Writing about tuberculosis, Susan Sontag argued that when the root cause was understood and effective treatment established, the mythology of the disease dissipated. She suggested that scientists might also find a clear and singular cause of cancer. But the more research has progressed, the less likely that seems—and the more useful imagery has become for understanding. In 2007, Johns Hopkins researcher Bert Vogelstein created a topographical map of the most common mutations in different types of cancer. On a bright green plane, a long winding path represents the human genome. Sharp purple spires rise from genes that are mutated in samples of breast and colon cancer. The maps are used to find patterns in genes turned on and off by cancer—to get beyond the sense of malignant cells as "disorganized mob," as Nuland put it. This is the old topographical language of cancer, turned on its head.
New maps have also given rise to new treatments, which suggest a narrow targeting of patients and should recalibrate our expectations of more than slow declines in cancer mortality. Mukherjee tells the inspiring tale of Gleevec, a highly-specific, nontoxic drug for chronic myeloid leukemia. First came the discovery of a chromosomal abnormality that is particular to this disease and that makes cells hyperactive. Then came tests of designer molecules to quell that hyperactivity. But this approach was never a sure thing: Chronic myeloid leukemia represents a "genetic tornado," as Mukherjee writes. Conceptually, it was never clear whether targeting the original abnormality would make a difference in the face of subsequent mutations.
And all along, there was the question of cost. The drug company Novartis, which owned the compound, hesitated to spend the hundreds of millions of dollars for clinical trials of a drug that might benefit only thousands of patients a year. The academic researcher Brian Drucker, who turned the drug into a personal mission, imagined that as a last resort, "I would make it in my own basement." Novartis did come around, though, in 1998, giving Drucker just enough for a small trial. The results were dramatic: Blood counts normalized, often rapidly. If, as Mukherjee imagines, cancer is our craftier, more clever doppelganger, Gleevec represents an even craftier, more clever response. Indeed today, chronic myeloid leukemia has gone from a fatal disease to a largely manageable and chronic one.
Yet Mukherjee is no Pollyanna when it comes to targeted treatments. Cancer's genetic signatures differ from one kind of cancer to the next, and even from one patient to another. "'Every patient's cancer is unique because every cancer genome is unique,'" he writes, quoting Vogelstein the mapmaker. To make a dent in cancer mortality, even in a best-case scenario, researchers would need a large number of customized molecules and long, expensive clinical trials. Meanwhile, cancers keep mutating. Some patients with chronic myeloid leukemia, for instance, have now developed resistance to Gleevec, and need new, updated treatments. It could be that cancer, "the scrappy, fecund, invasive, adaptable twin to our scrappy, fecund, invasive, adaptable cells and genes, is impossible to disconnect from our bodies," Mukherjee writes grimly.
Mukherjee does not discuss a new therapy, called Provenge, which the FDA approved in April. But it too seems a harbinger of things to come. Provenge relies on a patient's own immune cells to fight cancer. These cells are removed from the body, primed and fortified in a laboratory dish, then infused back in. Provenge has shown some success against advanced prostate cancer. But it also costs about $93,000. Gleevec can cost $4,500 per month. Revlimid, another cutting-edge treatment for multiple myeloma, can cost $10,000 per month. It's hard to see how these prices might come down when the market consists of patients increasingly fragmented not only by type of cancer but even by types of mutation. The answer to how well we manage at outsmarting our awful twin surely depends on our scientific maps and metaphors, but it may well have as much, or more, to do with money.