Although the overview above provides a partial molecular picture of the early stages of stepwise hepatocarcinogenesis. it should be emphasized that tumor and nontumor liver contain multiple changes, and that there is variability in their profile among different patients even within single studies. Variability in the number and types of genetic changes has also been observed geographically, and may be dependent upon the etiology of the tumor (viral, chemical or both). Interestingly, HBxAg inactivates tumor suppressors (such as p53 [by direct binding] and Rb [by stimulating its phosphorylation]) early in carcinogenesis that are mutated later during tumor progression. HBxAg also constitutively activates signal transduction pathways, such as those involving c-jun and ras, and activates oncogenes,such as c-nloc, that are otherwise activated by 3-catenin mutations. These findings suggest common molecular targets in hepatocarcinogenesis, despite different mechanisms of activation or inactivation. These observations need to be exploited in future drug discovery and in the development of new therapeutics. Heterogeneity in the mechanisms of tumor development, evidenced by the differences in the up- and down regulated genes reported in micro array analyses, as well as in the genetic loci that undergo mutation or LOH indifferent reports, has now been well documented. This suggests that there are multiple pathways to HCC, and that there is redundancy in the pathways that regulate cell growth and survival. These findings also reflect that,although hepatocarcinogenesis is multistep, the molecular changes that underpin histopathological changes in tumor development are likely to be different or only partially overlapping in individual tumors. Overall, the consequences of these changes suggest that the pathogenesis of HCC is accompanied by a progressive loss of differentiation, loss of normal cell adhesion, loss of the ECM, and constitutive activation of selected signal transduction pathways that promote cell growth and survival. Although mechanisms are important, attention also has to be paid to the target genes whose altered expression actually mediate the neoplastic phenotype. Other key avenues of work need to be explored. For example, it will be important to try to identify germline mutations in HBV-infected patients that are passed on to their children, resulting in the development of HCC in childhood. Clinical materials will also be important for the validation of new markers with diagnostic or prognostic potential. In this context, there is an urgent need to establish simple and low-cost tests based upon molecular changes that are hallmarks of HCC development. Identification of patients with early HCC will also significantly increase survival through its impact upon treatment. The discovery and validation of HCC markers may permit accurate staging of lesions, determine the proximity of such lesions to malignancy, and determine whether lesions with a particular genetic profile are still capable of remodeling through appropriate therapeutic intervention. The efficient reintroduction of the relevant tumor suppressors, or the inhibition of oncogene expression by siRNA, provide just some of the additional opportunities that will ultimately be useful in patient treatment. Together, these approaches will go far in reducing the very high morbidity and mortality associated with HCC.