Sorafenib is the first-line treatment of advanced hepatocellular carcinoma (HCC). However, there is a lack of validated biomarkers to predict sorafenib sensitivity. In this study we investigated the role of ACSL4, a positive-activating enzyme of ferroptosis, in sorafenib-induced cell death and HCC patient outcome. We showed that ACSL4 protein expression was negatively associated with IC50 values of sorafenib in a panel of HCC cell lines (R = -0.952, P < 0.001). Knockdown of ACSL4 expression by specific siRNA/sgRNA significantly attenuated sorafenib-induced lipid peroxidation and ferroptosis in Huh7 cells, and also rescued sorafenib-induced inhibition of xenograft tumor growth in vivo. We selected 29 HCC patients with surgery as primary treatment and sorafenib as postoperative adjunct therapy from a hospital-based cohort. A high proportion (66.7%) of HCC patients who had complete or partial responses to sorafenib treatment (according to the revised RECIST guideline) had higher ACSL4 expression in the pretreated HCC tissues, compared with those who had stable or progressed tumor growth (23.5%, P = 0.029). Since ACSL4 expression was independent of sorafenib treatment, it could serve as a useful predictive biomarker. Taken together, this study demonstrates that ACSL4 is essential for sorafenib-induced ferroptosis and useful for predicting sorafenib sensitivity in HCC. This study may have important translational impacts in precise treatment of HCC.
Keywords: ACSL4; ferroptosis; hepatocellular carcinoma; predictive biomarker; sorafenib.