Abstract
Quantitative structure-property relationship (QSPR) modeling has emerged as a pivotal tool in the field of medicinal chemistry and drug design, offering a predictive framework for understanding the correlation between chemical structure and physicochemical properties. Topological indices are mathematical descriptors derived from the molecular graphs that capture structural features and connectivity, playing a crucial role in QSPR analysis by quantitatively relating chemical structures to their physicochemical properties and biological activities. Lung cancer is characterized by its aggressive nature and late-stage diagnosis, often limiting treatment options and significantly impacting patient survival rates. This study focuses on the selection of drugs used to treat lung cancer, including dacomitinib, selpercatinib, tepotinib, trametinib, sotorasib, etoposide, alectinib, paclitaxel, dabrafenib, entrectinib, crizotinib, ceritinib, lorlatinib, afatinib, pralsetinib, brigatinib, erlotinib, adagrasib, gefitinib, vinorelbine, gemcitabine, docetaxel, and pemetrexed. Using molecular structural measures such as degree, neighborhood degree sum, and modified reverse degree, we have developed QSPR models to predict physicochemical properties through the topological indices derived from these structural measures. We then conducted a comparative analysis, incorporating correlation analysis, to identify the model with the highest predictive accuracy.