A prerequisite for the development of a cancer cell selective targeting adenovirus is the generation of adenoviral vectors that lack native receptor binding ability and additionally contain domains redirecting the vector to cancer cell specific receptors. Towards this goal, we have generated an E1B 55kDa-deleted oncolytic and coxoackie and adenovirus receptor (CAR)-binding ablated adenovirus, YKL-K420A. This newly engineered adenovirus resulted in a dramatic reduction of transduction efficiency compared to the control adenovirus, YKL-1, in all of the cell lines tested. The malaria circumsporozoite (CS) protein interacts with glycosaminoglycans (GAG) present on the liver cell surface, and plays a prominent role in sporozoite attachment and invasion into hepatocytes. To redirect the CAR binding ablated adenovirus YKL-K420A to hepatocytes, CS protein epitope (EWSPCSVTCGNGIQVRIK) was incorporated onto the C-terminus of the YKL-K420A fiber protein, generating an YKL-K420A-hepa. The In vitro efficacy and specificity of YKL-K420A-hepa was then evaluated by comparing the cytopathic effect in hepatoma and other cancer cells from different origins. In hepatoma cells, YKL-K420A-hepa exerted upto 20-fold higher cytolytic ability compared to the control adenovirus, YKL-1, in hepatoma cell lines. Treatment with YKL-K420A-hepa also significantly suppressed tumor growth in a hepatoma xenograft tumor model when compared to YKL-1. Taken together, these studies demonstrate that the strategy to alter adenovirus tropism may greatly improve adenoviral utilities in gene therapy applications.