Abstract
Background/Purpose: We sought to identify biomarkers associated with erosive disease and bone loss in an inception rheumatoid arthritis (RA) cohort receiving treat-to-target combination DMARD therapy without oral corticosteroids. Markers of osteoclast activation (RANKL) and inhibition [osteoprotegerin (OPG)], osteoblast inhibition [Dickkopf-1 (Dkk-1), sclerostin (SOST), osteopontin (OPN)], were serially tested. The degree of inflammation (TNFa, IL6, IL1b, ACTH); adipose tissue activity (leptin, insulin) and markers of bone turnover/damage [osteocalcin (OC), parathyroid hormone (PTH), fibroblast growth factor-23 (FGF23)], were also tested.
Methods: Patients with early RA (< 1 year; fulfilling ACR 1987 and/or 2010 classification criteria) received triple therapy (methotrexate, sulfasalazine, hydroxychloroquine) escalated to achieve DAS28 remission. Serum biomarkers were analysed using human bone panel Luminex kits in patients (n=112) at 0, 6, 12 months, and controls (n=33). Responses to treatment were analysed by mixed model regression. Principal component analysis (PCA) was performed on the within-individual correlations between treatment responses. Erosion and femoral neck bone density (BMD) data were available at 0, 1, 2, 3 years, and were analysed using the biomarker log(mean) as a predictor, adjusted for baseline covariates. Erosion progression was analysed using a zero inflated Poisson regression model.
Results: 53% of the cohort were anti-CCP (cyclic-citrullinated peptide) positive; mean (SD) age was 58(14) years, DAS28 5.5(1.3), 73% females, 60% current/past smokers and 20% had erosive disease. Table 1 summarises results. Nine biomarkers changed following therapy: IL6, TNFa, IL1b, RANKL levels were decreased, whereas leptin, SOST, PTH, OC, OPG were increased. IL6 and TNFa represented two different axes of the treatment response (PCA analysis, Figure 1). Anti-CCP positivity (p = 0.009), higher disease activity (p<0.001), OPN (p<0.001), and lower leptin levels (p=0.002) were associated with higher erosion counts, and older age (p=0.001) and higher TNFa (p = 0.022) were associated with a lower probability of remaining erosion-free. Higher TNFa (p=0.003) and lower leptin (p = 0.029) levels were associated with lower BMD.
Conclusion: Biomarkers from both treatment response axes (Figure 1) are associated with erosive disease and bone loss in RA. Low leptin (Component 1) may be a new biomarker for erosive disease and bone loss in RA, and its improvement with treatment was correlated with suppression of IL-6. OPN, which was invariant to treatment, may be a novel biomarker for active erosive disease.
Methods: Patients with early RA (< 1 year; fulfilling ACR 1987 and/or 2010 classification criteria) received triple therapy (methotrexate, sulfasalazine, hydroxychloroquine) escalated to achieve DAS28 remission. Serum biomarkers were analysed using human bone panel Luminex kits in patients (n=112) at 0, 6, 12 months, and controls (n=33). Responses to treatment were analysed by mixed model regression. Principal component analysis (PCA) was performed on the within-individual correlations between treatment responses. Erosion and femoral neck bone density (BMD) data were available at 0, 1, 2, 3 years, and were analysed using the biomarker log(mean) as a predictor, adjusted for baseline covariates. Erosion progression was analysed using a zero inflated Poisson regression model.
Results: 53% of the cohort were anti-CCP (cyclic-citrullinated peptide) positive; mean (SD) age was 58(14) years, DAS28 5.5(1.3), 73% females, 60% current/past smokers and 20% had erosive disease. Table 1 summarises results. Nine biomarkers changed following therapy: IL6, TNFa, IL1b, RANKL levels were decreased, whereas leptin, SOST, PTH, OC, OPG were increased. IL6 and TNFa represented two different axes of the treatment response (PCA analysis, Figure 1). Anti-CCP positivity (p = 0.009), higher disease activity (p<0.001), OPN (p<0.001), and lower leptin levels (p=0.002) were associated with higher erosion counts, and older age (p=0.001) and higher TNFa (p = 0.022) were associated with a lower probability of remaining erosion-free. Higher TNFa (p=0.003) and lower leptin (p = 0.029) levels were associated with lower BMD.
Conclusion: Biomarkers from both treatment response axes (Figure 1) are associated with erosive disease and bone loss in RA. Low leptin (Component 1) may be a new biomarker for erosive disease and bone loss in RA, and its improvement with treatment was correlated with suppression of IL-6. OPN, which was invariant to treatment, may be a novel biomarker for active erosive disease.
Original language | English |
---|---|
Article number | 1899 |
Number of pages | 4 |
Journal | Arthritis & Rheumatology |
Volume | 69 |
Issue number | Supplement 10 |
Publication status | Published - 17 Sept 2017 |
Event | ACR/ARHP Annual Meeting 2017 - San Diego Duration: 3 Nov 2017 → 8 Nov 2017 |
Keywords
- biomarkers and rheumatoid arthritis (RA)
- Bone