(FORTUNE Magazine) – Whenever major news develops in humanity's war against cancer, you can expect a lot of confusion. With its ability to pervert the body's genius for growth and regeneration into a relentless horror, cancer has to be death's favorite disease. The treatments are scarcely better: Surgery and doses of radiation and toxic chemicals cause great misery and pain, often unrewarded by lasting gains. No wonder positive news filtering in from the war on cancer tends to incite outbursts of wild surmise and wishful thinking, not to mention suspicions and accusations of quackery and hype.

Dr. Judah Folkman knows all too much about such outbursts. He has been the moving force behind what is arguably the most important single insight about cancer of the past 50 years--that tumors feed their own runaway growth by inducing angiogenesis, the formation of new blood vessels. In fact, tumors can't grow larger than BBs without luring new capillaries to sprout from nearby vessels and pipe in oxygen and nutrients. This has a momentous implication: Drugs to inhibit angiogenesis may halt cancer's growth with few side effects--starving a tumor of blood would subject it to the equivalent of a crippling stroke.

Over the past two years countless media spotlights have been trained on Folkman's lab at Children's Hospital in Boston. But they've brought more heat than illumination. The commotion began on May 3, 1998, when the New York Times ran a riveting front-page story describing how endostatin, an angiogenesis inhibitor discovered in Folkman's lab, withered tumors in mice. As it happened, that was old news, widely reported months before. But the writer, Gina Kolata, had a sensational grabber: She quoted James Watson, the Nobel laureate co-discoverer of DNA's structure, as proclaiming, "Judah is going to cure cancer in two years."

A few days later the Times published a letter from Watson stating that he didn't recall uttering the astonishing quote. (The newspaper stood by the story.) But the article and breathless TV and radio follow-ups raised impossible hopes among cancer victims. Voice-mail boxes at Children's Hospital were flooded by thousands of desperate callers. EntreMed, a small Rockville, Md., biotech company developing endostatin, saw its share price rocket from $12 to $85 before falling back to about $33.

Deeply chagrined, Folkman turned down hundreds of interview requests. Controversy dogged him all the same. Six months after the Times story, the Wall Street Journal weighed in with a scathing front-page piece portraying him as a visionary known for promising "too much too soon." The story reported that several scientists had had trouble replicating the renowned endostatin results and recounted earlier failed attempts to turn findings from Folkman's lab into anticancer drugs.

More headlines followed. Scientists outside Folkman's lab soon verified its finding: Endostatin does shrink tumors in mice. But then Bristol-Myers Squibb announced it was dropping work on angiostatin, a similar agent discovered in Folkman's lab that also had been played up in the press. Last fall hopes rose as EntreMed advanced endostatin to human tests in record time for a novel compound; but this May news reports questioned whether endostatin was working in the early trials, and Entremed stock got hammered again.

Watching this saga unfold, you might think the promise of angiogenesis research hinges on a quest to turn one or two iffy drugs into winners. But much bigger news involving scores of companies and new medicines has been building behind the headlines. As angiogenesis research matures in the coming years, cancer therapy is likely to undergo profounder changes than any single drug could bring about. An entirely new way of treating cancer is rapidly moving from the lab to the clinic. In the near term, it promises to make existing chemotherapy much more effective; within five years it could bring on potent new anticancer drugs.

In fact, drug and biotech companies have been at work developing Folkman-esque drugs for more than ten years. Besides fighting cancer, these medicines promise to arrest abnormal angiogenesis underlying a remarkably diverse array of diseases--rheumatoid arthritis, psoriasis, and diabetic retinopathy, a major cause of blindness, to name a few. Drugs to stimulate angiogenesis are also in the works, promising to replumb ailing hearts, cure ulcers, and speed wound healing.

Some 200 commercial labs worldwide are racing to deliver this therapeutic tsunami, according to the Angiogenesis Foundation, a nonprofit group in Cambridge, Mass., that tracks and promotes research on the subject. Nearly 50 antiangiogenic drugs are already in clinical trials for cancer alone, including 12 in final Phase III tests.

If even a fraction of the new antitumor medicines reaches the market, the big C should become a lot less scary. Unlike today's scorched-earth chemotherapies, which kill many types of normal cells as fast as they do cancerous cells, angiogenesis inhibitors promise to suppress tumor growth with little toxicity. That's because new blood vessels are rarely formed in the body, so blocking their growth isn't likely to cause dire effects.

