Safety assessments.

Adverse events (AEs) and serious AEs (SAEs) were classified using the Medical Dictionary for Regulatory Activities, version 12.1. SC injection reactions (occurring at the site of injection) were recorded; the following SC injection reactions were of special interest: erythema, pain, pruritis, hematoma, and swelling. Clinically relevant AEs probably associated with immunomodulatory drugs, including autoimmune events (prespecified), infections, and malignancies, were described.

Efficacy assessments.

The 28-joint Disease Activity Score using the C-reactive protein level (DAS28-CRP) and the Simplified Disease Activity Index (SDAI) were assessed throughout the study; a clinically meaningful improvement was defined as a DAS28-CRP reduction from baseline of ≥1.2 (23). Low disease activity state (LDAS; defined as a DAS28 ≤3.2 or an SDAI ≤11.0), remission (defined as a DAS28 <2.6 or an SDAI <3.3), and European League Against Rheumatism (EULAR) responses were assessed for patients treated in the LTE period (24–26). Physical function was evaluated by the Health Assessment Questionnaire (HAQ) disability index (DI) throughout the study (27).

Statistical methods.

Baseline demographics and clinical characteristics were analyzed descriptively for patients treated in the ST period and those who entered the LTE period, according to ST treatment group. For the primary study end point (percentage of patients with a positive immunogenicity response based on ELISA at month 4 [day 113] for the combination therapy versus monotherapy groups), point estimates and 95% confidence intervals (95% CIs) were calculated. A continuity-corrected chi-square test was the planned statistical test to compare between groups at the 2-sided 5% significance level. To detect an absolute difference of 30% in the immunogenicity rate between the treatment groups at day 113, a sample size of 100 patients was calculated to yield ∼90% power at the 5% significance level, based on a continuity-corrected chi-square test. This assumes an immunogenicity rate of 5% at day 113 in the combination group, and an overall discontinuation rate of 10% during the ST period.

In the ST period, the immunogenicity analysis included all patients who received ≥1 dose of SC abatacept and had ≥1 immunogenicity assessment, with an immunogenicity sample collected at baseline and on ≥1 visit while receiving treatment. In the LTE period, the immunogenicity analysis included all patients who received ≥1 dose of SC abatacept during the LTE period and had ≥1 LTE immunogenicity assessment. Immunogenicity samples were collected at each study visit of the ST period (days 1, 15, 29, 43, 57, 85, and 113) and at 12-week intervals during the LTE period. For patients who discontinued treatment, followup samples were taken 7, 28, 56, and 85 days after the last SC injection.

Efficacy was evaluated at each study visit in the ST period in all patients who received ≥1 dose of abatacept (n = 100); in the LTE period, efficacy outcomes were assessed quarterly for patients who entered the LTE period and received ≥1 dose of abatacept (n = 90). Efficacy data are shown for patients with available data at the visit of interest (as-observed analysis), according to the original ST treatment group. The mean change (95% CI) in the DAS28, SDAI, and HAQ DI and the proportion (95% CI) with clinically meaningful DAS28 response are shown from baseline to month 4 for all treated patients in the ST period, and from baseline to month 12 for all treated patients in the LTE period. The proportion (95% CI) with LDAS, remission, and EULAR responses is shown over time for patients who entered the LTE period.

Safety is shown for the ST period (n = 100, all treated patients in the ST period) according to treatment group, and for the LTE period (n = 90, all treated patients who entered the LTE period) for the combined population. For patients who discontinued at any time, safety was assessed at 7, 28, 56, and 85 days after the last SC injection. The ST summary includes events that occurred up to the first dose of LTE study medication, or up to 56 days after the last dose of study medication for patients who discontinued in the ST period. The LTE summary includes events up to 56 days after the last dose of study medication for patients who discontinued. Incidence rates (IRs; 95% CI) are shown as the number of patients with an event per 100 patient-years of abatacept exposure.

RESULTS

Patient disposition and baseline demographics.

