It's the Complexity, Stupid

Since President Richard Nixon declared war on cancer in 1971, scientists have unraveled the molecular systems that healthy cells use to multiply, grow, repair internal damage, and even die. And they've discovered the changes that allow cancer cells to subvert these systems. Yet all this information underscores the enemy's ingenuity and complexity--and explains why a cure still eludes.

Few wars in history have been as protracted as the war on cancer--or as costly. The search for a cure has consumed billions of dollars and thousands of scientists' careers, while the disease has killed millions of people.

Most of those victims are people over 65, cancer being by and large a disease of the elderly. According to one study, octogenarians are 1000 times more likely to die from cancer than are people in their 20s. The incidence rises with age because cancer cells contain mutations in multiple genes, and acquiring those hits can take time.

Like electronic components on a circuit board, these defective genes help form a complex network of overlapping pathways that drive the development of cancer. And as physicians and researchers working to treat cancer have found, short-circuiting that genetic web is not a trivial task.

"Cancer cells are not a foreign invader like a virus that we can develop a vaccine against," says Lance Liotta, chief of the Laboratory of Pathology and Section on Tumor Invasion and Metastases at the National Cancer Institute (NCI) in Bethesda, Maryland. Because cancer cells arise from the body, they are largely indistinguishable from their normal cousins. And that's one thing that makes treatment so challenging. "Anybody can kill a tumor," scoffs Andrew Yen, a cell biologist at Cornell University. "The problem is sparing the normal cells."

Cancer cells are also hard to defeat because they are genetically plastic: As they divide, they accumulate mutations--some of which offer them a survival advantage, allowing them to outgrow their neighbors or defy treatment. "Their genomes are mutable, which means they can continually be tinkering and trying out new permutations of genes that resist a therapy being imposed on them by an oncologist," says biologist Robert Weinberg of the Massachusetts Institute of Technology in Cambridge, who isolated the first cancer-causing gene from a human tumor in the early 1980s.

Since then, doctors have made progress in treating certain cancers. Many tumors can be removed surgically when they're caught early. "And we know how to treat some of the blood cancers therapeutically," says Weinberg. "But cancers of the lung, pancreas, prostate, and breast--when they're advanced--are as untreatable now as they were 30 years ago. So the question becomes, Does any of what we've done amount to more than a hill of beans?"

Weinberg's answer is a qualified yes. Only by understanding the molecular and genetic mechanisms of cancer, he says, can we develop effective treatments. For example, a drug called Gleevec, designed to block specifically the activity of a protein that promotes cell proliferation, has spurred remission in many patients with a particular form of leukemia. And two new studies appearing online in Science and the New England Journal of Medicine show that a drug called Iressa shrinks tumors in lung cancer patients whose tumors have a specific genetic alteration.

Although Iressa's effectiveness is limited to the 10% of patients with this specific alteration, scientists predict that its success offers the promise of targeted therapy for cancer in general. "There are likely Achilles' heels for almost all cancers, and we just haven't identified them," says Todd Golub, a geneticist at Harvard University.

But identifying a cancer's Achilles' heel does not guarantee victory, because even well-studied molecules can behave in unexpected ways. For example, in a paper published in March in Molecular Cell, researchers examined the behavior of NF-kB, a protein that controls the activity of many genes involved in proliferation and the cell's response to stress. Scientists have long observed that NF-kB is overactive in tumors, where it turns on genes that prevent cells from self-destructing. By keeping damaged cells from committing suicide, NF-kB can promote the development of cancer.

But in the new study, scientists discovered that NF-kB can just as easily behave in the opposite fashion--silencing genes that prevent cell death. Whether NF-kB promotes cell death or prevents it depends on how the protein is activated: Radiation and chemotherapy can persuade NF-kB to encourage cell suicide, whereas the inflammatory molecules and oxygen-poor environment found inside tumors prime the protein to stifle cell death. This discovery suggests that researchers need to learn more about how NF-kB is likely to respond in a given physiological situation before trotting out therapies that inhibit it. And it confirms a new consensus among scientists: Instead of pointing a finger at an individual gene or protein, researchers need to treat cancer as a complex biological system.

"Scientists tend to think in logical sequence; that's how we're trained," explains Yen. But the processes that orchestrate cell growth and division are not linear, he says; they're part of a "network of overlapping, redundant, and interacting processes."

Treating cancer-causing genes as part of a web might help scientists devise a way to gain the upper hand. "If we can design a therapy to hit that circuitry at different points in the cell, we think it will have much more effectiveness and less toxicity," says Liotta. Taking such a coordinated approach could also circumvent cancer cells' tendency to rewire themselves and evade treatment with single drugs.

NCI, in collaboration with the U.S. Food and Drug Administration (FDA), is conducting clinical trials in which researchers are tracking the activity of hundreds of proteins in patients' cancer cells before, during, and after treatment. "Not every patient is going to have the same wiring," says FDA microbiologist Emanuel Petricoin, who co-directs the joint research program, even for the same type of cancer. Instead of blasting all cancers with preset combinations of drugs and radiation, scientists now foresee a time when a cocktail of drugs can be tailored to destroy a tumor by targeting key points in its particular circuit board.

Nobody predicts an easy victory. Scientists agree that the war's new battles are going to be fought on multiple fronts, using years of hard-won intelligence to take out the enemy's lines of communication.

Sara Solovitch is a freelance writer in Santa Cruz, California, who has witnessed battles both won and lost.