By Elaine Schattner, November 1, 2013
Last month the World Health Organization’s cancer arm, the International Agency for Research on Cancer (IARC), declared that air pollution is a carcinogen. While physicians and others have long intuited a relationship between poor air quality and malignancy, few scientific organizations officially recognized the link. Proof was lacking.
The new designation rests on findings of a large analysis published in Lancet Oncology. The ESCAPE trial drew data from registries involving over 312,000 individuals in nine European countries followed for over 12 years, on average. The investigators used several models to evaluate possible associations between the sizes, types, and density of particles in air, or traffic patterns, and lung cancer cases. They checked for social and economic factors, like fruit consumed per person, age, sex, and education levels that might confound the data, and assessed smoking habits. A total of 2,095 lung cancer cases emerged within the study population. The researchers observed a clear, positive correlation between the amount of particulate matter in air sampled near a person’s address and the odds of developing lung cancer. They also identified a weaker but statistically significant link between the volume of road traffic near an individual’s home and the chances of a lung cancer diagnosis.
“It’s an important statement,” said Regina Santella, a professor of Environmental Health Sciences at Columbia University. What’s unusual about the IARC decision is that it refers to air pollution in general, rather than pointing to a specific chemical or toxin, she considered. “This is something the lay public will understand.”
“Exactly what’s in the air fluctuates from day to day. It varies within a city, and across neighborhoods,” Santella said. It’s a complex problem because air pollution includes a wide range of compounds. Potentially harmful chemicals enter air in cities from cars and trucks, industrial plants, and in rural areas from burning leaves and other sources. Pollutants vary not just by chemical composition, but by their size and capacity to penetrate deep in airways. “For example, you might want to look at particles that are less than 2.5 microns in diameter. Those tiny bits of airborne material have major health effects,” she said.
The U.S. National Toxicology Program (NTP) does not currently list air pollution as a cause of lung or other cancers. “We haven’t looked at air pollution as a single entity,” said John Bucher, a scientist and NTP director. The interagency group is charged with evaluating tens of thousands of chemicals for toxicity, and reporting those findings to the public. It works under the auspices of the National Institute of Environmental Health Sciences (NIEHS), and collaborates with the Environmental Protection Agency (EPA), Food and Drug Administration (FDA), and Occupational Safety branch of the Centers for Disease Control and Prevention (CDC). The NTP usually publishes a biennial report on carcinogens. The last issue appeared in 2011. Meetings to review information for the upcoming issue were delayed due to the government shutdown, Bucher said. The updated report will be released in 2014.
“The NTP is not just a cancer program. We look at all kinds of health harms,” Bucher said. It has examined particular components of air pollution, including mercury, lead, ozone, benzene, and other molecules. The problem of air pollution is vast, due to the sheer numbers of toxins emanating from diverse sources throughout the country. Coal-fired power plants may account for as much as 40 to 50 percent of U.S. air pollution, he said. In the ESCAPE trial, the researchers looked at the concentration of nitrogen oxides in air samples and did not find a relationship with lung cancer. The NTP has not specifically studied that class of compounds in air pollution, but it does plan for an analysis of sulfur dioxide—another product of coal-power plants.
Petrochemical plants generate thousands of distinct chemical compounds. Not all are harmful. The EPA offers an interactive map of reported emissions from industrial facilities. But those data don’t tell all. Polycyclic aromatic hydrocarbons (PAHs) represent another major factor in air pollution. Those, typically byproducts of combustion, occur in some 1,500 distinct forms. “But all we know about this large class of agents is based on studies of only a dozen or so compounds,” Bucher said. “Studies so far are inadequate.”
To get a more precise—and rapid—handle on potential toxins, the NTP is working with the NIH Chemical Genomics Center and EPA’s Computational Toxicology Center to develop high through-put screening assays. These tools won’t necessarily prove that a compound causes cancer or another illness. “But we can measure how thousands of chemicals alter gene expression in cells,” Bucher said. The modern assays, with tiny wells and computerized readouts, enable scientists to analyze many combinations of compounds, at varying doses. “We’re hoping, in the future, to use these systems to understand how chemicals affect molecular pathways that cause disease,” he said.
One detail that might draw an oncologist’s attention is the relationship between air pollution and a particular form of lung cancer called adenocarcinoma. In the Lancet Oncology report, the link with air pollution was strongest for this cancer type. Among non-smokers with lung cancer, adenocarcinoma is the most common pathology. Taken together, these findings support the contention that environmental toxins play a role in this form of the disease. In general, lung cancer remains the leading cause of cancer-related deaths, accounting each year for nearly 160,000 deaths in the United States, and some 7.6 million deaths worldwide.
Bernard Goldstein, a physician, environmental toxicologist, and former dean of the Graduate School of Public Health at the University of Pittsburgh, has authored hundreds of papers and testified before Congress on environmental concerns. “Absolutely, it’s a good thing,” he said, referring to the IARC decision to call air pollution a carcinogen. “To be able to call something known—you have to have large enough samples. Now we have those, and informatics to help us analyze the data.”
Knowing what environmental factors cause cancer, so that it might be prevented, may be an impossible task. Most clinical data on toxin exposure—based on surveys of patients and, in some studies, controls—is necessarily correlative. The exposome, a new and essentially theoretical concept, would reflect changes in genetic material, proteins, and other molecules, such as lipids, after exposure to chemicals, ionizing radiation, particles, or other substances including germs. The practical problem is that it’s hard to know what a person’s been exposed to, from early embryology, as in the first trimester in utero, or later. Your expososome is affected by the air you’re breathing and beverage you’re sipping now.
“Cancer can take 20 to 30 years to develop,” Santella said. In her laboratory, researchers study how chemicals alter and bind to DNA. “If you measure exposure to a possible toxin at the time of diagnosis, that may not matter so much,” she said. “Information from surveys, about the past, is unreliable.” If you ask people with cancer about past exposures, their responses are limited by biases in recollection, and limits of memory. “It’s hard to get prospective data,” she said. “To carry out a study of sufficient statistical power, you need to recruit lots of people and wait and wait.”