Background There is a need for reliable prognostic biomarkers in rheumatoid arthritis (RA) for identification of patients in urgent need of aggressive treatment. Serological assessment of type I collagen degradation (C1M) was demonstrated to be a biomarker for identification of RA disease progression in the phase 31 LITHE study. In the LITHE study C1M predicted JSN progression when corrected for confounders (including CRP and baseline level of joint space narrowing; JSN).
Objectives To support serum C1M as a biomarker for RA structural progression.
Methods Placebo groups were pooled (n=543) from the 24-week Phase 3, randomized, double-blind, placebo-controlled, parallel group studies of fostamatinib on a MTX background, OSKIRA 1, 2 and 3 (NCT01197521; NCT01197534; NCT01197755). Patients who did not achieve ACR20 at week 12 were given active drug and were considered escapers. Radiographic progression was assessed by increase in JSN, erosion (ERN) or modified total sharp score (mTSS) after 24 weeks. Only 270 patients had radiographic assessment at both baseline and at 24 weeks. C1M was assessed in fasting serum at baseline (BL). One-way ANOVA was used to assess differences in BL demographics. Associations between BL demographics were investigated by Spearman's correlation. Multiple regression included age group, gender, disease duration, BMI, CRP and BL radiographic status. Non-normalized data was log-transformed.
Results BL demographics of the 3 studies were comparable. 36% of patients were escapers, having BL elevated levels of C1M, CRP and disease activity compared to non-escapers (p=0.03, p<0.0001, p=0.006). 20% of all patients had radiographic progression by JSN, 24% by ERN and 31% by mTSS. BL CRP was significantly elevated in progressors compared to non-progressors (JSN p=0.02, mTSS p=0.005, ERN p=0.005). ERN progressors had a significantly increased level of C1M than non-progressors (p=0.02). JSN only progressors had significant increases in BL radiographic status compared to non-progressors (JSN p=0.003, mTSS p=0.007, ERN p=0.03).
At BL there was an association between C1M and radiographic status (JSN rho=0.14, mTSS rho=0.14, ERN rho=0.13,) and disease activity (DAS28 rho=0.44). In multiple regression the effect size β of C1M alone was -0.25, 0.90 and 1.18 for the association with the 24 week change in radiographic progression as determined by JSN, mTSS and ERN, respectively. Only for ERN the effect size reached statistical significance (p=0.0008). The beta value describes the effect of C1M on predicting the radiographic progression. The effect size decreased for JSN when adjusted for CRP (β=-0.68), but for both mTSS and ERN the effect was eliminated. When the other confounders were included the effect size decreased even further for JSN (β=-0.70).
Conclusions In this study we found that BL C1M was associated with structural status and disease activity and that BL C1M was a predictor of radiographic progression (JSN) after 24 weeks when adjusted for confounders. Secondly, escapers had higher level of C1M, CRP and disease activity indication a worse disease stage in escapers at baseline. This study provides supports C1M as an important emerging biomarker of progression in RA.
Arthritis Res Ther. 2013 Aug 14;15(4):R86.
Disclosure of Interest A. S. Siebuhr Employee of: NORDIC BIOSCIENCE A/S, A. Platt: None declared, M. Karsdal Shareholder of: NORDIC BIOSCIENCE A/S, Employee of: NORDIC BIOSCIENCE A/S, A.-C. Bay-Jensen Employee of: NORDIC BIOSCIENCE A/S