‘Carcinogen Hunters’ Investigate Causes Of Inexplicable Cancers
Frustrated cancer patients worry they may have been exposed to unknown carcinogens. Researchers are on the hunt.
The following is an excerpt from The Emperor of All Maladies: A Biography of Cancer by Siddhartha Mukherjee.
When you purchase products through the Bookshop.org link on this page, Science Friday earns a small commission which helps support our journalism.
Buy The Book
The Emperor of All Maladies: A Biography of Cancer
In the summer of 2008, I walked into the office of a New York editor with a large brown paper bag. In it were six hundred pages of my book The Emperor of All Maladies. My mother had looked at the brick-thick, black-bound pages as I left my house and called it the “telephone directory of Hell.”
I sat in the editor’s waiting room, high above Sixth Avenue, looking outside. It was one of those magical New York afternoons, when the balminess in the air becomes intoxicating, and crowds had gathered on the streets. But I was a trainee in oncology in Boston, and all I could see was a landscape of future grief and anxiety. The woman laughing breezily by the bakery might be diagnosed with cancer (ovarian? breast?) in a few years. I imagined the man smoking a cigarette with evident pleasure in a hospital gown as he went for a lung CT scan. The lenses with which I saw the world had forever been changed.
In the fifteen years since the publication of The Emperor of All Maladies, the collective lenses through which we all see cancer have also forever been changed. New horizons have emerged in the early detection of cancer, while innovations in treatment, built on extraordinary work in immunology and cell therapy, have widened the tantalizing world of personalized medicine. Some cancers considered untreatable only a decade ago have been eradicated in individual patients. And yet, the disease continues to surprise and frustrate us. There is no “one cancer,” and there will be no “one cure,” and with each apparent step forward, the vast tangle of cancer’s complexity comes more vividly into view. The complete eradication of cancer through prevention remains a fantasy: we hunt for new classes of carcinogens, yet the identification of carcinogens does not, on its own, explain why a tumor will develop in one person and not another. Technologies purporting the early detection of cancers sprout daily—but are yet to demonstrate that patients diagnosed with these technologies will have improved outcomes. And even with new therapies, many patients’ cancers remain and become resistant to treatment. Meanwhile, the cost of cancer treatment skyrockets, presenting an insurmountable barrier to the equitable distribution of medical care. Through all this, “cancer” remains a meme for everything that is malignant in our society and culture.
When I began writing The Emperor of All Maladies, I had in mind a kind of journal that would allow me to make sense of the world of cancer I experienced as a young physician in medical oncology. As I wrote, the book became something far greater than I first imagined: a “biography” of cancer that aimed to illuminate humanity’s engagement with this terrifying and beguiling ailment. I would tell the story through the intersection of lives and ideas, scientific theory and medical practice, culture and belief, traced across the sweep of history as well as within the intimacy of a single patient’s diagnosis and treatment.
In a sense, the chapters that follow constitute an entirely new Emperor. In a decade and half, there has been so much more learning that I realized that only a new book would capture the horizons ahead. The new chapters are organized simply, along the three pillars of our strategy against cancer: prevention, screening, and treatment, capped with a chapter on a new vision of cancer. New vistas have opened yet more new vistas—for exploration, science, and medicine, and for the evolving cultural metaphors that cancer evokes.
Prevention. Screening. Treatment. And the future. I welcome you to a new book.
The Carcinogen Hunters
In 2011, I spoke to an audience of cancer patients at various stages of their illness at a wellness retreat in upstate New York. My subject was the history of cancer, and I had just finished talking about carcinogenesis, mutations, and familial genetic risk when a woman with penetrating blue eyes and a blue turban raised her hand.
“I have ovarian cancer, but I never had any family history and was never exposed to anything unusual. How do you explain that?”
I mumbled an apologetic response about the possibility that her cancer might be the result of random mutations in her ovarian cells. Random, stochastic mutations in dividing cells do indubitably cause cancer.
But the woman found my answer ridiculous.
“And what do you doctors and scientists make of the fact that my aunt and my friend both have lung cancer? And that all of us happen to live next to one of the busiest turnpikes in New Jersey?” Her voice had risen a few pitches. There was a sharp accusatory note in it.
I apologized again.