The new drugs also promise to help overcome the long-standing problem of tumor resistance--cancer's nightmarish tendency to recur in unstoppable form. The novel drugs won't necessarily eradicate tumors. But when used in anticancer cocktails, they may turn many cases of rapidly fatal cancer into "manageable" chronic illness.

Research in Folkman's lab has already led to a treatment for tumorlike blood-vessel growths called hemangiomas that has saved hundreds of lives. And in a landmark animal study reported last month, his team demonstrated that widely available chemotherapy drugs can act as potent angiogenesis inhibitors when given frequently at lower-than-usual doses. This five-year mouse study, which has received little media attention, may well turn out to be one of the biggest cancer breakthroughs in years.

Led by oncologist Timothy Browder, the study also showed that relatively low doses of a conventional cancer drug, combined with an experimental antiangiogenic medicine, could eradicate "drug-resistant" Lewis lung carcinoma in mice, an aggressive cancer immune to chemotherapy alone. "That's something," says Folkman. "Nobody's ever been able to cure drug-resistant Lewis lung carcinoma before with anything."

Whether similar drug cocktails can save people with drug-resistant tumors isn't clear, of course. We may soon find out--such cocktails can be formulated with antiangiogenic drugs that are already approved for other indications, says University of Toronto researcher Robert Kerbel, who corroborated Browder's results strikingly in a separate study. One such drug is thalidomide, the notorious sedative that caused thousands of birth defects 40 years ago. In 1992, thalidomide was found to inhibit angiogenesis by ophthalmologist Robert D'Amato, a Folkman collaborator. The drug was reintroduced in 1998 by Celgene, a Warren, N.J., biotech company, to treat leprosy. Doctors now widely prescribe it to treat a bone-marrow cancer called multiple myeloma.

Another is Celebrex, a popular new painkiller sold by Pharmacia, the $16-billion-a-year drugmaker in Peapack, N.J. Even aspirin, which, like Celebrex, inhibits a tumor-associated enzyme called COX-2, apparently produces some antiangiogenic effects. That may be why arthritis sufferers who regularly pop aspirin and similar over-the-counter painkillers have a nearly 50% lower risk of death from colon cancer than nonpoppers.

It bears repeating that miracle cancer cures are not in the offing. Several angiogenesis-based drugs have flopped in clinical trials. But such setbacks don't faze Folkman in the least. At 67, he still has the energy, enthusiasm, and curiosity of a 5-year-old--qualities that have inspired a steady stream of gifted young researchers to launch their careers in his lab and push the envelope on angiogenesis research. Says Folkman: "When an angiogenesis inhibitor fails, everybody says, 'See, the idea doesn't work.' But that's not right. The principle that tumors are dependent on angiogenesis is well established. Translating it into products isn't."

That may soon change, thanks to the spate of antiangiogenic drugs in late stages of development. In a case Folkman finds particularly gratifying, an aggressive golf-ball-sized thyroid tumor in an Ohio patient melted away after he was injected last year with an experimental antiangiogenic drug called CA4P (see box). "I saw that form of cancer when I was a resident," says Folkman, "and thought it was horrible. It grows around the trachea and slowly strangles patients."

What most excites cancer experts about angiogenesis research is that it's rapidly multiplying the angles of attack on tumors. The two dozen or so anticancer drugs developed over the past 50 years and now in widespread use are basically aimed at just two molecular targets: They block the copying of DNA molecules, which cells must do in order to multiply, or they mess up tubulin, Tinkertoy-like molecules that form assemblies needed for cell division.

This paucity of targets is a major reason the war on cancer hasn't gone very well. After shrinking when first treated, tumors often regrow within a few months or years, armed with deadly resistance to therapies that previously worked. Their infernal resilience is due to the fact that tumors' deranged genes mutate at a furious pace, enabling their cells to readily evolve metabolic tricks that defeat any single drug. Like crazed dictators with crack bioweapons labs, the renegade cells can turn all the fantastic capabilities of our genes against us. Thus researchers have long sought additional key molecules involved in tumor growth--hitting several such targets at once with drug cocktails might wipe out cancer cells before they can evolve resistance.

Angiogenesis is rich with targets. The process begins when growing tumors start gasping for oxygen. That triggers their secretion of molecules representing a kind of siren call to endothelial cells, which form blood vessels' walls. These alluring chemicals, called growth factors, turn nearby endothelial cells into tumor-controlled zombies--they begin multiplying and migrating toward the tumors in order to form new blood tributaries.

