Knee osteoarthritis (KOA) is a serious disease and a leading cause of disability. Genicular artery embolization (GAE) is emerging as a minimally invasive option for refractory pain treatment; determining the optimal patient profile for this intervention continues to be a primary focus of current clinical trials. Current patient selection often relies on self-reported symptom scores and radiographs, which fail to capture the biological heterogeneity of KOA. The purpose of this review is to summarize imaging and biochemical markers that may optimize patient selection and predict outcomes for GAE. Magnetic resonance imaging-based biomarkers enable detailed assessment of bone marrow lesions and inflammatory burden. Phenotypes characterized by mild-to-moderate synovitis are associated with improved response to GAE. Biochemical markers such as CTX-II and hyaluronic acid reflect overall disease activity in osteoarthritis, while inflammatory mediators and angiogenic factors are linked to synovial inflammation and pain mechanisms. Post-GAE reductions in circulating vascular endothelial growth factor and interleukin-1 receptor antagonist were observed alongside sustained pain relief, suggesting that angiogenic and inflammatory processes may contribute to symptom improvement after embolization, although this evidence is currently limited and requires further validation. Integrating imaging phenotypes with molecular endotypes can offer a more precise framework for selecting GAE candidates and predicting outcomes.