“I’m sorry. Sometimes we find cancer clusters that are coincidental, like grains of rice that form seemingly organized patterns when thrown on the ground.” I was on the defensive, stammering, sweating. “But sometimes, the clusters are real, and we just don’t know.”
The epidemiological puzzle was no longer abstract. The audience stared at me, visibly restive, waiting for explanations, answers. I shifted uncomfortably on my feet, and moved on to further questions.
Another woman raised her hand. “I’m the only one in my family to ever have breast cancer. I have no risk factors,” she said. “What was I exposed to?”
The questions continued in a flood of anxiety and anger. Why do people who seemingly have no exposure to known carcinogens get cancer? Are there “hidden” carcinogens in our environment—in food, air, and water—that we are missing? Why haven’t researchers found ways to identify and prevent exposures to these hidden carcinogens? If prevention is the key to decreasing cancer incidence and mortality, and if finding carcinogens is crucial to prevention, then is it possible that we are we missing common carcinogens? What are the frontiers of cancer prevention?
And so it went. The conversation devolved into a series of accusations; and the evening came to a tense, halting end. I watched the men and women streaming out of the door, just as puzzled as they had been when they had walked in. The air was thick with exasperation.
The last phrase I overheard, a whisper from one of the attendees, was a damning, “A waste of time.”
The soaring frustration that I encountered that evening in the packed auditorium—the haunting image of the woman in a turban searching for an answer that neither she nor I could find—echoed a frustration that soared through the field of cancer prevention in the early 2000s. The premise that echoed repeatedly was accurate: prevention is the most impactful way to affect cancer-specific mortality. The identification of tobacco as a major preventable carcinogen was the single most significant driver of the decrease in cancer mortality in the United States between 1960 and 1990; other nations that had managed to control tobacco exposure also witnessed a striking decrease in cancer mortality.
A wave of exuberance followed this decrease, and by the early 1980s, the exuberance had morphed into pugilistic optimism about the future of cancer prevention. In 1984, Richard Peto, at Oxford University, then a young statistician and epidemiologist, wrote a defiant article about cancer prevention that was meant to provoke action (the article was accompanied by a photograph of Peto, with disheveled hair and a grim, determined face, looking like a boxer about to enter the ring). Why couldn’t the paradigms gleaned from the tobacco studies be applied to a host of other cancers? Peto asked. Why mightn’t fleets of epidemiological studies to document and understand every aspect of human behavior and exposure—“sexual, dietary, drinking or smoking habits, or [. . .] reproductive history”—provide clues to identifying preventable human carcinogens? What if there was a cosmos of environmental and behavioral carcinogens hiding in plain sight, and we just needed the energy, drive, and resources to find them?
Related Segment
‘A Multi-Headed Beast’: Telling The Story Of Cancer
How, then, might such “hidden” carcinogens be found? Broadly speaking, carcinogen hunters use two kinds of tools: epidemiological studies, which seek to identify cancer-causing agents by examining exposure and risk in human populations, and biological studies, which attempt to find a carcinogen using experimental methods. Each system of knowledge has its limitations. Epidemiologists can find correlations between an exposure and a heightened risk of cancer—but correlations are not causes (and correlations do not identify any mechanism by which a substance causes cancer). Biologists, in contrast, use experimental systems, such as lab experiments on animals and cells, to catch cancer-causing culprits—but extrapolating these exposures to human carcinogenesis is difficult. It is as if the two disciplines view carcinogenesis using two radically different lenses—one telescopic and one microscopic.
Yet, differences aside, between the 1980s and the late 1990s, both systems of knowledge produced seminal results. The Million Women Study—a prospective study that recruited more than a million women aged between fifty and sixty-four in the United Kingdom between 1996 and 2001—found that women on hormone replacement therapy (HRT), with estrogen alone or the estrogen-progestin combination, had a higher risk of breast cancer. Women who took estrogen-alone HRT, particularly those with obesity, also had a higher risk of uterine cancer. (Notably, these results were somewhat different from those of the Women’s Health Initiative, conducted in the United States, which showed that estrogen-progestin combination HRT is linked to a higher risk of breast cancer in postmenopausal women but not to endometrial cancer. Estrogen-alone HRT did not cause a higher risk of breast cancer, but did cause a higher risk of uterine cancer. These studies are still in progress, although some epidemiologists have argued that the UK study had potential biases and the causal association between HRT and breast cancer has not definitively been established.)
