Análises in sílico de quinazolinas como moléculas promissoras no desenvolvimento de fármacos para glioblastoma

Abstract

Glioblastoma multiforme (GBM) is a highly aggressive and invasive brain tumor, accounting for about 60% of primary neoplasms in adults. Its biological heterogeneity and the protection conferred by the blood–brain barrier (BBB) hinder therapeutic success, reinforcing the need to identify new drug candidates. In this context, in silico modeling represents an efficient strategy for screening and optimizing bioactive molecules.The aim of this study was to perform computational analyses of quinazoline derivatives to select compounds with potential therapeutic activity against GBM. A total of 57 molecules were evaluated against five protein kinases associated with tumor progression (AKT, FAK, SRC, mTOR, and MAPK-MK2), considering binding energy values and pharmacokinetic properties. Among the compounds with lower binding energy and multitarget profiles, molecules 54, 26, 52, 56, and 09 stood out. Metabolic analysis against cytochrome P450 isoforms revealed relevant differences: molecule 52 showed low metabolic interaction, absence of classification as a substrate or inhibitor of critical isoforms (including CYP3A4), and compliance with Lipinski’s Rules, suggesting greater safety and pharmacokinetic predictability. In contrast, molecules 54, 26, and 09 exhibited complex metabolism and potential risk of drug–drug interactions, while molecule 56 presented an intermediate profile. Taken together, the results demonstrate the importance of integrating molecular docking and pharmacokinetic property evaluation in the rational selection of candidates, highlighting molecule 52 as the most promising for further studies.