The study was initiated in December 2007 with an LTE database lock in December 2009. Of 100 patients treated, 96% (n = 96) completed the 4-month ST period; 94% (n = 90) entered the LTE period, with 83% (75 of 90) ongoing at database lock (up to 24 months after study initiation) (Figure 1).

Figure 1.

Patient disposition. * = this patient was discontinued from the study on long-term extension (LTE) day 196 but received their last abatacept dose on day 106, and no reason was given for the discontinuation; ST = short-term; SC = subcutaneous; MTX = methotrexate.

Baseline demographics, disease characteristics, and prior concomitant medications are shown according to the original treatment groups (Table 1). For the ST population, 38.8% (19 of 49) of the monotherapy group had never taken MTX; 57.1% (28 of 49) discontinued MTX ≥4 weeks before day 1, and 4.1% (2 of 49) discontinued MTX ∼3 weeks before day 1. The 2 patients who discontinued MTX <4 weeks before day 1 remained in the study, as they fulfilled prespecified criteria for inclusion in the immunogenicity analysis; the inclusion of these patients was also required based on the intent-to-treat principle. This shorter interval between prior MTX treatment and the start of study medication dosing was recorded as a relevant protocol deviation for these 2 patients. All of the patients in the combination group had received prior treatment with MTX.

Table 1. Baseline demographics and clinical characteristics for patients in the short-term (ST) period and for patients entering the long-term extension (LTE) period according to the original ST period cohort*
 ST periodLTE by ST period cohort
SC abatacept + MTX (n = 51)SC abatacept monotherapy (n = 49)Total population (n = 100)SC abatacept + MTX (n = 47)SC abatacept monotherapy (n = 43)Total population (n = 90)
  • *

    Values are the mean ± SD unless otherwise indicated. SC = subcutaneous; MTX = methotrexate; DAS28-CRP = 28-joint Disease Activity Score using the C-reactive protein level; HAQ = Health Assessment Questionnaire; DI = disability index; RF = rheumatoid factor; NSAIDs = nonsteroidal antiinflammatory drugs.

  • Data are shown according to the original treatment group in the ST period.

  • Patients in the original monotherapy group could receive MTX at the investigator's discretion during the LTE period.

  • §

    Concomitant medications summary at screening/enrollment for the ST period and at time of entry into the LTE period, respectively (nonbiologic disease-modifying antirheumatic drugs [sulfasalazine and cyclosporine] were received by 6.4% and 14.0% of patients and corticosteroids and NSAIDs were received by 61.1% and 74.4% of patients in the original combination and monotherapy groups, respectively).

  • MTX was discontinued ∼3 weeks prior to day 1 in these 2 patients.

  • #

    The patient did not start MTX until day 5 of the LTE period.

  • **

    For conditions other than rheumatoid arthritis.

Age, years55.7 ± 11.552.2 ± 10.254.0 ± 10.955.6 ± 11.551.7 ± 10.353.7 ± 11.1
Weight, kg84.6 ± 18.881.6 ± 22.483.1 ± 20.684.0 ± 18.982.6 ± 22.783.3 ± 20.7
Weight, median (range) kg81.6 (55.2–138.3)77.7 (44.9–142.0)81.2 (44.9–142.0)81.6 (55.2–138.3)81.0 (44.9–142.0)81.5 (44.9–142.0)
Sex, % female66.783.775.066.083.774.4
Race, % white82.469.476.085.169.877.8
Disease duration, years10.5 ± 11.69.6 ± 10.710.1 ± 11.19.7 ± 10.99.5 ± 11.19.6 ± 10.9
Tender joints23.9 ± 18.024.3 ± 14.224.1 ± 16.223.2 ± 17.122.3 ± 12.622.8 ± 15.1
Swollen joints16.2 ± 12.018.3 ± 12.217.2 ± 12.115.1 ± 10.316.9 ± 11.216.0 ± 10.7
Disease activity, DAS28-CRP5.1 ± 1.25.8 ± 1.55.4 ± 1.45.0 ± 1.15.6 ± 1.55.3 ± 1.4
HAQ DI score1.3 ± 0.71.5 ± 0.71.4 ± 0.71.2 ± 0.61.5 ± 0.71.4 ± 0.7
RF positive, %68.665.367.068.165.166.7
CRP level, mg/dl1.2 ± 1.63.0 ± 3.62.0 ± 2.91.0 ± 1.52.6 ± 3.41.8 ± 2.7
Antirheumatic medications, no. (%)§      
 MTX51 (100)2 (4.1)53 (53.0)47 (100.0)1 (2.3)#48 (53.3)
 Corticosteroids (oral and/or injectable)22 (43.1)23 (46.9)45 (45.0)19 (40.4)19 (44.2)38 (42.2)
 NSAIDs35 (68.6)34 (69.4)69 (69.0)32 (68.1)29 (67.4)61 (67.8)
 Chloroquine04 (8.2)4 (4.0)
 Cyclosporine2 (3.9)1 (2.0)3 (3.0)2 (4.3)**1 (2.3)**3 (3.3)**
 Hydroxychloroquine3 (5.9)7 (14.3)10 (10.0)
 Sulfasalazine3 (5.9)4 (8.2)7 (7.0)

