Background/Purpose: Early introduction of some biologic (b) DMARDs has been proposed to reduce anti-citrullinated protein antibodies (ACPA) titers in clinical trials settings. However, less information is available about which factors modulate ACPA titers in early arthritis patients under routine clinical practice, including the impact of synthetic (s) DMARDs. Aim: To analyze the factors that are related with the variation of ACPA titers in a seropositive early arthritis population under non-protocolized treatment with s and b DMARDs

Methods: We studied 124 patients enrolled in our prospective PEARL (Princesa early arthritis longitudinal) study. Appropriate ethical approval of the registry protocol and informed consent form from patients were obtained. Sociodemographic, clinical, laboratory, therapeutic variables and biological samples were collected by protocol at baseline, 6, 12, 24 and 60 months. Only baseline ACPA positive patients were include in the study (ELISA antiCCP2 IgG Euro Diagnostica Immunoscan CCPlus®, Arnhem, Holanda; positive >50U/ml). ACPA titers were assessed in 471 visits (3.8 visits/patient). The population was stratified into equal quartiles according to baseline CCP2 titers (median, range): Q1 (87, 50 – 160 U/ml), Q2 (308, 161 – 460 U/ml), Q3 (643, 461 – 1280 U/ml) and Q4 (2000, 1281– 5263) U/ml). Disease activity was assessed with DAS 28 and HUPI indexes. To estimate the effect of different variables (including DMARDs) on the variation of ACPA titers, multivariate linear regression models nested by visit and patients were fitted using the xtgee command of STATA 12.1.

Results: 86 % patients were female, with age at disease onset (median [p25-p75]) 52 [41-73] years, median disease duration at entry 6 [3,6-9] months; 83% of patients met 2010 ACR/EULAR classification criteria for RA at baseline and 93% after two years of follow up. Median DAS28 at baseline was 4,5 [3,5-5,6] and HAQ 1 [0,5-1,6]. No differences in demographic variables, disease duration, activity or functional status were detected between patients clustered in the baseline ACPA quartiles. A significant decrease in baseline ACPA titers can be detected at 6 months (median reduction 48% [15-70]) that was maintained at 1 (44% [12-68]), 2 (49% [15-71] and 5 years (62% [38-72]), (p=0,0002; 0,0049; 0,0001 and 0,0137, respectively). This reduction was also significant in all visits when population were analyzed by quartiles (p=0,0001), with increasing magnitude of change and number of patients with titers reduction in the Q4. In the multivariate analysis, active smoking was associated with higher ACPA titers along the follow-up (p<0,001) while reduction in disease activity was associated with decreasing ACPA titers (p = 0.005). After adjusting for these variables, both synthetic [methotrexate (p = 0.001), leflunomide (p = 0.002), sulfasalazine (p = 0.014)] and biological DMARDs [anti-TNF (P = 0.002), rituximab (p = 0.049)], also account for decrease in ACPA titers as independent factors.

Conclusion: In real clinical practice, an early and sustained reduction in antiCCP2 titers can be detected in seropositive early arthritis patients associated to the decline in disease activity. Both synthetic and biologic DMARDs can also independently explain the decline in ACPA titers.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/anti-citrullinated-protein-antibodies-titers-are-independently-modulated-by-disease-activity-and-synthetic-or-biologic-dmards-in-a-seropositive-early-arthritis-population/