Background Rheumatoid arthritis is a chronic autoimmune disease characterized by progressive inflammation and joint deterioration. Inflammation is driven largely by up-regulation of proteolytic enzymes such as MMPs and results in a greatly increased tissue turnover. Treatment with the αIL-6 receptor agent Tocilizumab (TCZ) results in decreased extracellular matrix (ECM) remodeling. Type VI collagen (Col VI) is found at the interface between the interstitial- and the basement membrane where it binds with both matrix proteins and support cell-cell interactions. The MMP mediated degradation of Col VI can be measured in serum by the protein fingerprint biomarker C6M.
Objectives The aim of the study was to investigate the predictive value of C6M in response to tocilizumab (TCZ) treatment on a methotrexate (MTX) background for 52 weeks in patients with rheumatoid arthritis in the LITHE study.
Methods C6M was measured in the LITHE study (n=740), a one year double blind, placebo controlled phase III parallel group study in patients receiving placebo, 4mg/kg or 8 mg/kg TCZ in combination with MTX. Col VI degradation was measured by the protein fingerprint biomarker C6M in serum at baseline, week 4, 16, 24 and 52. C6M's capacity to predict response to treatment was analyzed by calculating the odds ratio (OR) of ACR20, 50 and 70 criteria at 16 or 52 weeks, given a 20 or 50% reduction in C6M at 4 weeks, independent of treatment type and dose. Results were adjusted for age, sex, BMI, duration of disease and CRP.
Results Treatment with TCZ resulted in an early dose dependent reduction of C6M serum levels in both 4 and 8 mg/kg doses compared to placebo and the reduction persisted throughout the study. In patients responding to treatment after 52 weeks, C6M was significantly decreased in patients with ACR70 (p<0.01) response, throughout the treatment period, while C6M reduction in ACR20 and 50 responders was only significant at week 52 (p<0.05 and p<0.01, respectively). C6M was also significantly reduced in patient with DAS remission (score <2.6 at week 52) and DAS score reduction (>1.2).
To assess the short term predictive value of C6M we calculated the OR for a 20% or 50% reduction in C6M resulting in an ACR response already after 16 weeks. The OR of an ACR response given a 20% reduction in C6M was ACR20) 1.56 (CI 1.14 to 2.13), ACR50) 2.20 (CI 1.49 to 3.24) and ACR70) 2.69 (1.48 to 4.90) respectively, while a 50% reduction gave the OR 1.88 (1.30 to 2.71), 2.49 (1.66 to 3.72) and 1.37 (0.75 to 2.50). Similar findings were seen for week 52 responders.
Conclusions C6M was dose dependently and significantly decreased from week and throughout study in patients receiving either 4 or 8 mg/kg TCZ compared to placebo. Furthermore, C6M was significantly decreased in patients having ACR20, 50 and 70 response as well as DAS28-ESR reduction and remission compared to non-responders. Importantly, changes in C6M after 4 weeks were able to predict response to treatment by ACR score independent of treatment group. The data suggest that and early decrease in ECM turnover following TCZ treatment, measured by the protein fingerprint marker C6M, increases the odds of a positive patient outcome.
Disclosure of Interest C. Thudium Employee of: Nordic Bioscience A/S, N. Gudmann Employee of: Nordic Bioscience A/S, K. Musa Employee of: Nordic Bioscience A/S, M. Karsdal Shareholder of: Nordic Bioscience A/S, Employee of: Nordic Bioscience A/S, A.-C. Bay-Jensen Shareholder of: Nordic Bioscience A/S, Employee of: Nordic Bioscience A/S