We thank Drs. Hejazi and Rastmanesh for their interest in our article. The purpose of our study was to determine whether there is an association between tamoxifen therapy and diabetes, and our findings will need to be confirmed in other populations. Although our population-based databases have the advantage of providing a large sample size with which to address that question, they lack detailed clinical information such as that regarding weight gain and other risk factors.
We agree that the mechanisms postulated for this association are uncertain. We acknowledge that tamoxifen is a selective estrogen receptor modulator, and that its effects on estrogen vary by tissue site. As we discuss in our article,1 there is evidence that estrogen protects against beta-cell failure and diabetes,2-4 leading to speculation that the observed increase in diabetes may be related to its estrogen inhibitory effects at the beta-cell level. This hypothesis is supported by the finding that tamoxifen induces beta-cell apoptosis and insulin deficiency in mice through direct estrogen antagonism.3 However, as the authors point out, alternative hypotheses such as the potential inflammatory effects of tamoxifen also need to be considered. Further studies that include more comprehensive clinical and metabolic data are needed to explore this intriguing finding.