Pioneering <scp>QSAR</scp> Modeling Study of <scp>FAP</scp> ‐Targeting Radiopharmaceuticals Used in Oncology

Fibroblast Activation Protein (FAP) is highly expressed in the tumor microenvironment, promoting cancer growth and spread. FAP inhibitors (FAPIs) labeled with radionuclides are increasingly used for cancer diagnosis and therapy. The present study aims to explore how structural features relate to the inhibitory action of radiopharmaceuticals, representing a novel approach in the field of radiopharmacy. The 2D-QSAR using multiple linear regression analysis via the stepwise variable selection method showed promising results for both internal and external predictive ability of the model (R² _train = 0.877, Q² _LOO = 0.830, pred_R² = 0.740). This analysis based on the genetic algorithm was also robust (R² _train = 0.846, Q² _LOO = 0.768, pred_R² = 0.608). A 3D-QSAR model using partial least squares analysis showed better parametric results for CoMFA descriptors (R² = 0.988, Q² _LOO = 0.518 and pred_R² = 0.642) than the CoMSIA model as well. Our findings revealed that the steric, hydrophobic, and hydrogen-bonding properties notably impact the pIC₅₀ values of FAPI radiopharmaceuticals. Based on virtual screening on the FDA-approved drugs, 23 potential inhibitors of the FAP enzyme were identified. To the best of our knowledge, this is the first QSAR study on radiopharmaceuticals with FAP inhibitory action, the results of which can be helpful in designing more potent ones.