In this proposal, we will confirm of statins’ protective effects against knee OA progression statins in subjects with non-traumatic generalized OA (GOA) with no/minimal subchondral bone marrow lesions (BMLs) as “potential responders” and investigate the mechanism for statin protective effect on subchondral bone using dyslipidemia- associated non-traumatic OA mice models. While there is no universal consensus on its definition, GOA is commonly described as familial with polyarticular involvement and Heberden's nodes (HNs) in hand joints. Epidemiologic studies have shown that dyslipidemia and atherosclerosis are more prevalent among non- traumatic GOA (HN+) patients than non-OA subjects. Atherosclerosis may accelerate the GOA (HN+) through the ischemic effect on subchondral bone. A high-fat diet can also lead to OA-related subchondral bone damage both in humans and mice. Besides, OA-related bone marrow lesions (BMLs) detected by MRI are strongly associated with dyslipidemia. Experimental studies using dyslipidemia mice models (e.g., Apolipoprotein E- deficient (ApoE-/-) mice) have suggested a protective role for statins, first-line lipid-lowering drugs, on subchondral bone. While experimental animal studies have demonstrated a DMOAD role for statins, the overall observational clinical evidence for the protective effects of statins on OA outcomes have been inconsistent, which could be due to: 1) perhaps most importantly, heterogeneous subject selection with regard to OA etiology and baseline stage; 2) insufficient follow-up time; 3) use of only plain radiographs for OA progression rather than MRI. Our recently published results showed that only subjects with the presence of HN as the hallmark of non- traumatic GOA (HN+), but not HN– patients, show the lower hazard of radiographic OA progression compared to statins nonusers. In our further preliminary analysis of semiquantitative MRI BML measurements, only non- traumatic GOA (HN+) statin users with no/minimal BML demonstrated a lower risk of BML worsening at a 2-year follow-up MRI. Based on these results, we hypothesize that statin use has a protective effect on subchondral bone in non-traumatic GOA (HN+) subjects with no/minimal baseline BML, as “potential responders." Using per- protocol and incident user design and propensity score (PS) matching for potential confounding by indication variables (OA and statin indications), a selected subset of OAI participants with non-traumatic GOA (HN+) with no/minimal BMLs will be used. We further investigate the underlying mechanism in dyslipidemia-associated non- traumatic OA mice models through inhibiting subchondral bone vascular defects and associated subchondral bone marrow deformation. We will pursue the following aims: 1) Determine the association between statin use and protection against MRI, serum, urine biomarkers of cartilage loss, and pain worsening in "potential responders." 2) Examine the role of MRI-based OA-related subchondral bone changes as i...