Pos1046 in anti-ccp positive at risk individuals, which biomarkers change prior to progression to inflammatory arthritis?

Background The pre-clinical phase of inflammatory arthritis (IA) can be prolonged. Although many baseline biomarkers have shown predictive value for progression to IA, their evolution during the disease continuum remains unclear. Objectives To investigate which biomarkers change before progression to IA and assess their predictive value in the next 6 months. Methods In a single centre prospective cohort, 472 at-risk individuals were recruited with a positive anti-cyclic citrullinated peptide antibody (anti-CCP Ab) and a new musculoskeletal symptom, but without clinical arthritis. They were followed prospectively until IA occurred. Sequential data on early morning stiffness (EMS) duration, erythrocyte sedimentation rate (ESR), anti-CCP2 Ab value, rheumatoid factor (RF) value, number of small joints tender on examination, number of joints with power Doppler grade ≥1 (PD≥1), health assessment questionnaire (HAQ), visual analog scale for general health (VAS GH), global pain (VAS GP), and fatigue (VAS fatigue) were analysed. Three time-intervals were investigated within the progression (P) group: less than 6 months (P≤6M), 6 to 12 months (P 6-12M), and over 12 months before progression to IA (P>12M). Paired analysis (Wilcoxon signed rank and Mc-Nemar tests) were used to compare P≤6M and P >12M (Pα ), P≤6M and P6-12M (Pß ) within the P group. Independent analyses (Mann Whitney U and Fisher’s tests) were used to compare the non-progression (NP) and P groups. Time-intervals for NP visits were adjusted to P visits by time elapsed since first visit. Multivariate predictive analysis of change in biomarkers were done using random effects panel data using robust standard errors, adjusted for age and gender. Significant difference was defined by p<0.05. Results Paired analysis within P group showed significant increase in Pα for RF, EMS, ESR, number of small joints tender, HAQ, and VAS GP, and in Pß for EMS and VAS GP (Figure 1). After dichotomisation, significant increases were also found for EMS ≥30min, ≥1 joint PD≥1, VAS GP ≥50mm, ESR positive, and ≥1 small joint tender. Analysis between P and NP groups showed significant differences in anti-CCP2 Ab, RF, ESR, number of joints PD≥1, number of small joints tender on examination, and HAQ at multiple time-intervals (Figure 1). EMS and VAS GP were only different at P≤6M. Multivariate panel data analysis showed predictive value in the change of anti-CCP2 Ab ≥3x upper limit of normal (ULN) (OR 51, p=0.041), positive RF (OR 7, p=0.002), EMS ≥30min (OR 9 p=0.004), and abnormal ESR (OR 7, p=0.046) for progression to IA within 6 months. Conclusion In anti-CCP2 Ab+ at-risk individuals, a change in predictive biomarkers occurs in the 6 months prior to IA development. Whilst EMS and VAS GP only show a late increase, multivariate panel data analysis suggest that anti-CCP2Ab ≥3xULN, positive RF, EMS ≥30min and positive ESR predict IA development within the next 6 months. These observations may be of value for monitoring at-risk individuals. Table 1. Biomarkers characteristics NP group (N=330) P group (N=142) P >6M before IA P ≤6M before IA Anti-CCP2 Ab value ≥3xULN A 412/783 (53%) 170/206 (83%) 67/73 (92%) RF positive A 183/778 (24%) 120/189 (63%) 53/68 (78%) EMS minutes B 5 (0-30) 10 (0-30) 30 (3-67) ESR abnormal A 435/1307 (33%) 150/330 (45%) 67/115 (58%) Nb of small joint tender B 0 (0-2) 0 (0-2) 1.5 (0-5) Nb of joint PD≥1 B 0 (0-0) 0 (0-1) 0 (0-1) HAQ B 0.250 (0.0-0.875) 0.375 (0.0-1.0) 0.562 (0.125-1.250) GH-VAS mm B 22(7-47) 21 (8-47) 28 (10-50) GP-VAS mm B 27 (8-52) 27 (9-51) 35 (12-61) GP-VAS ≥50 A 375/1311 (29%) 100/353 (28%) 49/124 (40%) A n/N (%) B Median (IQR) N= number of visits with available data Figure 1. Biomarkers change through time. Un-paired 95% Confidence interval (95%CI) for the value of each biomarker with time; Progression (P) vs Non Progression (NP) groups. α Significant difference in paired analysis within P group between P>12M and P≥6M. ß Significant difference in paired analysis within P group between P 6-12M and P≥6M. δ Significant difference in independent analysis between P and NP groups. REFERENCES: NIL. Acknowledgements: NIL. Disclosure of Interests Laurence Duquenne: None declared, Kate Harnden: None declared, Leticia Garcia-Montoya: None declared, Andrea Di Matteo: None declared, Rahaymin Chowdhury: None declared, Jacqueline Nam: None declared, Navkiran Sidhu: None declared, Kulveer Mankia Speakers bureau: Abbvie, Lilly, UCB, Grant/research support from: Gilead, Lilly, Paul Emery Speakers bureau: Abbvie, Gilead, Lilly, Novartis, Consultant of: BMS, AbbVie, MSD, Pfizer, Novartis, and Roche, Grant/research support from: AbbVie, BMS, Lilly, Samsung.