Concomitant medication.

The average MTX dosage during the ST period was 16.4 mg/week at day 1 and 16.2 mg/week at month 4. The following medications were received by the combination therapy versus monotherapy groups during the ST period: corticosteroids (43.1% versus 53.1%), NSAIDs (68.6% versus 71.4%), and cyclosporine (3.9% versus 2.0%). The mean ± SD oral corticosteroid doses during the ST period were 3.4 ± 5.7 mg and 3.1 ± 3.6 mg in the combination therapy and monotherapy groups, respectively.

During the LTE period, concomitant MTX was initiated in 23.3% of patients (10 of 43) receiving monotherapy in the ST period; 1 patient receiving combination therapy discontinued MTX. During the LTE period, the mean ± SD doses of oral corticosteroids were 8.5 ± 4.5 mg and 7.4 ± 4.3 mg for the original combination therapy versus monotherapy groups, respectively.

Immunogenicity.

ST period.

A total of 95 patients had immunogenicity data available at day 113 (combination, n = 50; monotherapy, n = 45). No patient in either group was antibody positive based on ELISA at day 113; no statistical test was performed to compare immunogenicity between the treatment groups to reflect this.

Throughout the 4-month ST period, immunogenicity rates were low in both treatment groups (Table 2). For the 4 patients who were antibody positive based on ELISA, all events were transient and associated with low titers; 2 of these patients were from the combination group and 2 patients were from the monotherapy group. Of the 2 patients in the monotherapy group, 1 patient had never taken MTX and 1 patient had discontinued MTX 4 weeks prior to receiving the first abatacept dose. The MTX dosage received by the combination group patients with immunogenicity ranged from 10 mg/week to 17.5 mg/week during the ST period. After day 57, no patient receiving treatment was seropositive for anti-abatacept antibodies, as detected by ELISA. No effect on DAS28 or HAQ DI was seen for the 4 antibody-positive patients. One patient experienced AEs in temporal proximity to a transient positivity episode (mild or moderate paresthesia, mild rash, and moderate worsening of constipation). No further association between immunogenicity and safety was observed, and immunogenicity was not associated with SC injection reactions. Immunogenicity assessed by ECL during the ST period was consistent with ELISA (Table 2).

Table 2. Immunogenicity rates*
 Anti-abatacept, no./total (%)Anti–CTLA-4T, no./total (%)Ig and/or junction region, no./total (%)CTLA-4 and possibly Ig, no./total (%)Total, no./ total (%)
  • *

    ELISA = enzyme-linked immunosorbent assay; ST = short-term; SC = subcutaneous; MTX = methotrexate; ECL = electrochemiluminescence; LTE = long-term extension.

  • The patient was seropositive for anti-abatacept antibodies on days 29 and 43 and anti–CTLA-4 antibodies on day 1 and completed the ST period.

