When Cancer Escapes the Prostate: New Strategies for Prediction, Treatment

Some prostate cancer cells are practically homebodies; their growth is creeping, their advance local. But other cells can't wait to leave the nest -- to hitch a ride on the bloodstream headed for points north. This restlessness has a name -- micrometastasis -- and it can be lethal.

Micrometastatic cells specialize in the quiet exit; these tiny flecks of cancer slip out of the prostate in such small numbers that they're invisible -- and impossible for doctors to detect, even during surgery for what appears to be localized disease.

These cells are the scourge of prostate cancer treatment: Once they've managed this breakout, escaping via the bloodstream to the lymph nodes or spine, prostate cancer can no longer be cured.

But scientist John Isaacs, Ph.D., professor of urology and oncology -- every bit as persistent and determined as the enemy, metastatic cancer -- is undaunted. He's spent his career stalking these cells, working to master their habits and properties, and developing an impressive, multi-pronged plan of attack.

Which Cancers are Likely to Roam?

Pity the meteorologists trying to forecast the weather during tornado season in Kansas: Despite all the technological advances that allow them to track storm patterns, in the end, all they have is probability -- the odds that a tornado will develop, and an informed guess about when and where it might hit.

Scientists studying prostate cancer face a similar predicament. Although landmark tables developed at Hopkins using PSA level, Gleason grade and clinical stage (see story on page 14) have given men and their doctors an unprecedented ability to predict the extent of cancer, these markers are most helpful "when the cancer looks either terribly aggressive or extremely aggressive," says Isaacs. "The difficulty is, many men fall between those two extremes." Unfortunately, some of these men -- and there's no way to know for certain which ones -- diagnosed with curable disease have cancer that has already left the prostate. "There aren't any techniques now to detect that level of cancer, because it's just so small," says Isaacs. "But it means that local surgery alone isn't going to be curative. What we're trying to do is come up with a molecular mechanism for predicting cancer's aggressiveness."

Over the last decade, Isaacs has cultivated what probably is the world's richest nursery of prostate cancer cell lines -- nearly 30 distinct varieties -- which he uses in layers of experiments ranging from tissue cultures to a spectrum of animal models (including mice with no immune systems, in which human tumors can grow). "We've got the full range," he says, human and animal tumors that are "highly metastatic, not metastatic, hormone-independent, hormone-dependent, well-differentiated and poorly differentiated," and everything in between. (In fact, one rat model of prostate cancer, which Isaacs developed when he was a postdoctoral fellow in the lab of Donald S. Coffey, Ph.D., has become the most widely used system for prostate cancer in the world; Hopkins has supplied it to more than 200 research laboratories worldwide.)

John Isaacs, Ph.D.

Isaacs is using the most aggressive cells in his encyclopedia collection for a sophisticated series of experiments in tissue culture and animals, designed to find genetic restraints for cancer -- molecular fences to keep these cells from roaming. Through painstaking lab work, he and colleagues are adding human chromosomes from normal cells to these nasty, metastatic cancer cells. It's a form of "roll call," in which they're testing, one by one, all 23 pairs of chromosomes in the human body. They're looking for signs of inhibitory effect -- any clue that something (a gene or genes) on one of these chromosomes can either suppress the cancer's growth or its ability to metastasize. So far, they've earmarked for further study a handful that look promising, including chromosomes 5, 8, 10, 11, and 17. "We've been able to map areas of chromosomes with genes that actually suppress metastatic ability," Isaacs says. Among the most exciting is on chromosome 11, where "we've not only identified the region" (the cancer-suppressing "neighborhood" is on the chromosome's short, or petite arm, called the "P" arm) "but we've cloned and tested the genes. The first gene we identified is called KAI-1."

Like a sandbag that helps keep a hot air balloon on the ground, KAI (pronounced like the Greek "chi")-1 suppresses metastasis. It's product is found in normal cells. But at some point, on a cell's journey from normal to metastatic, it disappears; the cell stops making it. Isaacs and colleagues have developed special antibody stains that recognize KAI-1's distinctive handiwork (telltale proteins that it makes), to search for the gene in prostate cancer biopsy specimens. The idea is that if a cancer is not making enough of it, it may be well on its way to metastasis -- and this stain could help predict aggressive cancer.

Chromosome 11 has proved a fertile field; it's yielded another promising gene, called CD44. Here, too, is a gene, found in the prostate epithelial cells, that makes a metastatic-blocking substance. "When these cells become cancerous, and when they become highly metastatic, they turn off the production of this protein," says Isaacs. "The gene is still there. With both KAI-1 and CD44, the genes are physically still present -- they're just not expressed." Isaacs hopes to find a few more of these genetic markers. Then perhaps one day, if tests (such as the reagent stains his lab has developed) show that a cancer has systematically inactivated several of the genes that could keep it in check, this may be a strong indication for aggressive treatment (However, before this becomes a widely used form of testing, much more study is needed, Isaacs cautions.)

