Background Coll2-1 is a peptide located in the triple helical part of type II collagen and Coll2-1NO2 is the nitrated form. Fib3-2 is a fragment of fibulin-3, an extracellular glycoprotein highly expressed in OA.The levels of these biomarkers have been found to be elevated in serum of OA patients and to vary with severity.
Objectives To investigate soluble osteoarthritis (OA) biomarkers in the per-protocol (PP) population of the double-blind Multicentre Osteoarthritis interVEntion trial with Sysadoa (MOVES) comparing the efficacy and safety of Chondroitin Sulfate (CS) plus Glucosamine Hydrochloride (GH) versus CElecoxib (CE) in patients with knee OA.
Methods Coll2-1, Coll2-1NO2 and Fib3-2 were directly measured by immunassays (ARTIALIS SA, Liège, Belgium) in the serum of the PP population of the MOVES trial including 606 patients with knee OA receiving 400 mg of CS+500 mg of GH three times daily or 200 mg of CE once daily for 6 months.This PP population included 418 subjects (215 receiving CS+GH and 203 receiving CE) with at least one biomarker value at three time points (D0, D120 and D180).
Results Serum biomarkers values at baseline or at follow-up D120 and D180 were not associasted with age, sex, race, weight, height of BMI. In overall PP population, there were no statistically significant differences between CE and CS + GH groups for any of the three biomarkers at any time. However, there was a trend in favor of CS+GH in reducing Coll2-1 at D180 (p=0.069) and a trend in favor of CE to reduce Fib3-2 (p=0.055). When population was stratified according the radiological OA severity, the occurrence of synovitis, the WOMAC score or the sympromatic response to treatment some significant differences in biomarkers serum levels were observed between treatment groups. At D 180, CS +GH induced a significantly greater decrease of Coll2-1 in the subgroups of patients with the more severe radiographic disease (K&L III), with synovitis (at least one joint swelling or effusion event), in OMERACT-OARSI responders or in patients with WOMAC pain at baseline ≤369 compared to CE (p<0.05). For Coll2-1NO2, there was a trend in favor of CS + GH in the sub-population having at least one joint swelling event during the study period (p=0.077) at D180. CE was more effective than CS + GH in reducing Fib3-2 at D180 in patients with WOMAC pain at baseline >369 (p=0.042).
Conclusions CS + GH was more efficient than CE in reducing serum Coll2-1 particularly in a subgroup of patients with severe OA. This data indicates that CS + GH may down-regulate cartilage catabolism and are in accordance with the symptomatic benefits observed in the clinical trials with these therapies.
Disclosure of Interest None declared