Computational characterization of Paraphamide B as a potential EGFR-preferential dual EGFR/ALK-binding candidate for non-small cell lung cancer.
Raghuram Amirtesh A
Acquired resistance to selective inhibitors remains a major limitation in EGFR- and ALK-driven non-small cell lung cancer (NSCLC). Dual-target inhibition is a theoretically motivated strategy for addressing compensatory signaling between these pathways, but experimental kinase inhibition and cellular activity remain necessary to establish whether a computationally prioritized compound has biological activity. In this study, a multi-stage computational pipeline was used to identify candidate dual EGFR/ALK-binding molecules from the NPAtlas microbial natural product database, comprising 36,454 compounds. Drug-likeness and structural-alert filtering reduced this chemical space to 3912 compounds for virtual screening. Paraphamide B was prioritized based on its combined docking performance and predicted hinge-region interactions at Met769 of EGFR and Met1199 of ALK, with docking scores of - 9.0 and - 7.8 kcal/mol, respectively. Triplicate 200 ns molecular dynamics simulations were used to assess the reproducibility and persistence of ligand-protein interactions. MM-GBSA calculations yielded more favorable relative binding-energy estimates for EGFR (- 37.43 ± 4.82 kcal/mol) than for ALK (- 16.67 ± 1.55 kcal/mol). Conformational clustering and PCA-derived free-energy landscapes indicated that Paraphamide B-associated complexes sampled relatively confined conformational ensembles comparable to the reference inhibitor-bound systems. In silico ADMET profiling indicated favorable physicochemical and drug-likeness characteristics but also identified predicted CYP inhibition and toxicity liabilities requiring experimental assessment. DFT analysis characterized the electronic properties of Paraphamide B, including its electrophilicity and molecular electrostatic potential, providing physicochemical context for the predicted interaction geometry. Taken together, these results suggest that Paraphamide B may represent a promising computationally identified dual-target hit candidate, with an EGFR-preferential predicted interaction profile, warranting further experimental validation. The online version contains supplementary material available at https://doi.org/10.1007/s40203-026-00777-3.