Support the news
A dear friend died of lung cancer this week.
She didn’t smoke and she had no other risk factors. A retired nurse and avid cook, she ate an exemplary diet, maintained a lithe figure, and got an impressive amount of exercise climbing mountains and clearing trails.
Her death has left her many friends wondering “why her?” Her example undercuts the notion that any of us can ultimately do much to avoid an ugly death from cancer.
Dr. Bert Vogelstein of Johns Hopkins Medical Institutions says my late friend exemplifies the 50 percent or so of cancers he calls “unpreventable.”
"People want a reason why these cancers happen," Vogelstein says. "Well, here’s the reason: Cells make mistakes. That’s just bad luck."
Those mistakes are like typos in a giant manuscript. Each time a cell divides, it mis-copies several molecules in the long chain of DNA that makes up our genes. Most of the time these biochemical mistakes are harmless. But sometimes they occur in genes that keep cancer in check. Once some of these mutations accumulate, the cell breaks free of normal constraints on its growth -- that is, it becomes malignant.
"People want a reason why these cancers happen. Well, here’s the reason: Cells make mistakes. That’s just bad luck."Dr. Bert Vogelstein
Two-thirds of the mutations that give rise to cancer are these random copying errors, according to Vogelstein, biostatistician Cristian Tomasetti and their colleagues at Johns Hopkins University. Another 29 percent of cancer-causing DNA damage, they say, comes from environmental and behavioral factors such as cigarette smoking and obesity. The remaining 5 percent are hereditary defects.
Vogelstein is a leading light in cancer research. He pioneered the concept that there are genes that suppress cancer and identified mutations in those genes that underlie many cancers.
The Hopkins researchers have been saying for two years that most cancer-causing mutations are these inescapable random errors at the heart of our most fundamental cellular workings. But their 2015 paper was heavily criticized, partly because it considered only U.S. cancer cases -- other nations may have different or more environmental triggers. Also, the earlier report left out breast and prostate cancers, two of the most common types, so critics said it wasn’t valid to generalize the findings to all cancers.
Their new paper, just published in Science, includes breast and prostate cancer in the 17 cancer types they analyzed. And it incorporates evidence on cancer incidence from 69 countries that account for two-thirds of the world’s population.
"This is the very first time that someone has ever looked at the proportion of mutations within each cancer type and assigned them to these three factors," Tomasetti said during a press briefing, referring to environment, heredity and random copying errors. "I consider this a completely new result compared to our previous one and a really fundamental one."
Their argument relies on a mathematical model that estimates the number of stem cells in each tissue type (the cells that must replicate to renew the tissue) and the number of times these cells divide over a lifetime. The rate of random mutations is fairly constant in different organs, they say, but different tissues turn over at different rates.
For instance, cells that make up the lining of the large intestine renew themselves every four days. That gives them more opportunity to accumulate mutations in the genes that drive cancer than cells in, say, the bone or brain. So it makes sense that colorectal cancer is the fourth most common major cancer, after lung, breast and prostate cancers.
What it doesn't mean
The finding that two-thirds of cancer-causing mutations are due to random mistakes does not mean that two-thirds of all cancers are due to “bad luck.” That’s because many cancers arise from a mix of factors.
Since it takes several mutational “hits” to kick-start a cancer, one or more of these might arise from internal random copying errors while others might come from external environmental sources, such as tobacco smoke. An individual born with a mutation that predisposes her to cancer might not get the disease unless or until random mutations in cancer-driving genes accumulate.
Breast cancer is an example of mixed causes. Most of the underlying mutations are due to random mutations, Vogelstein says. Some are due to an environmental factor such as obesity, age of first menstruation or whether a woman breast-fed. And about 10 percent stem from a strong hereditary predisposition, such as a BRCA gene.
Vogelstein says my late friend appears to represent a "pure" case of random bad luck, since smoking and behavioral risk factors were apparently not in play and lung cancer is not known to have a significant hereditary component.
The Hopkins team’s findings, they say, are consistent with prior estimates that about 40 percent of all cancers are preventable, 5 to 10 percent are due to inherited cancer genes, and around half are not preventable -- since there’s no way to avoid random DNA copying errors.
But what about environment?
