Sorafenib, a multikinase inhibitor, is a new standard treatment for patients with advanced hepatocellular carcinoma (HCC). However, resistance to this regimen is frequently observed in clinical practice, and the molecular basis of this resistance remains largely unknown. Herein, the antitumor activity of sorafenib was assessed in 16 patient-derived xenograft (PDX) models of HCC. Gene expression analysis was conducted to identify factors that promote sorafenib resistance. Quantitative RT-PCR and immunoblotting were used to determine gene expression and activation of signaling pathways. Cell proliferation, clone formation, and transwell assays were conducted to evaluate drug-sensitivity, proliferation, and invasiveness, respectively. Kaplan-Meier analysis was used to evaluate the predictive power of biomarkers for sorafenib response. Differential gene expression analysis suggested that sorafenib resistance correlated with high karyopherin subunit alpha 3 (KPNA3) expression. Overexpression of KPNA3 in HCC cells enhanced tumor cell growth and invasiveness. Interestingly, KPNA3 was found to trigger epithelial-mesenchymal transition (EMT), a key process mediating drug resistance. On a mechanistic level, KPNA3 increased phosphorylation of AKT, which then phosphorylated ERK, and ultimately promoted TWIST expression to induce EMT and sorafenib resistance. Moreover, retrospective analysis revealed that HCC patients with low KPNA3 expression had remarkably longer survival after sorafenib treatment. Finally, we have identified a novel KPNA3-AKT-ERK-TWIST signaling cascade that promotes EMT and mediates sorafenib resistance in HCC. These findings suggest that KPNA3 is a promising biomarker for predicting patient responsiveness to sorafenib. Targeting KPNA3 may also contribute to resolving sorafenib resistance in HCC.
Keywords: drug resistance; epithelial-mesenchymal transition; hepatocellular carcinoma; patient-derived xenograft; personalized medicine.