Notes - The Emperor of All Maladies - A Biography of Cancer

Siddhartha Mukherjee | April 27, 2026

Part One: "Of Blacke Cholor, Without Boyling"

The Ancient Disease

The earliest known description of cancer appears in the Edwin Smith Papyrus (Egypt, ~3000 BC): "a bulging mass in the breast, hard and cool to the touch." The treatment column reads: "There is none."

The word "cancer" comes from Hippocrates, who used karkinos (Greek for crab) to describe spreading tumors—perhaps because the distended veins surrounding some tumors resembled crab claws.

For most of history, cancer was relatively rare—not because humans were healthier, but because most people died young from infectious diseases before cancer could develop. Cancer is primarily a disease of aging; the modern epidemic is partly an artifact of living long enough to get it.

Humorism

For 2,000 years, cancer was understood through Galenic medicine: caused by excess "black bile." Treatment was purging and balancing the humors. This theory was wrong in every detail but persisted because it was coherent and difficult to falsify.

The transition from humorism to modern medicine required not just new observations but a new ontology—a new way of understanding what disease is.

Farber's Gamble

In 1947, Sidney Farber—a pediatric pathologist at Boston Children's Hospital—attempted to treat childhood leukemia with antifolates (compounds that block folate metabolism, which cancer cells need to grow). His colleagues were skeptical: leukemia was considered uniformly fatal; pathologists didn't treat patients.

Farber's first patients showed temporary remissions. When he publislts in the New England Journal of Medicine, the response was hostility—he was accused of giving false hope. But Farber had found the first crack: cancer could be temporarily reversed by chemicals.

His insight was profound: if leukemia cells grew faster than normal cells because they metabolized folate faster, drugs that blocked folate would kill cancer faster than normal tissue. This was the beginning of chemotherapy.

The "Karnofsky Cocktail"

David Karnofsky at Memorial Hospital in New York was simultaneously developing multi-drug approaches to cancer. His work established that combination chemotherapy—attacking multiple metabolic pathways simultaneously—could produce more durable responses than single-drug approaches.

Part Two: An Impatient War

The National Cancer Act (1971)

Sidney Farber and Mary Lasker—a brilliant philanthropist and political operator—spent two decades lobbying for a "war on cancer." They secured congressional support and finally, in December 1971, Nixon signed the National Cancer Act with great fanfare.

The Act created:

  • The National Cancer Program with a budget of $1.5 billion
  • Direct reporting of the National Cancer Institute director to the President
  • Cancer Centers across the nation

The expectation: a moon shot–style effort would cure cancer within a decade or two.

Why the War Was Harder Than Expected

The early optimism was based on a misunderstanding of what cancer is. Researchers assumed cancer was like polio—a single disease caused by a single agent. If you could find the viral agent (the dominant theory in the 1960s-70s), you could make a vaccine.

Cancer turned out to be hundreds of different diseases with different causes, different genetics, and different responses to treatment. There was no single "cancer virus." The war on cancer would require understanding the fundamental biology of the cell.

VAMP and the Cure of Childhood Leukemia

Meanwhile, clinical results were spectacular in childhood ALL. James Holland, Emil Freireich, and Emil Frei developed VAMP (vincristine, amethopterin, 6-mercaptopurine, prednisone)—a four-drug combination.

The results: in the early 1960s, children with ALL were treated with VAMP at the National Cancer Institute. The 5-year survival rate went from ~5% to ~70%. Today it approaches 90%.

This was one of the great triumphs of medicine—achieved not by understanding the basic biology but through empirical clinical trial and error.

Part Three: "Will you turn me out if I can't get better?"

The Randomized Trial Revolution

Before the 1970s, cancer therapy was evaluated by anecdote and clinical impression. Surgeons claimed high cure rates based on patients who survived—without acknowledging that survivors were a biased sample.

Archie Cochrane and the British Medical Research Council established the randomized controlled trial (RCT) as the gold standard for medical evidence. Applied to cancer, RCTs produced shocking results:

  • Radical mastectomy (removing the breast, underlying muscle, and axillary lymph nodes) was NOT superior to simple mastectomy for breast cancer survival
  • Many established treatments had never been rigorously tested

Bernard Fisher's NSABP (National Surgical Adjuvant Breast and Bowel Project) trials demonstrated that breast cancer metastasizes early—before surgery can prevent it. The implications: less radical surgery was equally effective, and adjuvant therapy (chemotherapy after surgery) was needed.

Halsted's Paradigm and Its Collapse

William Stewart Halsted developed the radical mastectomy in the 1880s, based on the theory that cancer spread concentrically—from primary tumor outward through lymph nodes. If you removed enough tissue, you could cut ahead of the disease.

Fisher's trials showed this theory was wrong. Cancer spreads through the bloodstream, not concentrically. Radical surgery couldn't "outrun" the disease if it had already metastasized systemically.

This paradigm shift—from cancer as a local disease to cancer as a systemic disease—was one of the most important in 20th century oncology.

Part Four: Prevention Is the Cure

Tobacco and Lung Cancer

In 1950, Richard Doll and Austin Bradford Hill published a study in the British Medical Journal establishing that cigarette smoking caused lung cancer. The evidence was overwhelming: smokers had 14x the lung cancer rate of non-smokers; the more you smoked, the higher your risk; stopping smoking lowered the risk over time.

The tobacco industry's response: manufacture doubt. They funded counter-research, hired scientists to dispute the evidence, and maintained the "controversy" for 40+ years. The playbook was later used for climate change denial.

