7 December 1998 CapCure5\Sawyers.MS-- 1,100 words
Androgen-dependent cells not dead, just resting,
unless they have a HER-2/neu mutation
Lake Tahoe, NV--Cell biologists have a present for you: a better understanding of androgen independence. They're filling out the details of what happens when an androgen molecule docks onto its receptor, and transmits a signal down a pathway to the nucleus that orders the cell to divide.
They're filling out the details of why the cell divides even without an androgen molecule.
They now know, for example, that androgen-deprived cells don't always die--sometimes they're just resting.
And they're looking for drugs to repair these mistakes.
"The androgen receptor and downstream coactivator pathways are potential targets for second generation anti-androgen drugs," said Charles L. Sawyers, MD, at CapCURE'S 5th Annual Scientific Retreat here this September.
"When patients develop hormone refractory prostate cancer, the androgen receptor signaling pathway is revved up," said Dr. Sawyers. "So the question is, what did the cell do to sneak around the androgen receptor blockade?"
CapCURE's researchers are working on several pathways that bypass hormone control.
"In some cases," said Dr. Sawyers, associate professor of medicine at UCLA, "we found that the HER-2/neu oncogene gets upregulated."
"When we put HER-2/neu into LNCaP and LAPC-4 cells it made them androgen independent," said Dr. Sawyers. "It did so by turning on the androgen receptor."
"In my mind that was a real eye-opener," said Dr. Sawyers, "because why would you expect HER-2/neu to turn on the androgen receptor?"
HER-2/neu is not found in detectable levels in normal tissue, but found in some breast, kidney and bladder tumors. It is located at chromosome 17q21. In about 20% of breast cancers, its DNA is repeated, and the protein is overproduced, which is associated with a poor prognosis. In some prostate cancers, the gene is also overproduced, but the DNA is not repeated, and reports on its significance have been conflicting.
"HER-2/neu is a tyrosine kinase," said Dr. Sawyers. "Tyrosine kinases are a class of enzymes that are implicated in a range of cancers. Biochemically they phosphorylate certain proteins at the tyrosine. It's a major mechanism by which the proteins in these cascades are turned on and off. The phosphatases work in the other direction. They take the phosphate off tyrosine. So the kinase is the on switch, the phosphatase is the off switch."
Dr. Sawyers also clarified the role in prostate cancer of a phosphatase, a tumor suppressor gene called PTEN. PTEN was recently found at chromosome 10q23, a region which is deleted or unexpressed in 60% of advanced prostate cancers. PTEN inhibits cell growth, by blocking a pathway called HAT.
The evidence for PTEN's tumor suppression, said Dr. Sawyers, is that germ line mutations cause cancer, and knockout mice have cancer, including prostate cancer.
PTEN is "in my opinion the most commonly mutated gene in advanced prostate cancer," said Dr. Sawyers.
Dr. Sawyers described how 2 prostate cancer cell lines developed at UCLA didn't commit apoptosis after all.
"LAPC-4 and LAPC-9 are androgen dependent," said Dr. Sawyers. "After prolonged growth in castrate mice they can become androgen independent."
In SCID mice, LAPC-4 and LAPC-9 are a good model for metastases. "When implanted in the prostate, they circulate in the blood, and metastasize to the lung and bone marrow," said Dr. Sawyers. "When injected into the bone, they elicit osteoblastic reactions."
Androgen mediates a survival signal and a differentiation signal in prostate cells, Dr. Sawyers explained. "The secretory epithelial layer of the prostate is androgen-independent," he said. "The basal epithelial layer, upon exposure to androgen, differentiates to secretory epithelial cells."
If we inject 1 million LAPC-9 cells subcutaneously into an intact SCID mouse, it will develop a tumor with significant PSA levels in 3-4 weeks, Dr. Sawyers said. If we castrate the mouse, the PSA will plateau at 1/3 the level. After 1 year, a subset of cells will develop into an androgen-independent tumor.
In the traditional view, the androgen-dependent cells died, and the androgen-independent cells survived through selection. "To test that idea, we added androgen back to the animal after about 110 days," said Dr. Sawyers. "To our surprise, we saw an immediate and dramatic rise in the PSA as well as the tumor size."
So although the tumors were shrinking in response to castration, they weren't necessarily dying, he said. "They were just sitting there."
"If we inject 10,000 LAPC-9 cells sc into a female or castrate male, they never developed a tumor, and never developed androgen-independent cells after 110 days," said Dr. Sawyers. "So what happens to those 10,000 cells sitting there sc for 6 months? Do they die by apoptosis?"
"To our surprise," said Dr. Sawyers, "adding an androgen challenge to the animal at 6 months led to the immediate outgrowth of a tumor with a kinetics comparable with, if not faster than, what we would have gotten if we started the experiment by injecting the same number of tumor cells into an intact male mouse."
"When we talk about androgen independence," said Dr. Sawyers, "we should distinguish between the requirement of prostate cancer cells for androgen as a survival factor, versus a growth factor."
"At least on the secretory layer of the normal prostate gland, androgen is both a survival factor and a growth factor," said Dr. Sawyers, "because these cells die by apoptosis."
"However, in the early stage of prostate cancer which we call hormone dependence, in fact the tumor may have developed an independence of androgen from the survival requirement, yet remain dependent on androgen for growth," said Dr. Sawyers. In the traditional definition, hormone-dependent tumor requires androgen for both survival and growth.
Dr. Sawyers outlined several possible mechanisms for androgen independence.
(1) About 20-30% of androgen-resistant cancers have a mutation in the receptor. One mutation is the "promiscuous receptor." LNCAP's mutation in the hormone-binding domain allows it to respond to non-androgen steroids. But LAPC-4 has a normal androgen receptor, so there must be other mechanisms.
(2) The androgen receptor on the X chromosome could undergo genomic amplification, with excess receptor protein. But this occurs rarely.
(3) Alternative signaling pathways could bypass the androgen pathway, and activate the same downstream program.
The correct answer is (3). "The HER-2/neu receptor was consistently expressed at a higher level in androgen independent cell lines," said Dr. Sawyers. It operates synergistically with or independently of the androgen receptor. It "talks with" downstream molecules, on the ras, her, and other pathways.