Background Anti-citrullinated protein antibodies (ACPAs) are markers of RA and emerging evidence suggests they may play a role in disease progression. Analysis of biomarkers from AMPLE provides a unique opportunity to probe the differential effects of treatment with biologic DMARDs with distinct mechanisms of action (MoA).1 Here we characterize the changes in ACPA profiles over time and evaluate the relationship between ACPA and clinical outcomes.
Objectives To assess the relationship between changes in ACPA and disease activity in patients (pts) treated with either abatacept (ABA) or adalimumab (ADA).
Methods Pts in AMPLE were MTX failures but naïve to biologic DMARDs.1 Anti-CCP2+ status was determined based on a commercial anti-CCP2 ELISA established cut-off. ACPA analysis included 20 specificities and was performed using a custom assay.2 The anti-CCP2 ELISA was performed at baseline (BL) while the ACPA fine specificities were measured at BL, 85, 365 and 729 days. DAS28 (CRP) and ACPA changes were compared using an ANCOVA model with treatment as a factor and BL values and DAS28 (CRP) strata as covariates. Comparisons of pts who achieved or failed to achieve a Major Clinical Response at Day 729 (MCR729) were also made.
Results Of the available pt BL serum measures, 185/251 (74%) were anti-CCP2+ in the ABA arm and 203/257 (79%) were positive in the ADA arm. For both treatments, BL anti-CCP2+ pts showed greater improvements in both DAS28 (CRP) and HAQ than BL negative pts. For the BL anti-CCP2+ pts, mean change (95% CI) from BL in DAS28 (CRP) at Day 729 was −2.82 (−3.03, −2.62) and −2.72 (−2.92, −2.53) for ABA- and ADA-treated pts, respectively. For BL negative pts, mean change (95% CI) from BL in DAS28 (CRP) at Day 729 was −2.26 (−2.61, −1.90) and −2.13 (−2.52, −1.75) for ABA- and ADA-treated pts, respectively. HAQ showed a similar pattern, with a mean change (95% CI) at Day 729 in the ABA arm of −0.78 (−0.89, −0.68) for anti-CCP2+ pts and −0.59 (−0.76, −0.41) for negative pts. In the ADA arm, respective values were −0.82 (−0.91, −0.72) and −0.49 (−0.69, −0.29). Reductions in specific ACPAs were observed in both treatment arms over the 2-year study, independent of clinical response, although they followed different patterns (Figure). Certain ACPAs had a greater mean reduction from BL in ABA- than ADA-treated pts, most notably ACPAs against apolipoprotein E and histones 2A and 2B during Year 2 of treatment. In pts with MCR729, ABA treatment produced a continued decline in the median levels of most ACPAs beyond Year 1 of treatment (Figure – left). The median levels of most ACPAs rebounded after Year 1 in ADA-treated pts who reached MCR729 (Figure – right).
Conclusions AMPLE provided an opportunity to evaluate the molecular differences in RA pathology altered by treatment with two biologics with distinct MoA. Both agents resulted in a greater response in anti-CCP2+ pts and impacted the overall pattern of ACPA fine specificities. Among patients with similar sustained clinical response, abatacept and adalimumab induced different impacts on ACPA over time suggesting different effects on adaptive immunity.3
Schiff M, et al. Ann Rheum Dis 2014;73:86–94; 2. Sokolove J, et al. PLoS ONE 2012;7:e35296. 3. Cope A, et al. J Clin Inves 1994;94:749–760
Disclosure of Interest S. Connolly Shareholder of: BMS, Employee of: BMS, M. Maldonado Shareholder of: BMS, Employee of: BMS, M. Schiff Consultant for: Bristol-Myers Squibb, Abbvie, Speakers bureau: Bristol-Myers Squibb, Abbvie, M. Weinblatt Grant/research support: BMS, Crescendo Bioscience, UCB, Consultant for: BMS, Crescendo Bioscience, UCB, AbbVie, Roche, Janssen, R. Fleischmann Grant/research support: AbbVie, Amgen, Astellas, Astra Zeneca, BMS, Celgene, Dynavax, Genzyme, Janssen, Eli Lilly, Merck, Novartis, Pfizer, Roche, Sanofi Aventis, UCB, Xoma, Consultant for: AbbVie, Amgen, Astra Zeneca, BMS, Celgene, Janssen, Eli Lilly, Pfizer, Roche, Sanofi Aventis, UCB, W. Robinson: None declared, J. Sokolove Grant/research support: Lab receives research support from BMS