Overcoming transarterial chemoembolization resistance through metabolic reprogramming in hepatocellular carcinoma

Abstract

Background: In the United States, death from hepatocellular carcinoma (HCC) has risen 43% in the last 16 years. For patients not amenable to curative therapy, transarterial chemoembolization (TACE) is the locoregional therapy of choice. Despite a survival advantage to best supportive care, 40% of patients’ tumors will have no response to TACE. We hypothesize that TACE resistance is related to preferential oxygen independent glycolysis through the activation of pyruvate kinase muscle isoform 2 (PKM2), the rate limiting enzyme of glycolysis. Additionally, inhibition of PKM2 will induce cellular metabolic reprogramming and in resistant tumors induce the conversion to TACE responder. Methods: Through transcriptomic analysis of multiple patient cohorts a 15 gene signature, TACE Navigator, was identified which can reliably predict an individual's TACE response. Transcriptomic analyses of TACE responder vs. non-responders identified differentially expressed genes associated with both hypoxia and glycolysis pathways. Further bioinformatics analyses revealed PKM2 as the potential driver for TACE resistance. HCC cell lines were characterized as responder and non-responder based on both transcriptomic analyses and in vitro TACE modeling. Functional studies were performed in vitro using immunoblotting, quantitative polymerase chain reaction, glycolysis and hypoxia assays. Results: PKM2 expression was elevated in patients who did not respond to TACE compared to responders (p < 0.0001). Multivariable Cox regression demonstrated high PKM2 expression to be associated with worse overall survival (HR 3.02, CI 1.51–6.04, p < 0.01) and Kaplan-Meier analysis demonstrated decreased median overall survival (27.5 months to undefined, p < 0.0001). Relative mRNA expression of PKM2 was found to be 2.76 times higher in the non-responder (p < 0.01) and this difference increased to 8.03-fold higher in hypoxia as compared to the responder line (p < 0.0001). Additionally, baseline levels of glycolysis were 2.5 times higher in the non-responder (p < 0.01). In vitro TACE modeling demonstrated no change in cell proliferation in the non-responder line when exposed to doxorubicin in hypoxia compared to a 4.47-fold decrease in the responder. After PKM2 knockdown the non-responder resulted in a 20-fold decreased (p < 0.001) in proliferation with lentiviral shRNA and 6.77-fold decrease (p < 0.05) with pharmacologic inhibitor. Conclusion: Elevated PKM2 is associated with treatment resistance and worse overall survival in patients who receive TACE. In vitro data suggest that elevated PKM2 is associated with increased utilization of the glycolysis pathway, resulting in oxygen independent cell metabolism. Through PKM2 knockdown, non-responder cells can be reprogrammed to act as responders. By reprogramming the therapy resistant phenotype, TACE efficacy could be improved in patients.