Orosomucoid 2 serves for predicting but not for monitoring drug response in patients with knee osteoarthritis treated with chondroitin sulfate/glucosamine hydrochloride

Purpose: The prediction of drug responses based on the analysis of multiple clinical variables and omics data is mandatory for accomplishing the promise of precision medicine in rheumatology. This will benefit osteoarthritis (OA) patients by matching their proteomics profiles to the most effective therapy available. In the present study we explored potential protein biomarkers useful to predict the therapeutic response of OA patients treated with pharmaceutical grade Chondroitin sulfate plus glucosamine hydrochloride (CS + GH) or the COX-2 selective nonsteroidal anti-inflammatory drug Celecoxib, in order to optimize therapeutic outcomes in OA. Methods: A shotgun proteomic analysis was performed on sera from 80 patients with knee OA enrolled in the Multicentre Osteoarthritis interVEntion trial with Sysadoa (MOVES), employing the iTRAQ labelling technique (Sciex) followed by LC MALDI-MS/MS analysis. The patients were classified as responders and non-responders, either to Droglican or Celecoxib, according to WOMAC pain score and the OMERACT-OARSI criteria recorded after 6 months of treatment. A panel of 4 altered proteins was selected to be validated using commercially available ELISA Kit (Uscn Life Science) in the whole MOVES cohort at baseline. Logistic regression analyses, adjusted by confounder variables recorded during the MOVES trial and reported as significant in previous bivariate approaches, were used to analyze the contribution of the measured proteins to our prediction model of drug response in OA. Appropriate receiver-operating-characteristics (ROC) curves were also calculated. Results: We employed a quantitative proteomic approach to identify differentially expressed proteins between OA responders (R) and non-responders (NR) before starting the pharmacological treatment with chondroitin sulfate/glucosamine hydrochloride or celecoxib. Serum levels of orosomucoid 2 (ORM2), apolipoprotein A2 (APOA2), inter-alpha-trypsin inhibitor heavy chain H1 (ITIH1), and thrombospondin 1 (TSP1) were measured by ELISA assays in 451 OA patients at baseline (n = 229 for CS + GH group and n = 222 for Celecoxib group). As shown in Fig. 1A, in CS + GH group, baseline ORM2 levels were significantly lower in responders compared to non-responders according to the OARSI criteria (R: 192,8 ug/mL vs NR: 261,6 ug/mL; P = 0,042), while no statistically significant differences were found in the Celecoxib group and for the other 3 proteins (data not shown). 7 clinically relevant parameters recorded at baseline significantly influence patients' response regardless of treatment. Notably, if we include baseline ORM2 as covariate, we found a specific interaction between response to CS + GH and baseline protein levels (P = 0,007) thus increasing the power of our prediction model (Fig. 1B). In the CS + GH group, we also measured the levels of ORM2 after 6 months of treatment. We didn't observed any statistically significant differences in the responders group, thus indicating that ORM2 is not a useful biomarker for treatment monitorization. Conclusions: Using ROC curve analysis and prediction modelling, we showed that serum concentration of ORM2 at baseline combined with 7 clinical variables could efficiently predict patients' response to chondroitin sulfate/glucosamine hydrochloride with an accuracy of 84,3%.