Automated docking allowing a "protein-based" alignment was performed on a set of indole inhibitors of the GIIA secreted phospholipase A(2) (GIIA sPLA(2)). A correlation between the binding scores and the experimental inhibitory activity was observed (r(2) = 0.666, N = 34). All the indole inhibitors were docked in the active site of the GIIA sPLA(2) enzyme, and the best score docking pose of each inhibitor was used for the "protein-based" alignment of the compounds. A three-dimensional quantitative structure-activity relationship (3D-QSAR) model was then established using the comparative molecular field analysis (CoMFA) method. The set of 34 indole inhibitors was divided into two subsets: the training set, composed of 26 compounds, and the test set, consisting of eight compounds. The robustness and the predictive ability of the generated CoMFA model were examined by using the test set. A good correlation (r(2) = 0.997) between predicted and experimental inhibitory activity data allows the validation of the CoMFA model. Finally, the generated CoMFA model was used for the design and evaluation of new compounds. The new designed compounds exert improved predicted inhibitory activity and may be a target for the synthesis of new GIIA sPLA(2) indole inhibitors.