  • The patient was seropositive for anti–CTLA-4 antibodies only on days 29, 43, and 57 and completed the ST period.

  • §

    The patient was seropositive for anti–CTLA-4 antibodies only on day 57 and completed the ST period.

  • The patient was seropositive for anti–CTLA-4 antibodies by ELISA and anti-abatacept antibodies by ECL on posttreatment day 85, and was withdrawn from the study due to a lack of efficacy.

  • #

    The patient had a seropositive response at every assessment during the ST period, including baseline and day 113, with antibody titers similar to or lower than on day 1, suggesting that the seropositivity did not represent a drug-induced seroconversion in this patient.

  • **

    The patient was seropositive for anti-abatacept antibodies on day 15.

ELISA during 4-month ST period     
 SC abatacept + MTX (n = 51)     
  On treatment1/50 (2.0)1/51 (2.0)  2/51 (3.9)
  Posttreatment0/40/4  0/4
  Total1/50 (2.0)1/51 (2.0)  2/51 (3.9)
 SC abatacept monotherapy (n = 49)     
  On treatment0/491/49 (2.0)§  1/49 (2.0)
  Posttreatment0/41/4 (25.0)  1/4 (25.0)
  Total0/492/49 (4.1)  2/49 (4.1)
ECL during 4-month ST period     
 SC abatacept + MTX (n = 51)     
  On treatment  1/51 (2.0)#0/511/51 (2.0)
  Posttreatment  0/40/40/4
  Total  1/51 (2.0)0/511/51 (2.0)
 SC abatacept monotherapy (n = 49)     
  On treatment  0/491/49 (2.0)**1/49 (2.0)
  Posttreatment  0/41/4 (25.0)1/4 (25.0)
  Total  0/492/49 (4.1)2/49 (4.1)
ECL during 14-month LTE period     
 SC abatacept + MTX (n = 47)     
  On treatment  0/450/450/45
  Posttreatment  0/71/7 (14.3)1/7 (14.3)
  Total  0/471/47 (2.1)1/47 (2.1)
 SC abatacept monotherapy (n = 43)     
  On treatment  0/430/430/43
  Posttreatment  0/50/50/5
  Total  0/430/430/43

LTE.

During the LTE period (month 4 to month 18), immunogenicity was assessed only by ECL. One patient, originally stratified to the combination group and receiving 20 mg/day of MTX, demonstrated a seropositive response against abatacept after discontinuation during this time (for CTLA-4 and possibly Ig on days 56 and 85 following discontinuation). This patient had discontinued MTX on day 142 of the LTE period due to lack of efficacy, and had been negative as assessed by ECL throughout the study. No association between immunogenicity and lack of efficacy, AEs, or autoimmune events was determined for this patient.

Safety.

ST period.

A safety summary for the ST period is shown in Table 3. SAEs included Pneumocystis jiroveci pneumonia (PCP) and syncope (both considered severe in intensity) in the combination group, and cartilage injury and lower respiratory infection (both moderate in intensity) and pneumonia (considered severe in intensity) in the monotherapy group. The PCP event occurred on day 114 in a 76-year-old man in the combination group who had a nonserious upper respiratory tract infection (URI) on day 106 (final day of dosing). This patient received MTX (20 mg/week) continuously during the ST period and had a markedly low lymphocyte count throughout the ST period, ranging from 0.170 × 103 cells/μl at baseline to 0.410 × 103 cells/μl on day 29. On day 113, just prior to the PCP event, the patient's lymphocyte level was 0.190 × 103 cells/μl. This SAE resolved after 22 days with treatment; the patient discontinued prior to the LTE period. The PCP, occurring on day 104 in a 64-year-old man, resolved after 36 days with treatment; the patient discontinued as a result of this serious infection. Two other patients discontinued due to AEs in the ST period: 1 due to breast discharge and 1 due to angioedema of the throat and lips.

