Abstract The advent of syngeneic mouse tumor models provided the scientific foundation for cancer immunotherapies now in widespread use. However, in many respects, these models do not faithfully recapitulate the interactions between cancer cells and the immune systems of human patients who have solid tumors because they represent a very early stage in the immune response to the newly transplanted cancer cells compared with the relatively mature stage found in human patients at the time of treatment. The lack of translatability of syngeneic models is probably responsible for many failed clinical trials conducted at considerable expense, involving far too many patients with cancer who received no benefit. Better mouse models would substantially accelerate the pace of discovery of new immunotherapies. Until these models emerge, a better understanding of the differences between the existing syngeneic models and human cancers may provide a more efficient path for moving experimental drugs into clinical development. To accomplish this, we must consider mice transplanted with syngeneic tumor cells to be in vivo assays, potentially useful for understanding the mechanism of action of immunotherapies rather than disease models.