Will cancer ever be cured?

Not long ago, the defeat of cancer seemed inevitable. Decades of research would soon pay off with a completely fresh approach, an arsenal of clever new drugs to attack the very forces that make tumors grow and spread and kill.

No more chemotherapy, the thinking went. No more horrid side effects. Just brilliantly designed drugs that stop cancer while leaving everything else untouched.

Those elegant drugs are now here. But so is cancer.

The approach, which appeared so straightforward, has proven disappointingly difficult to turn into broadly useful treatments. Some now wonder if malignancy will ever be reliably and predictably cured.

The dearth of substantial impact so far suggests the fight against cancer will continue to be a tedious slog, and victories will be scored in weeks or months of extra life, not years. The full potential of the new approach may take decades to be realized.

The drugs, called targeted therapies, are intended to arrest cancer by disrupting the internal signals that fuel its unruly growth. Unlike chemo, which attacks all dividing cells, these medicines are crafted with pinpoint accuracy to go after the genetically controlled irregularities that make cancer unique.

Several have made it through testing, but despite their apparent bull's-eye hits, lasting results are rare. Instead, these new drugs turn out to be about as effective -- or as powerless -- as old-line chemotherapy. Aimed at the major forms of cancer, they work spectacularly for a lucky few and modestly for some.

But for most? Not at all.

Doctors have many theories about what's gone wrong. But it is clear that cancer is a surprisingly robust foe, packed with convoluted backup systems that kick in when threatened by the new drugs.

At best, experts now expect knocking down cancer will require an elaborate mixture of targeted drugs, assembled to match the distinct biology of each person's cancer.

The job is so daunting, especially for advanced cancers propelled by potentially dozens of nefarious genetic mutations, that scientists are even rethinking the goal of cancer research.

"Society as a whole, and most of the medical profession, have it wrong understanding we'll wake up one morning and find out cancer is cured. It won't happen. The public should give it up," says Dr. Craig Henderson, a breast cancer specialist at the University of California, San Francisco, and head of Access Oncology, a drug developer.

Henderson and many others have shifted their sights to something less -- converting cancer into a chronic disease, like diabetes or AIDS. Treatments might slow or even stop its worst effects so people survive for years reasonably free of symptoms.

Still, experts concede there is no firm evidence that targeted treatments will tame cancer to a chronic condition, either. Certainly, the ones tested so far do not often come close to this for the common varieties, such as lung, breast, colon and prostate cancer.

Although targeted therapies have their origins in basic cancer discoveries of the 1980s, the story for many began at a meeting of the American Society of Clinical Oncology in 1998. Researchers were thrilled to hear of the first convincing demonstration that a targeted drug could slow the course of cancer even a little. It was proof that the principle is sound.

Usually wary oncologists rhapsodized about a new era of treatment. "A tidal wave," one of them called it. Even then, no one predicted quick cures. But they clearly felt they at least had the key to getting inside cancer and fixing it.

The drug that caused the euphoria, Herceptin, became a standard treatment for spreading breast cancer, typically delaying progression by a few months in the quarter of victims with a particular genetic profile.

Since Herceptin, targeted drugs have become the prevailing approach in cancer research. Whenever any of these make slight progress, the news is widely and sometimes breathlessly reported. An estimated two-thirds of the nearly 400 cancer medicines in human study take this tack. Yet researchers do not envision successes any more spectacular from this pipeline than the modest effects of the handful already on the market.

"Right now, in the short run, we can bring an occasional miracle and have an overall small benefit," says Dr. John Glaspy, medical director of UCLA's surgical oncology center. "But there has not been a major improvement on what happens to them ultimately."

Furthermore, the dream of abandoning chemotherapy has largely evaporated. Even the targeted drugs' small benefits are typically seen only when combined with standard chemo.

Cancer doctors facing waiting rooms full of dying cancer patients, with little to offer but easing misery and perhaps a few extra months of survival, clearly had wished for more.

"The hope was that these targeted therapies would be the new magic bullet and would cure cancer," says Dr. David Decker, an oncologist at William Beaumont Hospital outside Detroit. "It's fair to say they haven't panned out the way we thought they would."

