Chimeric humanized liver rodent models are being used by those at the forefront of biomedical and drug discovery research. These humanized models allow for critical in-vivo research to evaluate everything from human-specific drug toxicity to gene therapies. One area where humanized liver rodent models could offer tremendous potential is in the study of diabetes-related treatments. A recent article published in Biomedical and Biophysical Research Communications not only discusses AAV-based gene therapy for the treatment of diabetes in humans, but it also describes the benefits of humanized liver models in pre-clinical research.

In 2000, the otherwise promising drug, troglitazone, was discontinued as an unexpected side effect of this treatment included fatal hepatic failure. Unfortunately, this secondary effect in humans was not observed in traditional animal experiments due to the significant differences between the way human livers and rodent livers metabolize troglitazone. Chimeric humanized liver models have been shown to be an effective tool to allow biomedical researchers to better detect toxicity issues, such as this drug induced liver injury (DILI), early on in the pre-clinical trials to help avoid future accidental deaths.

Beyond its ability to better predict hepatic reactions in humans, humanized liver mice were also determined to provide a good predictive model for human response to AAV-based gene therapies. Using adeno-associated virus (AAV) carrying the PDX1 gene for delivery of liver-specific transgenes that ultimately convert liver cells into insulin-producing cells, has the potential to be a game changer for people suffering from diabetes. Though there is still a long way to go before these gene therapies are ready for patient use, based on this preliminary study, the researchers hypothesize that this method could treat human diabetes without a secondary effect.

Therefore, it is reasonable to believe that humanized liver rodents, could be used to examine AAV transduction of human hepatocytes in vivo and more accurately predict clinical transduction efficiency as compared to non-humanized rodent models.