Table 3. Safety summary*
 ST period (4 months)LTE (up to 20 months of LTE)
SC abatacept + MTX (n = 51), no. (%)SC abatacept monotherapy (n = 49), no. (%)Total group (n = 100), no. (%)Total group (n = 100), IR (95% CI)All SC abatacept treated (n = 90), no. (%)All SC abatacept treated (n = 90), IR (95% CI)
  • *

    ST = short-term; LTE = long-term extension; SC = subcutaneous; MTX = methotrexate; IR = incidence rate; 95% CI = 95% confidence interval; AEs = adverse events; URI = upper respiratory tract infection; UTI = urinary tract infection; ND = not determined; SAEs = serious AEs.

  • AEs listed were reported in ≥5% of patients in any group during any treatment period.

  • Includes patients who discontinued from the study due to an AE up to 56 days after completing the ST period (n = 2 SC abatacept MTX and n = 1 SC abatacept monotherapy).

  • §

    Subset of SAEs.

AEs37 (72.5)32 (65.3)69 (69.0)460.4 (358.2, 582.7)78 (86.7)220.5 (174.3, 275.2)
Sinusitis1 (2.0)2 (4.1)3 (3.0)9.7 (2.0, 28.4)12 (13.3)12.0 (6.2, 21.0)
URI6 (11.8)2 (4.1)8 (8.0)26.6 (11.5, 52.4)11 (12.2)11.0 (5.5, 19.7)
UTI1 (2.0)3 (6.1)4 (4.0)13.0 (3.6, 33.4)11 (12.2)10.7 (5.3, 19.2)
Nasopharyngitis2 (3.9)2 (4.1)4 (4.0)13.2 (3.6, 33.9)10 (11.1)9.7 (4.7, 17.8)
Contusion1 (2.0)2 (4.1)3 (3.0)9.8 (2.0, 28.7)5 (5.6)4.8 (1.6, 11.3)
Tooth abscess0003.2 (0.08, 17.9)5 (5.6)4.8 (1.5, 11.1)
Diarrhea3 (5.9)1 (2.0)4 (4.0)13.1 (3.6, 33.7)3 (3.3)2.8 (0.6, 8.3)
Headache5 (9.8)3 (6.1)8 (8.0)27.4 (11.8, 53.9)2 (2.2)1.9 (0.2, 6.8)
Vomiting04 (8.2)4 (4.0)13.1 (3.6, 33.6)2 (2.2)1.9 (0.2, 6.8)
Cough03 (6.1)3 (3.0)9.8 (2.0, 28.7)2 (2.2)1.9 (0.2, 6.7)
Pyrexia1 (2.0)3 (6.1)4 (4.0)13.2 (3.6, 33.8)00
Injection site pruritus3 (5.9)1 (2.0)4 (4.0)13.2 (3.6, 33.9)00
Discontinuations due to AEs3 (5.9)1 (2.0)4 (4.0)ND4 (4.4)ND
SAEs2 (3.9)3 (6.1)5 (5.0)16.1 (5.2, 37.7)13 (14.4)12.9 (6.9, 22.1)
Deaths000000
Infections and infestations18 (35.3)14 (28.6)32 (32.0)126.3 (86.4, 178.3)54 (60.0)83.1 (62.4, 108.4)
Serious infections§1 (2.0)2 (4.1)3 (3.0)9.7 (2.0, 28.3)4 (4.4)3.8 (1.0, 9.7)
Malignancies00001 (1.1)0.9 (0.02, 5.2)
Autoimmune events00001 (1.1)0.9 (0.02, 5.2)

The most frequently reported infections in the combination and monotherapy groups during the ST period were URI, urinary tract infection (UTI), nasopharyngitis, sinusitis, bronchitis, pneumonia, and vulvovaginal mycotic infection. No cases of tuberculosis were reported.

SC injection reactions were reported in a similar proportion of patients in both groups (Table 4). All of the events were mild in intensity and none led to discontinuation. The majority of reactions occurred between day 1 and day 36. No malignancies, autoimmune events, or deaths were reported during the ST period (Table 3).