Meanwhile other molecules, called matrix metalloproteinases, or MMPs, chew through obstructing membranes so that the mesmerized endothelial cells can move toward tumors. Then Velcro-like molecules, called integrins, come into play to help the endothelial cells line up and form tubes. Several dozen molecules in this panoply have already been identified as drug targets, says Kerbel.

The take-home message: Even if no antiangiogenic drug becomes a stand-alone blockbuster, collectively the medicines promise anticancer cocktails of unprecedented effectiveness thanks to the multiplicity of targets they'll hit.

Like most big new ideas, Folkman's insight that tumors trigger angiogenesis met heavy resistance on its way to becoming obvious. His first attempt to publish an article supporting it was rejected by dozens of journals. For more than a decade after he finally got it into print in 1971, most cancer experts dismissed it as dubious conjecture. Says Kerbel: "One reaction was, 'Show us an angiogenesis stimulator secreted by tumors, and maybe we'll believe you.'"

The chemical "on switch" for angiogenesis proved very difficult to isolate. It turned out that there are multiple stimulators and inhibitors composing a complex check-and-balance system--tumors secrete various signaling molecules to tip the balance toward stimulation. Further complicating the picture, many of these molecules have nonangiogenic roles as well. And one of them, angiopoietin-2, is confusingly two-faced: It can either cause blood vessels to wilt or help bring forth new ones, depending on circumstances, according to studies at Regeneron, a biotech company in Tarrytown, N.Y.

In the mid-1980s, Folkman collaborators Yuen Shing and Michael Klagsbrun finally isolated a tumor-derived agent, called basic fibroblast growth factor, or bFGF, that clearly helps stimulate blood-vessel growth. But novel medicines remained a distant dream--it would be a decade before the drug industry fully embraced Folkman's idea that blocking such molecules with drugs might quell cancer.

"Pharmaceutical companies were used to looking for highly toxic drugs that kill cancer cells and cause tumor shrinkage within a few weeks," explains Donald Ingber, a Folkman collaborator at Children's Hospital. "They weren't interested in relatively nontoxic drugs that inhibit growth rather than shrinking tumors," as did the early antiangiogenic drugs developed in Folkman's lab.

But by the early 1990s a race to develop antiangiogenic drugs was quietly taking shape behind the industry's closed doors. One reason was a report in the New England Journal of Medicine that antiangiogenic therapy had saved a life. In 1988 a hemangioma was discovered growing in the lungs of a 12-year-old Denver boy named Tommy Briggs. The prognosis was abysmal--such lung growths had been fatal in all previously reported cases.

By a lucky twist of fate, the boy's doctor, Carl White of the National Jewish Medical & Research Center in Denver, attended a lecture on angiogenesis a few months after Briggs was diagnosed. White was galvanized by one of the findings that was trotted out. In 1980 a researcher in Folkman's lab, Bruce Zetter, had discovered that a drug called alpha interferon stopped blood vessels' endothelial cells from migrating, indicating it would hinder angiogenesis. After consulting with Folkman, White put Briggs on daily low-dose injections of the drug. "Within 24 hours of the first dose, he began coughing up blood," says White. "It scared the hell out of us."

But the episode soon passed, and within a few months the youngster's hemangioma was clearly withering--now 25, he's a sales associate at J.P. Morgan in Denver. After White's report, Folkman's team pioneered the use of the drug in newborns with life-threatening hemangiomas--although it can cause leg spasticity when used at an early age, it has saved some 300 lives worldwide.

Drugmakers became interested but still didn't pursue angiogenesis intensively--hemangiomas, although tumor-like, don't metastasize like cancer, and alpha interferon and other early inhibitors didn't show striking anticancer effects. What the companies really wanted was a key stimulator of angiogenesis in cancer tumors to use as a target. With such a molecule in hand, they could develop potent blockers of its activity, perhaps stopping cancer-related angiogenesis cold. In 1989 one such target was finally identified by Napoleon Ferrara, a young scientist from Sicily who had recently joined Genentech in South San Francisco. Dubbed vascular endothelial growth factor, or VEGF, the tumor-secreted substance triggers the release of other factors that orchestrate angiogenesis. Later it was shown to have another critical function: keeping fragile, newly formed endothelial cells perky. Harvard researcher Harold Dvorak also showed that VEGF makes blood vessels leaky--that's why cancer's first sign is often blood in the urine, stool, or sputum.

