Background It is unclear whether anti-cyclic citrullinated peptide (anti-CCP) and rheumatoid factor (RF) status are predictive of response to rheumatoid arthritis (RA) therapy.
Objectives To assess whether baseline anti-CCP/RF status is associated with treatment response in patients with RA initiating abatacept (ABA) or a tumour necrosis factor-alpha inhibitor (TNFi).
Methods Patients who initiated ABA or a TNFi within the Corrona RA registry between June 2002 and January 2015 were eligible if they had: anti-CCP and RF testing at or prior to initiation; a follow-up visit 6 months after initiation (±3 months); and Clinical Disease Activity Index (CDAI) measured at or within 4–6 months prior to initiation and after 6 months' treatment. Seropositivity was identified for anti-CCP (≥20 U/mL) and RF (≥40 U/mL). At 6 months, the primary outcome was ΔCDAI, with secondary outcomes being achievement of remission (CDAI ≤2.8 in those with low, moderate or high baseline disease activity) and low disease activity (LDA; CDAI ≤10 in those with moderate or high baseline disease activity). Unadjusted and adjusted linear and logistic regression analyses were performed by baseline anti-CCP/RF status: double positive (anti-CCP+/RF+), single positive (anti-CCP+/RF– or anti-CCP–/RF+) and double negative (anti-CCP–/RF–). Adjusted models controlled for baseline age, sex, BMI, CDAI score, co-morbidity index and number of prior biologic DMARDs.
Results 566 patients starting ABA (244 double positive; 167 single positive; 155 double negative) and 1715 starting a TNFi (774 double positive; 470 single positive; 471 double negative) were identified. Most patients were female (ABA: 79–83%; TNFi: 74–80%) and middle-aged (mean age 57–58 years in ABA, 54–57 years in TNFi), with established disease (mean disease duration 9–11 years in ABA, 6–8 years in TNFi) and moderate disease activity (mean CDAI 18–23 in ABA, 18–19 in TNFi). For patients who initiated ABA, double positive status was associated with a significantly greater response compared with double negative status on all outcomes (ΔCDAI −8.9 vs −4.5, p=0.002; LDA 43% vs 26%, p=0.002; remission 15% vs 5%, p=0.001). Additionally, single positive status was associated with a greater likelihood of remission as compared with double negative status for ABA users (12% vs 5%, p=0.018). Conversely, there were no significant differences in responses between anti-CCP/RF groups in the TNFi users (double positive vs double negative: ΔCDAI −7.5 vs −6.8, p=0.46; LDA 39% vs 35%, p=0.20; remission 16% vs 14%, p=0.38). In adjusted models, there are significant differences in the outcomes based on anti-CCP/RF status among ABA users but not TNFi users (Figure).
Conclusions These real-world data demonstrate that baseline anti-CCP+/RF+ is associated with better clinical response to ABA, with the strongest association observed for double seropositivity. In contrast, no relationship was evident between anti-CCP/RF status and response to a TNFi, suggesting that the predictive value of autoantibodies is dependent on drug mechanism of action.
Disclosure of Interest L. R. Harrold Grant/research support from: Pfizer, H. J. Litman Employee of: Corrona, S. E. Connolly Shareholder of: Bristol-Myers Squibb, Employee of: Bristol-Myers Squibb, S. Kelly Shareholder of: Bristol-Myers Squibb, Employee of: Bristol-Myers Squibb, W. Hua Employee of: Corrona, E. Alemao Shareholder of: Bristol-Myers Squibb, Employee of: Bristol-Myers Squibb, L. Rosenblatt Shareholder of: Bristol-Myers Squibb, Employee of: Bristol-Myers Squibb, S. Rebello Employee of: Corrona, J. M. Kremer Shareholder of: Corrona, Grant/research support from: AbbVie, Bristol-Myers Squibb, Genentech, Lilly, Novartis, Pfizer, Employee of: Corrona, Speakers bureau: Genentech (non-branded talks only)