One Way to Stop Cancer: Cut Off the Supply Line

So: After a radical prostatectomy, a man turns out to have micrometastases, invisible offshoots of tumor taking root at satellite locations in the body. His cancer lives; chances are, it will continue to grow. "What are you going to give him? Hormone therapy is very helpful, and in fact, it gives great palliation," says Isaacs.

"But it doesn't cure. It helps people, but it doesn't cure them." This brings us to phase two of Isaacs' cancer-fighting strategy: Putting prostate cancer cells on a leash, with highly-promised drugs called angiogenesis inhibitors.

Like Roman soldiers, advancing cancers pave the way before them, laying down a track of new blood vessels. This guarantees a ready-made supply of nutrients -- nourishing meals for the road -- which, it seems, the cancers absolutely cannot do without. Destroy this infrastructure, cut off the supply line, block these new blood vessels -- and the cancer cells starve.

Cancer cells make new blood vessels grow by subverting a normal process involved in wound healing. "Usually, once you become an adult, your blood supply is pretty stable, and -- except when your body's trying to repair an injury -- you don't really need new blood vessels," says Isaacs. "But in order for a cancer to grow, it has to stimulate its host to do a lot of things for it. A cancer isn't an autonomous machine that can grow anywhere; it's not like an air fern that just needs sunlight and water. It's very dependent on its host, and one of the major reasons why is because it needs vigorous growth of new blood vessels."

This process is called angiogenesis, and drugs to block it, called angiogenesis inhibitors, already exist. The good thing about these drugs, says Isaacs, is "that your other blood vessels -- supplying your heart, lungs, brain and normal tissue -- are already fully developed. Inhibitors of angiogenesis don't really produce any damage to them. They would target the blood vessels only in cancerous areas."

Isaacs and colleagues have been working with an angiogenesis inhibitor called Linomide, which has many qualities of a "dream" drug: It's inexpensive and already available, it can be given in pill form, it has low toxicity and hardly any side effects -- and it does a beautiful job of stalling tumor growth. Best of all, "there's really no way the cancer cell can become resistant to its requirement for blood vessels." That would be like a lung cells becoming resistant to oxygen.

"The disadvantage is that it's not something you could take only once and then never take again," says Isaacs. "The blood vessels are constantly being stimulated to grow by the tumor, so you'd have to take this chronically -- like someone with high blood pressure who takes medication every day." But many men might find this a tiny price to pay for the potential benefits -- putting a cancer's growth in slow-motion for years, perhaps even decades. "Say a man has very limited, micrometastatic disease," says Isaacs, "we know that, untreated, it might take five or six years for this cancer to produce symptoms. But an anti-angiogenic medication might be able to prevent this happening in 20 years. If the man is 60 years old, that may allow him to not die from prostate cancer. He may still have prostate cancer cells in his body -- this doesn't eliminate all of them -- but it will allow him to survive his cancer."

For Men with Extensive Disease, A Molecular Bomb

But an angiogenesis inhibitor won't do enough to combat more advanced disease. Starting the drug once cancer has become entrenched -- when it starts producing such symptoms as bone pain -- would be like closing the proverbial barn door after the horse has already galloped away: Too little, too late. "What these anti-angiogenic agents do is inhibit the growth of a tumor," Isaacs explains. "If a man has very extensive disease, they won't cause the tumor to regress and melt away."

So how to help these men, who need an effective long-term treatment most of all? This is phase three of Isaacs' research program: A molecular grenade that only detonates in cells that make PSA.

"We're taking advantage of two attributes of prostate cancer here," Isaacs says. "One is that it makes PSA, and the other is that PSA is an enzyme that can -- like a pair of molecular scissors -- clip protein." PSA recognizes certain strings of amino acids, the building blocks of protein, and cuts them up. (The specific proteins are involved in making a sperm-trapping gel, which is part of the semen; the prostate's main job is to contribute part of the fluid for semen.) Isaacs and colleagues are designing a drug by genetically doctoring a potent toxic molecule, hooking it chemically with this protein carrier -- so that's it's activated when PSA goes into its protein-clipping mode. Then the PSA, recognizing this sequence of proteins that it's supposed to cut will, in effect, pull the pin on its own grenade: One clip and boom! Out comes the toxic molecule.

The secret is an unlikely terminator, derived from an innocuous-looking member of the parsley family. "It's a compound called thapsigargin, isolated from the thapsia gargancia plant, found in the Mediterranean" says Isaacs. (He is working in collaboration with Soren B. Christensen. the medical chemist from the Royal Danish School of Pharmacy who first isolated, characterized and named thapsigargin.)

For nearly 2,000 years, resin from this plant has been a staple of Arabian medicine; it's a natural irritant, easily absorbed through the skin, which can ease the pain of rheumatism. Thapsigargin works by burrowing its way into a cell and targeting a protein that acts as a calcium pump: Like someone bailing water out of a leaky rowboat, this pump keeps calcium from rising above a certain level inside a cell.

