Article Text

SAT0474 A Proteomic Panel Predicts Drug Response To The Combination of Glucosamine and Chondroitin Sulfate
  1. V. Kraus1,
  2. H. Martinez2,
  3. M. Herrero2,
  4. J. Verges2
  1. 1Duke Molecular Physiology Institute, Durham, United States
  2. 2Clinical R&D Area, Bioiberica S.A., Barcelona, Spain


Background Biomarkers can help identify patients that are more likely to respond favorably to a given therapy (a Predictive or patient stratification biomarker). Biomarkers can also be used to provide objective proof of a biological response to a drug (Efficacy of Intervention biomarker).

Objectives The overall goal of this project was to identify protein biomarkers that predict likelihood of response to glucosamine hydrochloride and chondroitin sulfate (GH+CS), both predictive and efficacy of intervention biomarkers. We hypothesized that a subset of peptides in a mass spectrometry panel, developed to predict Osteoarthritis (OA) progression, would identify responders to GH+CS, knowing that CS is an NFkB activity and inflammation inhibitor.

Methods This study analyzed a subset of 80 baseline serum samples from the MOVES (Multicentric Osteoarthritis Intervention Study with SYSADOA) trial1. The MOVES trial was an international multicentric, phase IV, double-blind, non-inferiority, randomized trial to compare the efficacy and safety of the combination of glucosamine hydrochloride (GH 1.500mg/day) and chondroitin sulfate (CS 1.200mg/day) known as Droglican® (Bioiberica), versus celecoxib Celebrex (200mg/day) in patients with Kellgren Lawrence grades 2–3 knee OA with severe pain. The sample selected for this study consisted of half drug responders and half non-responders by OARSI-OMERACT criteria. A total of 25 peptides from 23 proteins were analyzed by mass spectrometry. This panel was based on results of discovery proteomic studies in synovial fluid (n=23), urine (n=45) and serum (n=40) from knee OA subjects to identify biomarkers indicative of an OA diagnosis and OA progression.

Quantitative LC/MRM was performed on a 1 μg of protein digest spiked with 10 fmol of 25 stable isotope labeled peptides using a nanoAcquity UPLC system (Waters Corp) coupled to a Waters Xevo TQ-S triple quadrupole mass spectrometer via a nanoelectrospray ionization source. Technical variability (based on spiked in yeast alcohol dehydrogenase was 10.4%). Logistic regression was used to evaluate the association of demographics (age, gender and BMI) and the biomarkers with drug response (yes/no) after 180 days of treatment. P values <0.05 were considered significant.

Results The sample was 90% female, mean age 64 (SD 7) years, 55% with BMI >30 kg/m2. Higher age (p=0.004) but neither gender, nor KL grade were associated with likelihood of drug response. Age plus the marker panel significantly predicted drug response (overall model p=0.0007). In the multivariable logistic regression, adjusted for age, a total of 12 peptides, corresponding to 11 proteins, were significantly independently associated with drug response (p<0.0001 by Effect likelihood Ratio tests).

Conclusions These biomarkers in theory could be used to stratify different patient groups in terms of likelihood of clinical response to GH+CS in efforts to develop personalized treatment algorithms for OA patients.

  1. Hochberg et al. ARD 2015.

Disclosure of Interest None declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.