SUMMARY Mouse models of cancer are invaluable for obtaining detailed knowledge about tumor development and for screening therapeutic and preventive approaches. Mesothelioma is an unusual cancer because the same carcinogen, asbestos, causes a similar disease in both humans and animals. Unlike most other cancers, murine mesothelioma can therefore be regarded as a disease homolog, rather than a model as such. However, because asbestos-induced cancer has low penetrance and a long lag time, most translational studies have utilized more efficient models such as tumor transplantation. In consequence, many promising results have not translated into positive findings in patients. Here, we describe the widely used murine mesothelioma models and critically discuss their relative advantages and disadvantages. We emphasize the use of the appropriate model for the specific research question and the need to use multiple models in order to obtain robust and translatable data.