The most interesting thing here is the calcium, which also happens to be a key that turns the engine of a genetic process called programmed cell death; the Greek name for this is apoptosis, which refers to leaves dropping off a tree. "This gives cells very specific signals to activate a process of suicide," says Isaacs. "Normally, calcium is almost 10,000-fold higher outside a cell than within it. If too much of the calcium gets inside it, it causes the cell to reprogram itself and activate this suicide pathway." The effect is like cranking up the gauge on a pressure cooker.

Programmed cell death is certainly not a new concept. It's fundamental to how babies develop -- the way certain cells in limb buds die, for instance, so that fingers and toes can form. It's the reason a tadpole looses its tail and becomes a frog. "If you look at these developing limb buds (in an embryo), the cells that are going to live are right next to the cells that are going to die. What could control such a tightly orchestrated pattern? For a long time, it was assumed that the microenvironment around the cell basically murdered it -- in other words, that a bad environment killed the cells. But it's now clear that the cells that are dying are in a very happy environment: They've got plenty of nutrients, plenty of oxygen -- they've got everything that they need to live. But they've been given a signal, and that signal says: Don't live. Die." And that pathway to death is apoptosis.

Now imagine a medieval fortress under siege. The enemy is outside; but one soldier scales the walls and opens the mighty gates, and this is all it takes to change the course of battle. By interfering with the crucial pump, thapsigargin allows that calcium outside the cell to sneak inside; it reaches too high a level, disrupts the cell, and activates this pathway of death. "The DNA inside the cell's nucleus gets all chewed up, becomes degraded to the point of not being useful for any information. The nucleus itself becomes fragmented, then the cell becomes fragmented," Says Isaacs. The grand finale is an act of cannibalism: These little fragments, called apoptotic bodies, are then consumed by neighboring cells.

"The great thing about this," says Isaacs, "is that the cell has no way of preventing its own activation of this pathway." Another bonus is that this death pathway -- unlike many chemotherapeutic drugs -- doesn't require rapidy dividing cells. It can kill any cell, within 24 to 72 hours.

Further Reading

"KAI1, a Metastatic Suppressor Gene for Prostate Cancer on Human Chromosome 11.p11.2," Science, Vol. 268, pp. 884-886, May 12, 1995. Jin-Tang Dong, Pattie W. Lamb, Carrie W. Rinker-Schaeffer, Jaminka Vukanovic, Tomohiko Ichikawa, John T. Isaacs, and J. Carl Barret.

"Down-Regulation of the KAI1 Metastasis Suppressor Gene during the Progression of Human Prostatic Cancer Infrequently Involves Gene Mutation or Allelic Loss," Cancer Research, Vol. 56, pp. 4387-4390, Oct. 1, 1996. Jin-Tang Dong, Hiroyshi Suzuki, Sokhom S. Pin, G. Steven Bova, Jack A. Schalken, William B. Isaacs, J. Carl Barret, and John T. Isaacs.

"Implication of Cell Kinetic Changes during the Progression of Human Prostate Cancer," Clinical Cancer Research, Vol. 1, pp. 473-480, May 1995. Richard B. Berges, Jasminka Vukanovic, Jonathan I. Epstein, Marne CarMichael, Lars Cisek, Douglas E. Johnson, Robert W. Veltri, Patrick C. Walsh, and John T. Isaacs.

"Antiangiogenic Effects of Quinoline-3-Carboxamide Linomide," Cancer Research, Vol. 53, pp. 1833-1837, April 15, 1993. Jasminka Vukanovic, Antonino Passaniti, Takahiko Hirata, Richard J. Traystman, Beryl Hartley-Asp, and John T. Isaacs.

"Antiangiogenic Treatment with Linomide as Chemoprevention for Prostate, Seminal Vesicle, and Breast Carcinogenesis in Rodents," Cancer Research, Vol 56, pp. 3404-3408, Aug. 1, 1996. Ingrid B.J.K. Joseph, Jasminka Vukanovic, and John T. Isaacs.

"Linomide Inhibits Angiogenesis, Growth, Metastasis, and Macrophage Infiltration within Rat Prostatic Cancers," Cancer Research, Vol. 55, pp. 1499-1504, April 1, 1995. Jasminka Vukanovic and John T. Isaacs.

"Human Prostatic Cancer Cells are Sensitive to Programmed (Apoptotic) Cell Death Induced by the Antiangiogenic Agent Linomide," Cancer Research, Vol 55, pp. 3517-3520, Aug 15, 1995. Jasminka Vukanovic and John T. Isaacs.

"Role of Programmed (Apoptotic) Cell Death During the Progression and Therapy for Prostate Cancer," The Prostate, Vol 28., pp. 251-265, 1996. Samuel R. Denmeade, Xiaohui S. Lin, and John T. Isaacs.

"The Role of Calcium, pH, and Cell Proliferation in the Programmed (Apoptotic) Death of Androgen-independent Prostatic Cancer Cells Induced by Thapsigargin," Cancer Research, Vol. 54., pp. 6167-6175, Dec 1, 1994. Yuzo Furuya, Per Lundumo, Alison D. Short, David L. Gill, and John T. Isaacs.