Abstract Since the landmark studies conducted by Huggins and Hodges in 1941, a failure to distinguish the etiological role of testosterone from its role in prostate cancer progression has led to the prevailing opinion that high levels of testosterone increase the risk of prostate cancer. To date, this claim remains unproven and epidemiological evidence is inconsistent. In this study, we present a novel dynamic model regarding the relationship between testosterone and prostate cancer to advance our understanding of the current findings and to guide future prostate cancer research. In our model, we theorize that it is the magnitude of the age-related declines in testosterone rather than the static level of testosterone measured at a single point that may trigger and promote the development of prostate cancer. After body testosterone falls below a threshold level, the prostate cells cannot maintain their normal functions and eventually some turn into tumor cells. Individual threshold of testosterone may be positively associated with a peak testosterone level of this individual during young adulthood. Our dynamic model can satisfactorily explain the observed age pattern of prostate cancer incidence, the seemingly conflict of findings from epidemiological studies on testosterone and risk of prostate cancer, racial disparities in prostate cancer incidence, many risk factors associated with prostate cancer, and the role of testosterone in prostate cancer growth. Our dynamic model may also have implications for testosterone replacement therapy. We conclude that this dynamic model provides a new perspective to enhance our understanding the role of testosterone in the onset and development of prostate cancer, to guide future studies, and to promote and improve testosterone replacement therapy in prevention of prostate cancer.