The regulatory response took decades:

  • 1964: US Surgeon General's report conclusively linked smoking and lung cancer
  • 1965: First cigarette health warning labels (US)
  • 1988: Tobacco advertising banned from TV in UK
  • 1998: $246 billion settlement with state attorneys general

The public health impact: US smoking rates fell from 45% in 1965 to about 15% today. This prevented millions of cancer deaths.

The Logic of Prevention

Prevention turned out to be the most effective cancer intervention. Tobacco causes ~30% of all US cancer deaths. The HPV vaccine (targeting cervical cancer) is one of the most effective cancer prevention tools ever developed. Sun protection prevents most melanomas.

The challenge: prevention is invisible. You can't see the cancer that didn't happen. But the statistical impact is enormous.

Part Five: The Molecular Revolution

Oncogenes

In the 1970s, researchers studying cancer-causing viruses discovered that these viruses didn't bring entirely foreign genes—they captured and mutated normal human genes called proto-oncogenes.

Harold Varmus and Michael Bishop showed that the src gene in the Rous sarcoma virus was actually a mutated version of a normal cellular gene. They won the Nobel Prize in 1989.

The implication: cancer is a disease of mutated normal genes. Every cancer cell contains the mutated remnants of your own genome.

Tumor Suppressor Genes

Alfred Knudson's "two-hit hypothesis" (1971): some cancers required two genetic "hits" to develop. In familial cancers (like retinoblastoma), patients were born with one hit already in place; a second somatic mutation triggered the disease.

The RB1 gene was the first tumor suppressor identified—it normally prevents uncontrolled cell growth. When both copies are mutated, the brake is released and cells proliferate.

The Oncogene Decade

The 1980s-90s saw rapid discovery of cancer-causing mutations:

  • ras mutations: found in ~30% of all human cancers
  • p53 mutations: the most common cancer-associated genetic change; found in >50% of human cancers
  • BRCA1/BRCA2: breast cancer susceptibility genes
  • HER2/neu: amplified in ~25% of breast cancers

This genomic revolution transformed cancer from a mysterious black box to a disease that could be categorized, analyzed, and targeted.

Part Six: Targeted Therapy

Gleevec: The First Targeted Drug

Chronic myelogenous leukemia (CML) is caused by a specific chromosomal rearrangement—the "Philadelphia chromosome"—that fuses the BCR and ABL genes, creating an abnormally active tyrosine kinase.

Brian Druker, Nicholas Lydon, and Charles Sawyers developed imatinib (Gleevec): a small molecule that fits into the ATP-binding pocket of BCR-ABL and turns it off.

Clinical results (2001 paper in New England Journal of Medicine): complete hematologic response in 98% of CML patients in chronic phase. Durable responses—many patients still in remission a decade later.

Gleevec was the proof of concept for targeted therapy: if you know the molecular driver of a cancer, you can design a drug to turn it off.

Limitations of Targeted Therapy

Cancer rapidly develops resistance to targeted drugs through mutation. In CML, resistance emerged as BCR-ABL mutated its binding pocket. Second-generation drugs (dasatinib, nilotinib) were developed to overcome specific resistance mutations.

The more fundamental problem: most solid tumors have multiple driver mutations, not a single target. Unlike CML (which is essentially defined by BCR-ABL), lung cancer or breast cancer may have 5-10 different driver mutations—requiring different drugs for different patients.

Epilogue: Lessons of the Emperor

What Cancer Is

Cancer is not a foreign invader—it is the product of mutations in our own cells. Normal processes—cell growth, DNA repair, programmed cell death—go wrong in specific, predictable ways. Cancer is the body attacking itself.

The mutations accumulate over decades, driven by:

  • Carcinogens (tobacco, UV radiation, certain chemicals)
  • Oncogenic viruses (HPV → cervical cancer; hepatitis B → liver cancer; EBV → some lymphomas)
  • Random copying errors during cell division (responsible for ~2/3 of cancer mutations)
  • Inherited susceptibility (BRCA1/2, Lynch syndrome, retinoblastoma gene, etc.)

What Has Changed

From Sidney Farber's first remissions in 1947 to Gleevec's targeted precision in 2001, cancer medicine has been transformed:

  • Childhood ALL: 5% survival → 90% survival
  • CML: uniformly fatal → manageable chronic disease with near-normal life expectancy
  • Hodgkin lymphoma: near-uniformly fatal → ~85% cure rate
  • Cervical cancer: preventable with HPV vaccine
  • Melanoma: immunotherapy producing durable remissions in many patients

What Has Not Changed

Most common solid tumors—lung, colon, pancreatic, breast cancer—remain difficult to cure once they've metastasized. The fundamental challenge: by the time most cancers are diagnosed, they have already spread.

Early detection remains the most promising area for improving outcomes. When most cancers are caught early, they're highly curable; when caught late, they're not.

The Central Insight

Cancer is not one disease but hundreds—each driven by a unique constellation of genetic mutations, each requiring tailored treatment. The "war on cancer" failed not because the effort was insufficient but because the target was wrong. There is no single cure for cancer—there are hundreds of potential cures, one for each cancer type and subtype.

The optimistic view: the molecular revolution has given us the tools to understand each cancer's specific biology. The era of truly personalized cancer medicine—treating each tumor based on its genetic signature—has begun.

The humbling truth: we are still in the early stages. Cancer remains the emperor of all maladies—ancient, ubiquitous, and formidable. But for the first time in history, we know what it is.