It's well known that male sex hormones promote the growth of prostate cancer. That's why doctors use hormone therapy -- chemical or physical castration -- to shut off these tumor-promoting androgens.
But Chang's team finds that in different types of prostate cancer cells, androgens actually inhibit prostate cancer. When these tumor cells don't get androgens, they become more aggressive and more invasive.
The lining of the prostate is made up of epithelial cells. The fibrous body of the prostate is made up of stromal cells. On their surfaces, both cell types have triggers -- androgen receptors -- that fire when they encounter sex hormones. Triggering androgen receptors has different effects in each cell type.
"The androgen receptor in the stromal cells always turns the cancer on," Messing tells WebMD. "The androgen receptor in the epithelial cells, at least in the animal models we studied, tends to inhibit cancer."
This, Messing says, helps explain why hormone therapy always works at first but then tends to lose its cancer-inhibiting effect over time.
Since the cancer-promoting effect of androgens is strongest in the earlier stages of cancer, hormone therapy does more good than harm. But as the cancer spreads to distant sites, Messing says, the cancer-inhibiting effect of androgens may become more important. At this point, hormone therapy may do more harm than good.
How can the same hormones have two opposite effects?
"Anyone who has been around teen boys and older men knows that androgen receptors in different parts of the body cause different effects," Messing says. "Androgen receptors on the scalp make older men lose their hair, while androgen receptors on the face make teenagers grow beards. So androgen receptors can do different things in different places."
Doctors have long known that hormone therapy has different effects at different times in different parts of the body, says Peter Nieh, MD, director of the Uro-Oncology Center at Emory University, Atlanta.
"We'd all like to find a silver bullet that attacks one thing but does not hurt anything else. The problem is there is always collateral damage," Nieh tells WebMD.
Chang's team demonstrated the opposite effects of androgen receptors in cell-culture studies and in studies of prostate-cancer-prone mice that lacked androgen receptors only in their prostate epithelial cells. These mice had much more aggressive cancer, apparently because they lost the ability to respond to the cancer-inhibiting effects of androgens.
The researchers also point to studies of prostate glands removed from men with prostate cancer. There were significantly fewer androgen receptors in metastatic prostate cancers than in early prostate cancers or in normal prostate cells.