Abstract

Background: Predicting drug response would be useful but has proven difficult. Two Norwegian cohorts of early or established rheumatoid arthritis (RA) patients included in a Nordic multi-centre study exploring personalized medicine in RA, the NORA/ScandRA consortia, were used to evaluate the predictive value of inflammatory markers as well as clinical assessments for treatment remission. Objectives: To explore calprotectin, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR) and clinical disease activity assessments as predictors of primary treatment remission. Methods: The early RA patients were from the step-up, treat-to-target randomized control trial ARCTIC (1) (230 patients, mean (SD) age 51.5 (13.7) years, disease duration 0.6 (0.5) years, 61% women, 81% anti-CCP positive). The established RA patients (ULRABIT (2)) included 209 patients (mean (SD) age 53.3 (13.2) years, disease duration 10.0 (8.8) years, 81% women, 80% anti-CCP positive) initiating a biologic disease-modifying anti-rheumatic drug. Both cohorts were assessed by clinical (tender/swollen 28 joints (TJC28/SJC28), physician's global assessment (PhGA), patient's global assessment (PGA)), laboratory and comprehensive ultrasound (US) examinations (grey scale and power Doppler of 36 joints). We presently included results from baseline, 1 and 3 months (half of the ARCTIC cohort had US only at baseline and at 12 months). Calprotectin was assessed by a fluoroenzyme immunoassay (FEIA), CRP and ESR by inhouse methods. Calprotectin, CRP, ESR, and clinical examinations (combination of TJC28, SJC28, PhGA and PGA) were assessed at treatment start and at 1 and 3 months, and were explored (using Area Under the Curve (AUC) measures) as potential predictors for remission at 6 and 12 months. Remission was defined as power Doppler sum score = 0 in ARCTIC and ≤3 in ULRABIT, SJC28 sum = 0 in ARCTIC and ≤ 1 in ULRABIT, and CDAI ≤ 2.8. 95% confidence intervals for the AUC were obtained by bootstrapping. Results: In the ARCTIC (Table 1) as well as in the ULRABIT (Table 2) cohorts, models combining clinical assessments and either of the inflammatory markers had higher AUCs than models only containing clinical measurements. When each domain was assessed separately, clinical assessments had generally higher AUC than either of calprotectin, CRP or ESR. Conclusion: Information on calprotectin, CRP or ESR alone during the first three months after treatment start had only limited predictive capacity on primary treatment-induced remission at 6 and 12 months. However, the results of clinical assessments combined with either of calprotectin, CRP or ESR during the first three months of follow-up provides a good indication of a subsequent remission as defined by power Doppler, SJC28 or CDAI at 6 and 12 months. REFERENCES: [1] Haavardsholm EA et al. BMJ. 2016;354:i4205. [2] Hammer HB et al. Arthritis Care Res. 2018;70(5):703-12. Acknowledgements: This project was supported by Vinnova, Innovationsfonden Denmark and The Research Council of Norway, under the frame of Nordforsk (Grant agreement no. 90825, Project NORA) and by The Swedish Research Council, Innovationsfonden Denmark, the Federal Ministry of Education and Research in Germany and The Research Council of Norway, under the frame of ERA PerMed (Project ScandRA). Disclosure of Interests: Hilde Berner Hammer AbbVie, Novartis, Lilly, UCB, Joe Sexton: None declared, Siri Lillegraven: None declared, Linda Mathsson-Alm Linda Mathsson-Alm: Employee of Thermo Fisher Scientific, Isabel Gehring Isabel Gehring: Employee of Thermo Fisher Scientific, Bente Glintborg: None declared, Johan Askling Agreements between Karolinska Institutet (with JA as PI) and Abbvie, BMS, Eli Lilly, Galapagos, MSD, Pfizer, Roche, Samsung Bioepis, Sanofi, mainly for the national safety monitoring of rheumatology immunomodulators in Sweden (ARTIS)., Espen A. Haavardsholm Pfizer, AbbVie, Celgene, Novartis, Janssen, Gilead, Lilly, UCB.