Issue of
June 28, 2000

When cells listen to the wrong hormones, prostate cancer turns deadly  

BY MITCH LESLIE

Prostate cancer escapes control and turns deadly when the malignant cells start listening to the wrong hormones, Stanford researchers have discovered. The team ­ led by professor of medicine David Feldman, MD ­ found that, because of a mutation, the stress hormones cortisone and cortisol can trigger the growth of later-stage prostate cancer cells. The results explain why prostate cancer eventually becomes impervious to a commonly used treatment.

Early in the course of the disease, the growth of prostate cancer cells is under the control of testosterone and related hormones ­ collectively known as androgens. These hormones deliver a strong message to a prostate cancer cell: Be fruitful and multiply. As they reproduce, the cells spill out a protein marker called prostate specific antigen, or PSA. High levels of PSA in the blood ­ detectable with a widely used test ­ are often the first sign of prostate cancer.

To check this rampant growth, doctors may resort to castrating the patient ­ either surgically or chemically ­ to slash blood androgen levels. Though it sounds extreme, the treatment usually works as long as the tumor remains "androgen-dependent," or requires androgens to grow, Feldman said. "In the absence of androgens, androgen-dependent cancers usually shrink, the patients do well, and PSA levels fall," he said.

But unless the patient dies of something else first, Feldman said, the cancer eventually metamorphoses into an "androgen-independent" state in which it can grow without stimulation from these hormones. No longer halted by castration or androgen deprivation, the cancer begins to proliferate, and currently there are no other treatments to stop it from spreading to the bones or other organs and eventually killing the patient. About 40,000 American men die every year from androgen-independent prostate cancer. Among cancers, only lung cancer kills more men.

What transforms the cancer from treatable to untreatable? To find out, the researchers began studying a lineage of metastatic prostate cancer cells isolated from a patient who later died from the disease. Initial experiments suggested there was a defect in the cells' responsiveness to androgens. So Feldman and his team sequenced the gene for the androgen receptor, a protein in the cell's nucleus that functions as an androgen sensor. Androgens latch onto the receptor molecule, triggering metabolic changes within the cell that stimulate cell division.

Writing in the June issue of Nature Medicine, Feldman and colleagues report that the androgen receptor gene in the metastatic cancer cells contained two mutations that transformed its activity. Androgens no longer bound tightly to the receptor. Instead, cortisol and cortisone, the body's "stress hormones," did. Though released in large amounts during stressful situations, cortisol and cortisone also help regulate metabolism during quieter times and are secreted continually by the adrenal glands atop the kidneys.

"These two mutations changed the androgen receptor structure so that now cortisol and cortisone could bind to the mutated receptor and act like 'pseudo-androgens,' " Feldman said. Presumably, the mutations contorted the receptor molecule into a new shape that closely matched the shapes of cortisone and cortisol. By recreating the mutant receptors in the laboratory, Feldman and his colleagues proved that the double mutation converted the androgen receptor into a receptor that responds inappropriately to cortisol, cortisone and other hormones.

To show that the mutated cancer cells responded to the cortisone and cortisol, Feldman's team grew the cells in a culture medium and exposed them to increasing doses of the hormones. Sure enough, the more cortisol and cortisone present, the faster the cells divided and the more PSA they released ­ a sign that cortisol and cortisone were activating the same metabolic pathways that androgens normally would trigger, Feldman noted.

For patients, this change in the shape of the androgen receptor would bring on disaster. Levels of cortisone and cortisol in the body are high enough to keep the cells permanently stimulated, Feldman noted. That means that the prostate cancer cells are deluged with signals to divide, which could account for their rapid multiplication in later-stage cancers. However, Feldman added, the team is not sure yet how common these mutations are among prostate cancer patients.

Some good news emerged from the study. Though it would be difficult to banish cortisol and cortisone from a patient's body without creating further problems, it may be possible to block the mutated androgen receptor and prevent these hormones from stimulating growth of cancer cells. Feldman and colleagues are already pondering ways to do this.

Feldman's Stanford co-authors include research associate Xiao-Yan Zhao, PhD; senior research scientist Peter Malloy, PhD; research associate Aruna Krishnan, PhD; postdoctoral fellow Srilatha Swami, PhD; and associate professor of urology Donna Peehl, PhD. Nora Navone, MD, PhD, of M.D. Anderson Cancer in Houston also collaborated with the Stanford team. The study was funded by grants from the National Institutes of Health, the American Institute for Cancer Research, and the U.S. Army Medical Research Acquisition Activity. SR