Progression of osteoarthritis in patients based on a marker of cartilage degeneration and pain can be identified with a proteomic signature
Using a model based on a plasma proteomic signature, US researchers have been able to distinguish between patients with osteoarthritis whose disease is progressive and this model was superior to currently used methods of assessment.
In a global analysis from 2017, it was found that the annual incidence rate of osteoarthritis was 181.2 per 100,000 patients and the authors noted that the burden of disease is increasing in most countries. Currently, there is a need to develop OA prognostic markers and in 2016, researchers identified a method for diagnosing radiographic OA based on a series of serum biomarkers. More recently, a number of systemic biomarkers were identified and deemed to have promise as predictors of both pain and structural worsening of OA. To date, measurement of urinary carboxyl-terminal cross-linked telopeptide of type II collagen (uCTXII) appears to be the strongest predictor of clinically relevant osteoarthritis progression. In the current study, the US team used uCTXII as the ‘best-in-class’ biomarker to evaluate the performance of the proteomic signature they had identified back in 2016, for predicting clinically relevant knee OA progression, defined in terms of both joint structure and pain worsening, over a period of 48 months.
Using a cohort of 596 individuals with OA, the US team set out to measure the effectiveness of their proteomic signature (based on the area under the receiver operating curve, AUC) at distinguishing between progressors and non-progressors.
Osteoarthritis proteomic signature and identification of disease progressors
Data were available for 596 individuals with knee OA with a mean age of 61.6 years (58.7% female) and who at baseline, had moderate to severe radiographic knee OA.
Based on the proteomic signature, containing 13 distinct proteins, the AUC was 73% for differentiating between progressors (based on a measure of both radiographic joint space loss and pain scores). In contrast, the uCTXII model only had an AUC of 58% which was comparable to a model based only on baseline structural OA and the severity of pain (59%).
The researchers went a step further and assessed their proteomic model (but with only 11 proteins) in a second OA cohort and determined an AUC of 70% for distinguishing progressors from non-progressors.
They concluded that their plasma biomarker signature was able to effectively identify clinically relevant knee OA progressors from non-progressors, adding that the proteomic signature may be of value during clinical trials to identify those with the greatest need for treatment.
Citation
Zhou K et al. A “best-in-class” systemic biomarker predictor of clinically relevant knee osteoarthritis structural and pain progression. Sci Adv 2023
