Structure-based pharmacophore modeling and 3D-QSAR analysis toward the discovery of novel PAK1 inhibitors for the treatment of colorectal cancer.

p21-Activated kinase 1 (PAK1) is frequently overexpressed or excessively activated in a wide range of cancer types, playing a crucial role in promoting tumor cell survival and proliferation. Its amplification is particularly notable in malignancies such as breast, prostate, and colorectal cancers. In this study, we employed structure-based pharmacophore modeling and 3D-QSAR analysis to perform virtual screening of 51,000 compounds from the Maybridge database, aiming to identify novel PAK1 inhibitors for the treatment of colorectal cancer. The five most promising compounds were selected based on their predicted IC50 values and subjected to in vitro antiproliferative assays using prostate cancer (PC-3) and colorectal cancer (HT-29) cell lines. Among the tested compounds, PD00521 demonstrated moderate antiproliferative activity, with IC50 values of 143.31 μM and 85.01 μM against PC-3 and HT-29 cells, respectively, whereas S14362 exhibited selective activity toward HT-29 cells (IC50 = 66.82 μM). Although the potency of these hits remains in the micromolar range, their selective cytotoxicity toward the colorectal cancer cell line, together with their structural novelty compared to known PAK1 inhibitors, supports their potential as preliminary lead scaffolds. Collectively, this study demonstrates the utility of integrating structure-based pharmacophore modeling with 3D-QSAR analysis as a rational strategy for identifying and optimizing novel PAK1-targeted chemotypes.