But the work has been controversial up to now, and may remain contentious. It disturbs those who believe that changes in environment and behavior deserve top priority in the war against cancer.
“Concluding that ‘bad luck’ is the major cause of cancer would be misleading and may detract from efforts to identify the causes of the disease and effectively prevent it,” Dr. Christopher Wild, director of the World Health Organization’s International Agency for Cancer Research, complained after the first Vogelstein-Tomasetti paper came out.
Anne McTiernan of the Fred Hutchinson Cancer Research Center in Seattle remains unconvinced. "They say randomness is a major cause of cancer, and they said it last time," she told The Washington Post. "But the data doesn’t convince me."
One group from Stony Brook University in New York did an analysis of the 2015 paper and concluded that “extrinsic” factors, such as environmental toxins, account for 70 to 90 percent of cancers while “intrinsic” ones – the random copying errors when cells divide – are to blame for only 10 to 30 percent.
The leader of the Stony Brook team says he’s not convinced by the new report.
Vogelstein is philosophical about his critics. “It’s never easy to overturn dogma,” he said in an interview. “Like every new concept or idea that goes dramatically against the current paradigm, it will take awhile. It will take other people doing measurements we haven’t performed yet or even thought of. And hopefully people will accept it.”
The implications for the way we think about cancer, and the strategies to fight it, are profound.
Most immediately, it might lessen the self-blame many cancer victims and their families feel -- the sense there’s something they did, or didn’t do, that led to their diagnosis.
“We hope this research offers comfort to the literally millions of patients who have developed cancer but who have led near-perfect lifestyles -- who haven’t smoked, who have avoided the sun without sunscreen, who eat perfectly healthy diets, who exercise regularly, who’ve done everything that we know can be done to prevent cancer but they still get it,” Vogelstein says.
Self-blame is particularly poignant for parents of children who get cancer. The Hopkins group believes that childhood cancers are entirely due to random mutations -- except for the small percentage related to inherited gene defects.
When such parents read online that cancer is caused by environment or heredity, they often conclude they’re somehow to blame. "That causes a tremendous amount of guilt," says Vogelstein, a practicing pediatrician in his early career.
But the larger implication concerns the best strategy for fighting cancer.
Not preventing but nipping in the bud
It’s not that classic prevention efforts, especially smoking cessation and healthier diets, don’t deserve our continuing attention. But if around half of all cancers are due to random mutations -- the price of biological mechanisms necessary for life -- then there’s no way to prevent those. That means we need to rely on early detection and treatment to maximize cures and minimize deaths.
“We believe the first step in developing these strategies is simply recognizing these enemies exist, there are a lot of them, and they’re already here, inside us,” Vogelstein says.
Right now, he says, funding for early detection and treatment is “minuscule” compared to that devoted to curing advanced, late-stage cancers.
One of the hottest areas of cancer research focuses on spotting the earliest signs of cancer, using such things as “liquid biopsies” to screen bodily fluids for evidence of cancer-causing mutations or telltale proteins produced by errant genes.
Vogelstein is prominent in this effort and has a personal stake such research. He’s the co-founder of two companies, PapGene and Personal Genome Diagnostics, that are trying to develop such high-tech diagnostics.
But early detection and treatment of cancer is fraught with pitfalls, as controversies over the risk-benefit ratio of mammography to detect early breast cancer and PSA testing for prostate cancer amply show. In large part that’s because such tests often find cancers that don’t need to be treated -- that is, they’d never cause trouble if they weren’t treated. Doctors call this “over-diagnosis.”
“There is certainly a problem with potential over-diagnosis,” Vogelstein acknowledges. “But that doesn’t mean we should abandon the effort, especially now that we’re cognizant that many cancers aren’t preventable.”
If Vogelstein and Tomasetti are right, they say, there’s really no choice but to focus much more effort on nipping cancers in the bud. That's a long-term project.
Right now, for instance, many cancers are out of the gate and down the road before they're discovered. My friend's lung cancer, like so many with this fearful diagnosis, was too far advanced by the time it was found to even be slowed down by the most aggressive treatment.
So after living the healthiest life you can imagine, she spent her last days coping valiantly with the debilitating effects of the best cancer therapy her doctors could offer. It wasn't enough.