Antitumor effects of sorafenib, bevacizumab and cetuximab as single agents or in combination with an MEK, mTOR or bcl-2 inhibitor, in a SNU-398 human hepatocellular tumor xenograft model.

1332 Current therapy for advanced hepatocellular carcinoma (HCC) is still limited. Both surface receptors like EGFR and VEGFR, as well as intracellular molecules can be targeted with new therapies. Early clinical trials in HCC investigating the efficacy of anti-EGFR agents like cetuximab, or anti-angiogenic agents including sorafenib and bevacizumab show single agent activity in HCC. Recent results of a phase III trial of sorafenib vs. placebo have shown an increase in time-to-progression of 3 months which granted approval by the FDA and EMEA. A phase II trial of the MEK inhibitor AZD6244 in HCC is ongoing. The combination of targeted therapies has been shown to be feasible in phase I studies and have entered phase II testing in many solid tumors.
 The goal of this study was to test different combinations of approved and novel targeted therapies, using both a rational and empirical approach. We combined inhibitors of the surface receptors EGFR (cetuximab) and VEGFR (sorafenib, or bevacizumab) with either an apoptosis activator (the bcl-2 inhibitor GX15-070) or with downstream signaling inhibitors (the mTOR inhibitor temsirolimus, or the MEK inhibitor AZD6244) in the human SNU-398 human HCC model.
 Measured endpoints for these studies included tumor growth inhibition (TGI), partial or complete tumor regression, and agent toxicity as measured by weight loss, deaths, and gross observation.
 Significant single agent activity was observed with sorafenib (42% TGI), suggesting some correlation with clinical observations. Combined inhibition of the EGF or VEGF receptor with mTOR or MEK inhibition produced significant additive or synergistic effects. Sorafenib combined with either AZD6244 or temsirolimus respectively produced a 69% and 79% TGI and were considered the most active in this model. No drug-related deaths were observed and no significant weight loss was seen in any combination groups when compared to single agent treatment. Additional pharmacodynamic studies evaluating the effect of these combinations on the intracellular feedback loops are underway.
 Overall, these studies confirms SNU-398 as a useful model for testing of potential HCC treatments. In addition, these data strongly support further investigation of the combination of sorafenib with either AZD6244 or temsirolimus in HCC. AZD6244 and GX15-070 were obtained through collaboration with the NCI-CTEP.