Discovery of novel FGFR1 inhibitors for oral squamous cell carcinoma using a multi-class QSAR model, virtual screening, and molecular dynamics simulations

Oral squamous cell carcinoma (OSCC) is a highly aggressive cancer with poor prognosis and limited response to conventional therapies. The fibroblast growth factor receptor 1 (FGFR1) has emerged as a pivotal molecular target among the oncogenic drivers of OSCC because of its critical role in tumor cell proliferation, migration, and chemoresistance. This research employed a comprehensive multi-tiered computational drug-discovery approach, integrating multi-class QSAR modeling, virtual screening, and molecular dynamics simulations, to identify novel small-molecule FGFR1 inhibitors with therapeutic potential for OSCC. The ChEMBL database was utilized to create a dataset of 3,222 distinct inhibitors, subsequently categorized into four bioactivity classes. Exploratory data analysis revealed that more potent compounds had a higher average molecular weight, an increased number of hydrogen bond acceptors, a higher count of rotatable bonds, and a higher. The ROS technique was employed on the training set to address the issue of dataset imbalance. We employed 10 distinct machine learning techniques to develop and assess multi-class QSAR models. These models explain how the chemical structures of substances connect to their biological functions mathematically. The Extra Trees (ET) classifier had the best performance, achieving a test set accuracy of 0.926 and MCC of 0.902. This made it the optimal model for our upcoming virtual screening. We employed the validated ET model to examine a repository of FDA-approved drugs and identified high-priority potential drugs. Molecular docking studies in the FGFR1 active site (PDB ID: 6MZW) followed by 200 ns molecular dynamics simulations demonstrated the stability of the top candidates. The study identified two significant lead compounds, CHEMBL155526361 and CHEMBL155529723, exhibiting robust binding affinities and strong interactions. This study provides a robust computational framework and remarkable molecular scaffolds for further preclinical investigation. This will expedite the search for innovative therapeutics for OSCC.