Animal experimentation is a prerequisite for preclinical evaluation of treatments such as chemotherapy. It's strictly regulated with the purpose of reducing the number of experimental animal as well as their pain. Small animal imaging should provide a painless longitudinal follow up of tumor progression on a single animal. The aim of the study is to validate small animal imaging by microscanner (μscan) in longitudinal follow up of a hepatocellular carcinoma (HCC) and to demonstrate its interest for in vivo evaluation of tumor response to different therapeutics. An HCC model achieved by orthotopic graft of the MH3924A cell line in ACI rats was followed using a Imtek/Siemens microscanner (μscan) with contrast agents (Fenestra(®) LC/VC). The procedures giving the optimal enhancement of the liver as well as a reliable determination of tumor volumes by μscan were validated. Three protocols for therapeutic assessment through μscan longitudinal follow up were performed. Each consisted in three groups testing a chemotherapy (gemcitabine, gemcitabine-oxaliplatine or sorafenib) versus two control groups (placebo and doxorubicine). Comparison was done on tumor volumes, median and actual survivals. There was a significant correlation between tumor volumes measured by μscan and autopsy. Treatment by sorafenib, at the contrary of gemcitabine alone or with oxaliplatine, resulted in a significant reduction in tumor volumes and prolongation of actuarial survival. These results are consistent with available clinical data for these diverse therapeutics. In conclusion, small animal imaging with μscan is a non-invasive, reliable, and reproducible method for preclinical evaluation of antitumor agents.