Table 4. SC injection reactions*
 ST period (4 months)LTE period (20 months)
SC abatacept + MTX (n = 51)SC abatacept monotherapy (n = 49)Total group (n = 100)All SC abatacept treated (n = 90)
  • *

    Values are the number (percentage). SC = subcutaneous; ST = short-term; LTE = long-term extension; MTX = methotrexate.

Local injection site reactions3 (5.9)4 (8.2)7 (7.0)0
Pruritis3 (5.9)1 (2.0)4 (4.0)0
Pain1 (2.0)1 (2.0)2 (2.0)0
Hematoma01 (2.0)1 (1.0)0
Paresthesia01 (2.0)1 (1.0)0

LTE.

A safety summary for the LTE period (up to 20 months of LTE) is shown in Table 3. A total of 13 (14.4%) of 90 patients experienced 20 SAEs. Two patients discontinued due to AEs (goiter and stomatitis) and 2 discontinued due to SAEs (serious gliosis identified by magnetic resonance imaging and myocardial infarction and cerebrovascular accident).

Infections reported in ≥5% of patients included sinusitis, URI, UTI, nasopharyngitis, and tooth abscess. The 4 serious infections were bronchopneumonia, diverticulitis, pelvic inflammatory disease, and Haemophilus pneumonia.

No SC injection reactions were reported during the LTE period. One malignancy (basal cell carcinoma) and 1 prespecified autoimmune event (autoimmune thyroiditis) were reported during the LTE period, and both were considered mild in intensity (Table 4). No deaths were recorded during the LTE period.

For all abatacept-treated patients, the IR of overall AEs did not increase during the LTE period relative to the ST period. IRs of the most frequently occurring AEs, SAEs, and overall infections and infestations, including serious infections, also did not increase during the LTE period relative to the ST period (Table 3).

Clinical efficacy.

Efficacy over 4 months for patients in the ST period (n = 100).

Patients treated in the ST period in either treatment group experienced reductions in DAS28-assessed disease activity from baseline to month 4 (day 113); reductions in the DAS28-CRP from baseline were observed as early as day 15 (−0.4 [95% CI −0.7, −0.1] and −0.7 [95% CI −0.9, −0.4] for the combination and monotherapy groups, respectively) and continued to month 4 (Figure 2A). SDAI-assessed disease activity also improved to month 4: mean changes from baseline in SDAI score were −17.3 (95% CI −21.9, −12.7) and −22.5 (95% CI −28.2, −16.9) for the combination and monotherapy groups, respectively. At month 4, 62.5% (95% CI 49.2, 75.8) and 66.7% (95% CI 53.5, 79.9) of patients in the combination and monotherapy groups, respectively, demonstrated a clinically meaningful DAS28 improvement. Improvements in the HAQ DI score were observed in both the combination and monotherapy groups (mean changes from baseline to month 4: −0.3 [95% CI −0.4, −0.2] and −0.6 [95% CI −0.7, −0.4], respectively).

Figure 2.

Maintenance of clinical efficacy, disease activity, and remission. A, Mean change from baseline 28-joint Disease Activity Score using the C-reactive protein level (DAS28-CRP) score over 4 months for the short-term (ST) population (n = 100). B, Mean change from baseline in DAS28-CRP score over 18 months for patients entering the long-term extension (LTE) period (n = 90). C, Proportion of patients in DAS28-defined low disease activity state (LDAS) over 18 months for patients entering the LTE period (n = 90). D, Proportion of patients in DAS28-defined remission over 18 months for patients entering the LTE period (n = 90). Error bars show the 95% confidence interval. LDAS is defined as a DAS28-CRP ≤3.2 and remission is defined as a DAS28-CRP <2.6. Data shown are for patients with data available at the visit of interest (as-observed analysis). The vertical broken line indicates the end of the ST period at the end of month 4 and the horizontal broken line (A and B) indicates the threshold of clinically meaningful change in DAS28-CRP score (−1.2). The addition of selected disease-modifying antirheumatic drugs and modification to methotrexate (MTX) dose were permitted during the LTE period. SC = subcutaneous.

