Small-animal imaging with positron emission tomography (PET) has become a valuable tool for evaluating preclinical models of breast cancer and other diseases. In this review, we examine a number of issues related to preclinical imaging studies with PET, using transgenic models of ductal carcinoma in situ and metastasis as specific examples. We discuss imaging components such as reconstruction, normalization, and extraction of quantitative parameters. We also analyze the effect of longitudinal correlations on cohort size and present some simple statistical techniques for determining cohort sizes that may be helpful in designing preclinical imaging studies. We describe studies that are greatly facilitated by access to non-invasive imaging data including a study involving multiple endpoints and another investigating metastasis. We conclude with a brief survey of emerging approaches in small-animal PET imaging.