The Emperor of All Maladies: A Biography of Cancer

The summary

In December 1947, Sidney Farber waited in a basement room under Boston’s Children’s Hospital for vials of aminopterin, a chemical mimic of folic acid. He injected it into Robert Sandler, a two-year-old whose leukemia had already snapped a bone in his thigh. Within weeks the boy’s white-cell count fell toward normal, and for a month he looked just like his healthy twin, Elliott. Then the leukemia returned and killed him. Farber published anyway: sixteen children, ten responses, five alive at six months. In leukemia, six months was an eternity. A chemical had pushed back a cancer — and that door opening, slamming, and a profession deciding it already knew why, is the pattern of this whole history.

Galen’s black bile never really left

Galen taught that cancer was trapped black bile, a systemic fluid, so cutting it out was futile — and surgeons obeyed for fourteen centuries, until Vesalius went hunting for black bile and found none in the body, nor Matthew Baillie inside tumors.

William Halsted then inverted Galen exactly. Cancer, he argued, spun outward from one point like a pinwheel, so a surgeon who cut wide enough would catch it. His radical mastectomy took the breast, the pectoral muscles, the nodes under the armpit, eventually the collarbone, leaving shoulders caved inward and arms swollen — surgical elephantiasis, he called it. His own 1907 figures held the refutation: of sixty women whose lymph nodes were clean, forty-five were alive at five years; of forty whose nodes held cancer, three. Survival tracked how far the disease had already traveled, not how much tissue came out. Halsted saw it and turned away. Not until 1981 did Bernard Fisher’s randomized trial show that radical mastectomy and lumpectomy-with-radiation gave identical outcomes. Half a million women had the operation first.

A war declared before anyone could describe the enemy

Farber’s second discovery was political: a disease has to be sold before it can be cured. In 1948 he and the Variety Club renamed a boy with intestinal lymphoma “Jimmy,” put him on the radio with the Boston Braves crowding his hospital room, and raised $231,000. Mary Lasker, who thought research deserved the advertising budget of toothpaste, supplied the rest. After Apollo 11 the campaign had its metaphor: a moon shot for cancer. Ann Landers ran one column in April 1971, a million letters buried the Senate mailroom, and Nixon signed the National Cancer Act that December.

Nobody could yet say what cancer was. James Watson predicted “a massive expansion of well-intentioned mediocrity,” and the hunt for human cancer viruses swallowed close to $500 million and produced almost nothing. The era’s largest real win came from a mailed questionnaire: Richard Doll and Austin Bradford Hill wrote to 59,600 British doctors, 41,024 answered, and by 1954 thirty-six of them had died of lung cancer — every one a smoker. Prevention research got a fifth of the institute’s grants.

Poison has a ceiling, and the ceiling is us

The cures were real. Frei and Freireich’s four-drug VAMP regimen was called insane and cruel — “we could have killed all of those kids,” Freireich admitted — yet it cured about one child in twenty. Pinkel’s two-year “total therapy” in Memphis left 80 percent of 278 children disease-free, MOPP cured more than half of forty-three advanced Hodgkin’s patients, and cisplatin made metastatic testicular cancer curable.

Then came the generalization: if hard works, harder must work better. “The smiling oncologist,” Rose Kushner wrote, “does not know whether his patients vomit or not.” Across 1984 and 1985, six thousand chemotherapy papers appeared without a single new cure for an advanced solid tumor. The end of the road was STAMP — megadose chemotherapy rescued by the patient’s own frozen marrow — a $4 billion business with waiting lists before any randomized trial reported. The trial that seemed to vindicate it was fabricated; the real one found no survival benefit, nine women dead of transplant complications, and nine more with leukemias caused by their treatment. In 1986, age-adjusting the national figures, John Bailar found cancer mortality had risen 8.7 percent since 1962 and called thirty-five years of effort “a qualified failure.”

The enemy turned out to be a copy of us

The correction came from a chicken tumor: Rous sarcoma virus carried a gene, src, that forced cells to divide. Harold Varmus and Michael Bishop went looking for src’s distant relatives in normal cells and found src itself — in ducks, quail, sheep, cows, a newborn emu at the Sacramento zoo, and in us. The virus had not delivered a cancer gene; it had stolen one on the way out. Cancer, Varmus said in Stockholm, is “a distorted version of our normal selves.”

That explained the ceiling. Chemotherapy attacked rapid growth, which is what living cells do; the way past it was to find something a cancer cell needs and a normal cell does not. Brian Druker spent years pushing a reluctant Novartis to make enough of a kinase inhibitor to test; 53 of his first 54 patients went into complete remission on Gleevec. Herceptin, aimed at the Her-2 gene, lifted survival in early breast cancer by about a third. But when Jerry Mayfield’s leukemia relapsed in 2003, it had changed the shape of the target so the drug could no longer bind. A differently shaped drug put him back into remission. We run to stay in place.

The bottom line

Cancer is not an invader. It is our own growth machinery with the accelerator jammed and the brakes cut, which is why every approach that treated it as a foreign object — bile to purge, tissue to excise, poison to escalate — met the same wall, and why the drugs that finally worked came out of decades of biology nobody could sell as a war. Mukherjee’s measure of victory, borrowed from Richard Doll, is deliberately modest: death in old age is inevitable, death before old age is not.

Read it to watch medicine correct itself in real time — slowly, expensively, and almost always over the objections of the people who were most certain.