Efficacy over 18 months (4-month ST period plus 14-month LTE period) for patients entering the LTE period (n = 90).

For patients entering the LTE period, reductions in disease activity achieved by month 4 were maintained with continued SC abatacept treatment (Figure 2B). Mean changes from baseline in SDAI scores were −18.6 (95% CI −22.9, −14.3) and −23.6 (95% CI −29.2, −17.9) at month 4 and −19.3 (95% CI −24.5, −14.1) and −31.0 (95% CI −38.1, −23.9) at month 18 for the combination and monotherapy groups, respectively. A total of 53.8% (95% CI 38.2, 69.5) and 88.6% (95% CI 78.0, 99.1) of the combination and monotherapy groups, respectively, achieved a clinically meaningful improvement in the DAS28-CRP through to month 18 with continued abatacept treatment.

For patients who entered the LTE period, LDAS, remission, and EULAR responses were evaluated. The proportions of patients achieving DAS28-defined LDAS and remission over time were comparable between the original treatment groups (Figures 2C and D). Similarly, comparable proportions of the combination and monotherapy groups achieved SDAI outcomes: 48.9% and 44.2% achieved LDAS and 6.4% and 11.6% achieved remission at month 4, respectively; 57.5% and 80.6% achieved LDAS and 15.0% and 36.1% achieved remission at month 18, respectively. For patients treated in the LTE period, 76.1% and 81.0% of patients in the combination and monotherapy groups, respectively, achieved a good or moderate EULAR response at month 4; this response was maintained through month 18 in 76.9% and 91.4% of the combination and monotherapy groups, respectively.

For patients entering the LTE period, improvements in physical function achieved in the ST period (mean changes from baseline to month 4: −0.3 [95% CI −0.5, −0.2] and −0.6 [95% CI −0.8, −0.5] for the combination and monotherapy groups, respectively) were maintained with continued abatacept treatment in the LTE period (mean changes from baseline to month 18: −0.35 [95% CI −0.5, −0.2] and −0.7 [95% CI −0.95, −0.5], respectively).

DISCUSSION

This study was designed to assess the immunogenicity to SC abatacept, with and without concomitant MTX and in the absence of an initial IV loading dose of abatacept. Treatment with SC abatacept 125 mg/week resulted in low-titer, transient immunogenicity over 4 months, regardless of whether patients received MTX; ELISA and ECL assays showed similar, low immunogenicity rates. At the end of the 4-month ST period, no patient was seropositive for abatacept or CTLA-4T antibodies and no difference in immunogenicity was observed between the combination and monotherapy groups throughout. For patients who were seropositive while receiving treatment, immunogenicity to SC abatacept was not associated with AEs, loss of efficacy, or discontinuation. These findings are in contrast to immunogenicity with other biologic agents, which can be associated with treatment discontinuation due to inefficacy and failure to sustain disease activity reductions (3, 28, 29). The ACCOMPANY study confirms the lack of clinically meaningful immunogenicity with abatacept, both with and without concomitant MTX and in the absence of an initial IV loading dose of abatacept. These findings support the good long-term retention rate observed with abatacept treatment; the low immunogenicity also correlates with the low frequency and severity of injection site and hypersensitivity reactions seen with abatacept.

Weekly SC abatacept was well tolerated during the ST period, regardless of whether patients received treatment as monotherapy or combination therapy. LTE data showed that SC abatacept was also well tolerated in the longer term. The IRs of overall AEs and SAEs did not increase with increasing SC abatacept exposure. The types of events reported were consistent with the broader experience of SC abatacept and IV abatacept across clinical trials (17, 18, 30–33). The occurrence of serious infections, malignancies, and autoimmune events was low throughout the trial and consistent with previous IV abatacept studies (31, 32).

