Abstract Hepatocellular carcinoma (HCC) is the fifth most common cancer in the world and ranks second in cancer deaths. HCC is associated with chronic liver damage due to viral infection, excess alcohol, and most recently, non-alcoholic hepatitis (NASH), linked to obesity. Despite the recent success of the multikinase inhibitor sorafenib in HCC, for those patients with advanced stage disease the prognosis remains poor. A promising target in the treatment of HCC is mTOR, which is hyperactivated in 50% of HCCs. However, the FDA-approved mTOR-allosteric inhibitor, RAD001 has failed as a single agent in HCC, most likely due to its incomplete suppression of mTOR Complex 1 (mTORC1). Consistent with this finding we recently showed that RAD001 had minimal effects on the proliferation of HCC cells in culture, but when combined with BEZ235, a PI3K/mTOR-ATP site competitive inhibitor, synergistically blocked the growth of HCC tumor cells in culture and caused tumor regression in a mouse model approximating human HCC with bad prognosis (Elnakat et al., 2012). Further studies revealed the down regulation of a number of genes implicated in autophagy, which acts as a tumor suppressor in liver. This led to the finding that RAD001/BEZ235 causes the induction of autophagy in culture and induced mitophagy in tumors. Despite the finding that RAD001/BEZ235 caused tumor regression in vivo, it is argued that a small population of human CD133+ HCC stem-like cells (HSCs) is protected by mTOR inhibitors. In contrast, the biguanides, phenformin and metformin, also potent inhibitors of mTORC1, are reported to selectively suppress the proliferation of human cancer stem cells (CSC). These effects are argued to be through inhibition of mitochondrial oxidative phosphorylation, blocking ATP production, activating AMPK, and inducing autophagy. Given the ability of biguanides to inhibit mTORC1 and to selectively inhibit the proliferation of CSCs, we set out to determine its effects on tumor progression, using human HCC cell lines in a mouse orthotopic model, as compared to the RAD001/BEZ235. In parallel, we analyzed the effect of either drug treatment on whole cell energy consumption and autophagy. In our preliminary studies, similar to our results in the syngeneic mouse model, we found that orthotopic human HCC tumor progression is strongly impaired by RAD001/BEZ235, and also by phenformin as a single agent. RAD001/BEZ235 treated tumors show on-target effects as decreased phosphorylation of RPS6 and 4E-BP1, and lower proliferation indices. Similar on-target effects are observed after phenformin treatment. In vitro, we found that RAD001/BEZ235 inhibits mitochondrial respiration (25-40%), with mass decreasing (10-20%), but that mitochondria conserve their coupled vs uncoupled ATP oxygen consumption ratio, with no apparent compensatory shift towards glycolysis. In contrast, phenformin induces a greater decrease in mitochondrial respiration (80%) and a more rapid loss of mitochondrial mass (30-40%), with only 50% of mitochondrial oxygen consumption being coupled to ATP production, and a stronger induction of ROS. This strong mitochondrial impairment leads to higher glycolitic capacity, but is not sufficient to prevent cell death. Combining mTOR inhibition and mitochondrial complex 1 targeting results in even a greater decrease in mitochondrial function (>80%), with the same uncoupling and mass loss seen in phenformin treatment alone. Finally, using either a cargo-based or flux autophagic assay, implied, unexpectedly, that the benefits of phenformin in suppressing tumor progression are independent of autophagy, but consistent with mitochondrial dysfunction. Note: This abstract was not presented at the conference. Citation Format: Sonia Pereira Da Veiga, Carol Mercer, Ge Xuemei, Hala Elnakat Thomas, Sara Kozma, George Thomas. Dual inhibition of mTOR in hepatocellular carcinoma: Autophagy friend or foe? [abstract]. In: Proceedings of the AACR Special Conference: Targeting the PI3K-mTOR Network in Cancer; Sep 14-17, 2014; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(7 Suppl):Abstract nr A53.