Epidemiological studies have also revealed that alcohol is a carcinogen. Alcohol in combination with cigarette smoking has long been known to cause cancer of the esophagus. But the independent risk of alcohol for a variety of cancers—particularly for head and neck cancer, and colorectal cancer—is a relatively new finding, leading the National Cancer Institute to release a special directive to reduce alcohol consumption.
A second branch of investigation—one that predates the epidemiological studies described above—also revealed the identity of a potent preventable carcinogen. In the mid-1960s in Dusseldorf, a young virologist named Harald zur Hausen, frustrated with the pace of his own research, was on a soul-searching walkabout in his lab when he retrieved a crumpled letter from a wastepaper basket. The letter, from a prominent scientist in America, was soliciting a young researcher to work in virus-isolation from cancer cells in Philadelphia.
He applied and was accepted, and in 1966, zur Hausen began working at the Children’s Hospital of Philadelphia. There he developed methods to isolate the Epstein-Barr virus—the first cancer-linked virus—from lymphoma cells (the virus had been identified in lymphomas by Anthony Epstein, Bert Achong, and Yvonne Barr). He was deeply influenced by the Polish-American virologist Hilary Koprowski, who strongly believed that some cancers were caused by yet undiscovered viruses.
Zur Hausen moved back to Germany in 1969. His interest turned to genital warts, which, he knew, were sexually transmitted and could eventually develop into cervical cancer in women. His decades-long immersion in virology and cancer had sharpened his instincts: a con-tagious, transmissible cancer-causing virus—a “relatively poorly studied human wart virus”—must be involved. Working with two scientists, Georg W. Bornkamm and Lutz Gissmann, he obtained surgical samples of human warts, pulverized them by grinding them with sea sand, and spun them in a simple centrifuge. Under powerful microscopes, he detected particles of novel virus types—papilloma, or wart-related, viruses.
Continuing his work with his colleagues, zur Hausen created molecular “probes” to detect signs of this new virus in tissues, hoping to seal their case linking cervical cancer to a papilloma virus. But the tests were strikingly negative: the virus, called HPV6, was not present in cervical cancer. It was a devastating result. The ridicule of his work by many scientific colleagues haunted him: “Everyone knows that warts and papillomaviruses are harmless,” critics had insisted. Had zur Hausen spent decades chasing a mirage?
But then an incidental strand of thought gripped zur Hausen: What if there were multiple strains of the virus, and only some caused cervical cancer—and he had just stumbled on the wrong strain? In a last-ditch effort to save a decade of work, zur Hausen and his students found another batch of genital warts, and set out to find new molecular probes. In 1983, using new probes against different HPV strains, zur Hausen’s team found a second strain, which they called HPV16. These probes were strikingly positive: the virus was present in nearly 50 percent of cervical cancer samples. Subsequent work revealed that two of about a dozen strains of HPV can be detected in a large percentage of cervical cancer samples.
Zur Hausen’s discovery set off a race to find a vaccine for can-cer-associated HPV strains. In the early 2000s, a team led by Doug Lowy and John Schiller, working at the National Cancer Institute, helped create a vaccine that would prevent HPV infection by multiple strains. Vaccination against a particle expressing proteins from the pathological variants of HPV proved to be so remarkably effective that even zur Hausen was taken by surprise. In a Swedish study, for instance, 1,672,983 vaccinated and unvaccinated girls and women were fol-lowed from 2006 through 2017. Cervical cancer was diagnosed in 19 women who had received the HPV vaccine and in 538 women who had not received the vaccine—a striking difference. More recent controlled studies have found a nearly 100 percent prevention of cervical cancer in women vaccinated against HPV. Zur Hausen won the 2008 Nobel Prize for establishing the connection between HPV and cervical cancer.
Excerpted from THE EMPEROR OF ALL MALADIES by Siddhartha Mukherjee. Copyright © 2025 by Siddhartha Mukherjee. Reprinted with permission of Scribner, a Division of Simon & Schuster.
Meet the Writer
About Siddhartha Mukherjee
Dr. Siddhartha Mukherjee is a cancer physician and an associate professor of medicine at Columbia University. He is the author of the Pulitzer Prize-winning book, The Emperor of All Maladies: A Biography of Cancer.