6A). To examine whether ABT-737 has an antitumor effect in the presence of sorafenib, we administered ABT-737 and

sorafenib together to nude mice bearing Rucaparib Huh7 xenograft tumors. Although even sorafenib alone significantly suppressed tumor growth compared with the vehicle alone (Supporting Fig. 3), coadministration of ABT-737 and sorafenib led to significant further suppression of tumor growth compared to administration of sorafenib alone (Fig. 6C). Similar to administration of ABT-737 as a single agent, coadministration of sorafenib and ABT-737 also induced mild thrombocytopenia and ALT elevation (Fig. 6D). However, coadministration did not induce further morbidity or mortality in mice, suggesting that this regimen is safe and effective for controlling HCC progression. Tumor cells have two characteristic features: uncontrolled proliferation and apoptosis resistance. Uncontrolled proliferation, driven by activation of a variety of oncogenes, is directly linked to tumor growth. Apoptosis resistance is believed to be required for the oncogene-induced aberrant proliferation, because without it, tumor cells tend to undergo apoptosis.24 However, the direct link between apoptosis resistance and growth of solid find more tumors in vivo has not been well studied. Clarifying this point is very important, especially because a very recent study by Weber et al.25 produced the

contradictory finding that aged hepatocyte-specific Mcl-1 knockout mice develop HCC-like lesions, suggesting a link between hepatocarcinogenesis and increased proliferation resulting from increased apoptosis. In the present study, we used conditional expression of Bcl-xL in tumor cells to show that Bcl-xL MCE overexpression, which

is frequently found in human HCC, can be directly linked to tumor growth in vivo, although it did not promote significant cell growth in vitro. Our results suggest that tumor cells encounter a variety of cellular stresses and require antiapoptosis to survive in vivo rather than in vitro. Thus, we consider antiapoptosis to be an important mechanism for progression of a solid tumor, even if it may not be the case for tumor development as suggested by Weber et al.25 Our finding also provides proof of the concept that Bcl-xL may be a target for therapy against HCC progression. In the present study, we showed that, unlike hematopoietic malignancy, hepatoma cells are relatively resistant to ABT-737. Although ABT-737 dose-dependently induced apoptosis in hepatoma cells, a relatively high dose of ABT-737 (more than 8 μM) was required to suppress tumor growth in vitro. Importantly, administration of an in vivo effective dose of ABT-737 (50 mg/kg) failed to suppress xenograft tumors. We found increased expression of Mcl-1 in cultured hepatoma cells as well as xenograft tumors upon ABT-737 treatment.