The targeted drugs have been most impressive against cancers of the blood and immune system, which are easier to control than the more common organ tumors. For instance, about half of patients getting Rituxan for non-Hodgkin's lymphoma have at least a 50 percent reduction in their cancer, and the improvement lasts an average of a year before the disease progresses again.

The one striking success, Gleevec, unfortunately works only against two rare blood and digestive cancers that involve unusually simple signaling pathways, offering ideal targets. Even so, Gleevec's stunning effects -- close to 90 percent initially get better -- often wear off in time.

In 2001, Dean Gordanier, a Boston tax attorney with a cancer-swollen belly, was "pretty much dying" when he started on Gleevec. For 18 months, it was a miracle. He gained weight and went back to work. Then, the day before Christmas, he learned the tumor was growing again.

For most who win a cancer reprieve, that would be the end. But as it turned out, another experimental targeted drug was available at the Dana Farber Cancer Institute, and Gordanier's cancer is in retreat again, although he expects this medicine, too, will eventually fail.

"I feel like I'm riding the crest of a wave," he says. "It could dump me at any time, but right now I'm cruising."

Dr. Brian Druker of Oregon Health and Sciences University, one of Gleevec's developers, says, "What Gleevec tells us is if we have the right target and the right drug, we will have spectacular results. Until then, we will be mired in incremental gains."

Those gains seem especially incremental against the far more complicated common cancers. For instance, the drug Iressa was approved in May on evidence that it temporarily shrinks advanced lung cancer in just 10 percent of patients. Doctors often assume drugs will work better if given earlier in the disease. But when Iressa was combined with chemotherapy in newly diagnosed lung cancer, the patients did not respond to it at all.

Others in the pipeline seem hardly more potent.

At this June's clinical oncology meeting, doctors reported the results with two targeted drugs for advanced colon cancer: A growth signal blocker called Erbitux (the same medicine that ensnarled Martha Stewart in a Wall Street scandal) temporarily shrank tumors in a quarter of patients. And Avastin, intended to stop cancers from building blood vessels, improved average survival by four months.

In theory, all of these drugs should work better, because they block the communication pathways hijacked by the genetic mutations that become cancer.

Cancer occurs when five to 10 ordinary genes develop mutations in a single cell over a person's lifetime, the consequence of biological insults like smoking or just bad luck. As a result, the genes may get stuck in hyperdrive, churning out huge amounts of growth-stimulating proteins, or perhaps they get turned off inappropriately, robbing cells of their brakes. Whatever the problem, the result is cells that divide over and over; that take root in parts of the body where they don't belong; that lose the normal instinct to self-destruct when their wiring goes awry; and that they simply never die.

This mess is fueled by hundreds, even thousands, of individual proteins. The process of making new blood vessels alone may entail 30 or 40 of them lined up in several signaling pathways.

With targeted drugs, scientists envisioned bringing the entire process to a halt by chemically knocking out a single protein link in one chain of communication. Like a cut phone line, no signal gets through. Therefore, no more growth and no new cancer.

Although the process has not worked out to be this simple, many in the field believe that eventual progress is likely. Maybe the best-case scenario is spun out by the Dana Farber's Dr. William Kaelin. He theorizes that a few key pathways will prove important in many kinds of cancer, and drugs will be created to effectively block them.

Several dozen drugs may offer enough choices to cover all the typical genetic combinations at work in cancer. "This is the belief that is driving all of us," says Kaelin. With the right mix of three or four drugs for any individual, "we will make a major inroad into the common adult tumors, such as lung, colon and breast cancer."

Though they hoped for more, many cancer specialists seem relieved for now to have anything new to offer at all. Even small advances are welcome in a profession where progress often seems glacial.

"It is slower progress than we'd like, but that's the nature of medicine," says Dr. Donald Trump, chairman of medicine at Roswell Park Cancer Institute in Buffalo, N.Y. "A disease process as complex as cancer is unlikely to yield to blockbuster effects."