ned. Sorafenib has likewise been shown to counteract the cytotoxic effects of other DNA damaging agents, such as platinum based chemotherapy. Mechanistically, the cell-autonomous radioprotective effect we observed could be explained by sorafenibmediated reassortment into cell cycle phases of relatively greater radioresistance and a generalized cell cycle AZ-6102 web slowing allowing increased time for sublethal damage repair. Classically, cells were observed to be most radioresistant during S phase and most radiosensitive during G2 and M phases, with intermediate radiosensitivity during G1 phase. More recent data suggest that radiosensitivity varies even within a given phase of the cell cycle, probably in 
part due to fluctuation in activity levels of different DNA damage repair pathways within individual phases. What appears clear from these studies, however, is that cells are most radiosensitive during late G2 or M after they have passed the G2 checkpoint, likely as a result of reduced time to perform sublethal damage repair prior to mitotic catastrophe. Our data demonstrate that sorafenib consistently reduced the number of cells in G2 and M among the three diploid HCC cell lines examined. Sorafenib has been observed to exert a similar effect on the cell cycle in some non-HCC cell lines, which is not unexpected given its well-established inhibition of the proproliferative MAP kinase pathway. Therefore, a possible explanation for the radioprotection observed in vitro is that the proportion of cells in late G2 or M at the time of irradiation was reduced by treatment with sorafenib. In a fashion similar to our in vitro findings, examination of treated HCC tumor xenografts showed that concurrent sorafenibradiation therapy resulted in reduced persistence of double-strand breaks compared to radiation alone. We also observed a lower mitotic index among HCC cells in tumor xenografts following sorafenib treatment in vivo, recapitulating the reduced rate of 10608278 proliferation seen in vitro by cell cycle analysis. These data suggest that the interaction between sorafenib and radiation in HCC cells may be similar in vitro and in vivo. In spite of these observations, no difference in efficacy was seen between concurrent sorafenibradiation and radiation alone when tumor growth delay was measured in vivo. This lack of difference is unexplained by our study. However, one feasible hypothesis is that cell-autonomous radioprotective effects of sorafenib on tumor cells were balanced in vivo by additional anti-cancer non-cell-autonomous 10914735 effects that were unobservable in vitro, resulting in no net difference from radiation alone. Whatever the true mechanistic explanation may be, it remains interesting and instructive for present and future clinical trials that concurrent sorafenib-radiation in vivo was not superior to radiation alone in producing tumor growth delay, despite the fact that each treatment had clear efficacy when administered separately. Using both clonogenic survival assays in vitro and the HCC tumor xenograft model in vivo, we observed sequential radiationsorafenib to be a superior regimen compared to sorafenib alone, radiation alone and concurrent sorafenib-radiation. The results of our Annexin V/propidium iodide flow cytometric assay suggest that increased apoptosis may result from a sequential regimen compared to a concurrent regimen or either agent alone. More investigation, however, is required to elucidate the mechanistic underpinnings at play. In a liv