SC injection reactions have been frequently reported among patients receiving anti-TNF agents in RA treatment, with 10–20% of etanercept-treated patients (during month 1) and 6–12% of adalimumab-treated patients (over 26 weeks) experiencing these AEs (34–36). Prespecified SC injection reactions were few (5.9% and 8.2% for the combination therapy and monotherapy groups, respectively) and mild with weekly SC abatacept and did not lead to discontinuation; no SC injection reactions were reported after month 4.

Improvements in disease activity were observed with SC abatacept treatment to month 4 and maintained throughout the LTE period, regardless of concomitant MTX treatment in the ST period. For patients who entered the LTE period, high proportions of both groups achieved DAS28- and SDAI-defined LDAS and remission by month 4, with outcomes maintained at month 18. Approximately 80% of abatacept-treated patients who entered the LTE period achieved and maintained a good/moderate EULAR response over 18 months regardless of the original treatment group. Similarly, improvements in physical function observed during the first 4 months of treatment were maintained during the LTE period for both original treatment groups. These findings are consistent with previous experience with IV abatacept (31, 32).

The SC regimen utilized in other studies included an IV loading dose of abatacept (∼10 mg/kg, according to weight range) given on day 1 to offset the initial low serum concentrations expected immediately following the start of SC dosing, and to reduce the time to steady-state drug concentrations (17, 18, 30, 33). The ACCOMPANY study omitted the loading dose so the impact of concomitant MTX on immunogenicity to SC abatacept could be clearly observed. Efficacy data shown here indicate that patients already had decreased disease activity at week 2, as measured by the DAS28-CRP, and >60% of patients demonstrated a clinically meaningful DAS28 improvement by month 4, sustained with continued treatment. These data suggest that an IV load may not be necessary to achieve a clinically meaningful response.

The safety and efficacy benefits observed with SC abatacept monotherapy in the absence of an IV loading dose are relevant observations for clinical practice, particularly for those patients who cannot receive MTX because of tolerability issues, and those patients whose medical history precludes administration of IV abatacept (16). These findings are supported by high patient retention in this study, with more than 96% of patients completing the ST period and more than 80% of patients who entered the LTE period remaining on treatment at the time of database lock. This is consistent with observations from clinical trials with IV abatacept across a range of patient populations (14, 32).

The results shown here should be interpreted within the limitations of the study: a relatively small number of patients were treated and these patients were not randomized, but rather stratified to their open-label treatment group according to MTX use. However, the primary end point was the assessment of immunogenicity, determined using an objective laboratory evaluation that was not influenced by prior knowledge of the treatment group. Together with minimal differences in the patient demographics and characteristics at baseline, this reduces the impact of the nonrandomized study design.

In summary, the immunogenicity rate observed with SC abatacept was low irrespective of MTX use and, where it did occur, was transient with a low titer. Immunogenicity rates were similar to IV abatacept with no increase over time and with no impact on SC abatacept safety and efficacy. The SC abatacept safety and efficacy profile observed in this study is consistent with previous experience with IV abatacept and with other SC abatacept studies (17, 18, 30, 33), reflecting the optimization of the SC formulation to maintain comparable immunogenic potential to IV abatacept. These findings indicate that the SC formulation of abatacept provides increased treatment options for RA patients and their physicians.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Nash had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Nayiager, Genovese, Rodriguez, Delaet, Elegbe, Corbo.

Acquisition of data. Nash, Nayiager, Genovese, Kivitz, Oelke, Ludivico, Palmer, Rodriguez, Delaet, Elegbe.

Analysis and interpretation of data. Nash, Genovese, Kivitz, Oelke, Rodriguez, Delaet, Elegbe, Corbo.

ROLE OF THE STUDY SPONSOR

Bristol-Myers Squibb employees (as authors) were involved in the study design, data collection, analysis, review, and approval of the manuscript. Editorial support was provided by Bristol-Myers Squibb. All authors, including those from Bristol-Myers Squibb, approved the content of the submitted manuscript, and submission only occurred on approval from all authors involved.

Acknowledgements

The authors would like to thank Dr. Vicky Reynolds, Medicus International, for editorial assistance.

Ancillary