Genentech soon developed a monoclonal-antibody drug whose molecules selectively stick to VEGF, blocking its signal. The drug, combined with conventional chemotherapy, is now in Phase III clinical trials for colon and lung cancer. Other VEGF blockers are being clinically tested by Sugen, a Pharmacia unit; ImClone Systems in New York City; and Ribozyme Pharmaceuticals in Boulder, Colo.

During the past decade commercial interest in angiogenesis has exploded as its full import has sunk in. Perhaps the most significant possibility--and the reason the mouse study with endostatin originally caused a stir--is that antiangiogenic therapies may be able to beat the lethal problem of tumor resistance.

Toronto's Dr. Kerbel was the first to spell out why that's a reasonable hope. In a 1991 paper he argued that since endothelial cells in tumors' blood vessels are normal, they should be genetically stable and hence have little chance of mutating into drug-resistant monsters. So drugs that target such cells shouldn't lose their potency over time, as medicines aimed at quick-changing tumor cells often do.

Kerbel's idea would be useless unless drugs that target tumors' blood vessels leave other blood vessels unharmed. As it happened, Folkman had explored that issue years before. In the early 1980s he'd shown that a combination of cortisone and a blood thinner called heparin was toxic to newly formed endothelial cells, but not to the mature ones in normal blood vessels. That indicated such newborn vessels are weaklings that might be wiped out without destroying the body's well-established arteries and veins--a possibility later confirmed in other studies.

Drawing on his knack for folksy analogies, Folkman compares treating cancer to maintaining a tennis court. Laid out flat, the endothelial cells in an adult's blood vessels would cover an entire court surface, he says. Like clay, "normal endothelial cells almost never grow," hence are relatively invulnerable to antiangiogenic drugs. "Tumor angiogenesis is like a single weed growing on the court."

Proof of Kerbel's theory emerged only after another six years of painstaking research in his and other labs. For starters, the researchers needed an especially potent angiogenesis inhibitor, one that would actually shrink tumors rather than merely curb their growth. That would enable them to conduct a telling study: First, implant tumors in mice and suppress the cancer with the drug. Next, halt the treatment, allowing the tumors to roar back. Then inject the drug again. If it shrank the resurging tumors--and continued to do so without losing potency in successive cycles of treatment and regrowth--Kerbel's idea would be golden.

Once again, Folkman's lab led the way. In 1996, Michael O'Reilly, a postdoctoral fellow in the lab, isolated endostatin, one of the body's potent angiogenesis inhibitors. That paved the way for the telling study on tumor resistance, spotlighted two years later in the overwrought New York Times piece. It also led to a troubling mystery.

Within a few months of O'Reilly's finding, he and colleagues had implanted the gene for endostatin in bacteria, causing the microbes to produce enough of the substance to try in mouse studies. In November 1997 the team published its electrifying results in the journal Nature: Endostatin shrank virulent Lewis lung tumors in mice--and continued to do so through six cycles of treatment and regrowth. After the sixth treatment the mice appeared to be permanently cured.

At first the results seemed too good to be true, so O'Reilly repeated the study, then tried endostatin against two other kinds of cancer in mice. Each time it worked. But when researchers at the National Cancer Institute produced their own endostatin and injected it in mice, things took a perplexing turn. The endostatin in the NCI experiment didn't work. After the NCI team had struggled for months to replicate his results, O'Reilly stirred up a batch of endostatin, tested it in mice to confirm its potency, and shipped it overnight to the NCI packed in dry ice. Somehow it lost its antitumor effects on the way. The same thing happened when he shipped endostatin to a San Francisco colleague.

Frustrated and baffled, he packed some of his endostatin in dry ice, drove around with it in his car for three days, then injected it in mice. It didn't work. Then he tried FedExing it to himself. The endostatin came back inactive. "We didn't have a clue," says Folkman. "But then last August we saw a stunning paper in Nature Medicine. It's now enshrined in the lab."

The paper showed that biomolecules shipped in plastic tubes in dry ice, as O'Reilly's had been, can be rendered inactive as carbon dioxide evaporates from the ice and leaches through the plastic, making the solutions containing the substances highly acidic.

The vexing mystery seemed solved. But by then it was moot--other researchers had confirmed endostatin's antitumor effects in mice by switching to versions made in yeast and human cells. Such versions are water-soluble, making them easier to formulate as drugs than the insoluble, bacteria-made endostatin O'Reilly had used.

Toronto's Kerbel was delighted about the endostatin studies, for they proved that, just as he had predicted, antiangiogenic drugs can defeat the tumor-resistance problem--at least in mice. But one implication of his theory remained strangely at variance with what was known about resistance.

