Prediction of Potential Drug-Resistant ALK Kinase Mutations against Fourth-Generation Inhibitors NVL-655 and TPX-0131

Anaplastic lymphoma kinase (ALK) rearrangements have been identified as oncogenic drivers of multiple malignancies. Although at least three generations of ALK inhibitors have been developed, the occurrence of drug-resistant mutations remains a major barrier to drug efficacy. In this study, we develop a multiscale computational pipeline that combines hotspot residue identification, saturation mutagenesis screening, and alchemical free energy calculations to predict the impact of point mutations on the binding affinity of fourth-generation ALK inhibitors NVL-655 and TPX-0131. The hotspot residue analysis reveals that L1122, V1130, V1180, L1196, L1198, M1199, D1203, and L1256 contribute the most to ligand binding. A total of 9 and 13 mutations are identified as drug-resistant candidates with significantly decreased binding free energy of over 3 kcal/mol for NVL-655 and TPX-0131, respectively. Specifically, V1180W, M1199W, and L1256S are the common mutations with decreased binding free energy concerning both inhibitors. These findings highlight important residues and mutations that may impact the clinical efficacy of NVL-655 and TPX-0131, and this pipeline provides an efficient and accurate framework to predict drug-resistant mutations and facilitate the rational design of next-generation ALK inhibitors.