We have been interested in the design, synthesis, and evaluation of novel adenosine A2A agonists. Through the use of comparative molecular field analysis (CoMFA) we have generated a training model that includes 78 structurally diverse A2A agonists and correlated their affinity for isolated rat brain receptors with differences in their structural and electrostatic properties. We validated this model by predicting the activity of a test set that included 24 additional A2A agonists. Our CoMFA model, which incorporates the physiochemical property of dipole and selects against A1 receptor activity, generated a correlated final model (r2 = 0.891) that provides for enhanced A2A selectivity and predictability. Synthesis, pharmacological evaluation, and modeling of four novel ligands further validate the utility and predictive power (r2 = 0.626) of the CoMFA model.