Conventional chemotherapy drugs kill not just tumor cells but all rapidly multiplying cells in the body, including blood-cell precursors in the bone marrow and hair-follicle cells--that's why the drugs are so toxic and cause hair loss. Endothelial cells rapidly multiply when forming new blood vessels to feed tumors. Thus such cells should also be highly vulnerable to chemotherapy. So why doesn't chemo continue to exert strong anticancer effects, even after mercurial tumor cells become resistant to it, by stopping the nonmercurial endothelial cells in their tracks?

This question puzzled many cancer experts until Dr. Browder, the oncologist in Folkman's lab, carried out his laborious, five-year study. Browder was haunted by the puzzle after treating a 14-year-old boy with rapidly spreading tumors. In the typical pattern, the boy's cancer had initially responded to chemotherapy, then relapsed and spread. In desperation Browder prescribed chemotherapy that recent studies had suggested was antiangiogenic. It failed--the boy died three weeks later. "I felt very bad. I couldn't understand why it didn't work," says Browder.

Then it clicked: Chemotherapy is given in brief bursts at high doses, typically followed by multiweek drug-free periods that are needed to let patients' bone-marrow cells recuperate from the toxic effects. During the drug-free weeks, Browder realized, proliferating endothelial cells hurt by chemotherapy also recuperate. So he devised a new way to use chemo: Give it frequently--which necessitates lower, less-toxic doses--to prevent the cells from recovering enough to form new vessels.

It sounds simple, but making it work was onerous. Browder spent months trying different dose levels and injection schedules in tumor-ridden mice before finding one that suppressed angiogenesis without much toxicity. When he finally got it right, the results were stunning. Continually injecting a conventional anticancer drug called cyclophosphamide every six days knocked out Lewis lung tumors. In contrast, mice with similar tumors that got the drug on the customary regimen--high doses followed by three weeks off--died within two months.

In other experiments the team pitted the new continual-dosing method against drug-resistant Lewis lung tumors in mice. According to past studies, animals with such tumors were doomed. But when injected with cyclophosphamide on the new regimen, plus an experimental antiangiogenic medicine called TNP470, an astonishing 84% of mice were cured. After waiting almost two years to see if their tumors would regrow, "we had a geriatric ward of mice," says Folkman. The study was reported in the April 1, 2000, issue of Cancer Research.

Meanwhile, Toronto's Kerbel, who had heard of the new regimen from Folkman, conducted a similar mouse study. By continually injecting very low doses of an anticancer drug called vinblastine, plus an experimental antiangiogenic drug developed by ImClone Systems, his team was able to wither aggressive tumors called neuroblastomas with virtually no sign of drug toxicity. Kerbel's study, co-authored by researcher Giannoula Klement, was reported in the April 2000 issue of the Journal of Clinical Investigation.

The two studies may well lead to breakthroughs in treating human cancers. But conventional chemotherapy regimens shouldn't be abandoned in favor of the new one, Browder cautions. One reason is that the artificially implanted tumors the scientists eradicated in mice are easier to treat than human cancers--the dramatic mouse results may not be borne out in people. Further, a number of cancers respond well to conventional therapy.

Still, several oncologists have already tried the new antiangiogenic regimen against refractory cancers. At least one patient, a woman with tumors that had spread to her liver, has had encouraging preliminary results. Her doctor, Donald Sutherland at Sunnybrook and Women's College Health Sciences Center in Toronto, heard about Kerbel's mouse study earlier this year. When conventional therapy failed to stop the woman's cancer, he tried combining a conventional anticancer drug with Celebrex, the antiangiogenic painkiller. Since then she's had "some stabilization" of her cancer, he says, "but it will take months to see if the change is lasting." Recently the patient was able to resume jogging--a small victory, maybe temporary, or maybe a sign of a turning point in a very large war.

With so many branches of angiogenesis research beginning to bear fruit, Folkman looks increasingly likely to add a Nobel Prize to the many awards he's already racked up. But don't expect a sudden revolution in cancer treatment, he cautions. "To those who ask me, 'When is cancer going to be cured?' my answer is, 'When was infection cured?'" he says. "The history of treating infection is one of steady progress as we've lowered the toxicity and improved the efficacy of antibiotics and vaccines. That's how it's likely to go with cancer."

Not the stuff of banner headlines, perhaps. But considering that incremental progress against infectious disease has saved tens of millions, if not hundreds of millions, of lives over the past 50 years, perhaps it should be.

FEEDBACK: dstipp@fortunemail.com