First published April 2004
Northern California's Marin County has long been lampooned as trendy and hedonistic, a place where personal comfort rules. While these stereotypes are easy to dismiss, Marin dwellers do, in fact, enjoy greater wealth, a better climate, grander scenery and more lifestyle choices than just about anyone else on the planet.
Yet there also is something else, unbefitting such a paradise, that stamps Marin County as special: The number of new breast cancer cases is 40 percent higher than the national average. Spurred by activists, researchers -- including many from UCSF -- are working like never before to find out why.
A study published two years ago, led by Christina Clarke of the Northern California Cancer Center (NCCC), found that the yearly tally of new breast cancers in Marin rose six times faster throughout the 1990s than it did in the rest of the state. Indeed, Marin may be at the forefront of a deadly trend; the National Cancer Institute has announced that breast cancer incidence rates are rising nationwide, in contrast to what government statisticians previously thought.
The NCCC report and the efforts of Marin activists -- including a sharp-edged TV ad in which shoppers reach for organic produce over the dead body of a cancer victim -- have raised awareness to a fever pitch and gotten new players involved. Politicians are asking questions. More government dollars are being squeezed out for research to help solve this puzzle, bringing activists and scientists together in partnerships that sometimes begin awkwardly.
For public health officials, awareness-raising tactics -- like those TV ads -- can become scare tactics, and when it comes to breast cancer and Marin, it doesn't take much to raise fears. Even without seeing the lawn placards that suggest a cancer-causing threat from radioactive waste dumped 30 miles offshore, breast cancer already looms large in the minds of many; so much so, in fact, that it can crowd out the reality that for now, heart disease and lung cancer are bigger killers of women.
The same lack of awareness obscures known breast cancer risks, which include getting older, early age of first menstruation, late menopause, delayed or no childbearing, not breast feeding, early exposure to radiation and a family history of the disease. Nor are modifiable behaviors that have been identified more recently as breast cancer risks -- drinking, not exercising, carrying excess body weight after menopause, and taking hormone replacement therapy -- as well known to women as physicians would like.
Still, even if women knew all scientists know, or think they know, about breast cancer risk, and did all they could to lower their exposure to known risks, they still would have cause for concern. Established risk factors, modifiable and nonmodifiable, account for only about half of breast cancer cases.
"Everybody throws their hands up at the other 50 percent," says Jeanne Rizzo, executive director of the San Francisco-based Breast Cancer Fund. "That's not OK. There's reason to be afraid. The numbers are there. The risk is great."
The Breast Cancer Fund, the Marin Cancer Project (the TV-ad sponsor) and other grassroots activist groups in Marin suspect environmental causes for breast cancer. For their part, epidemiologists do not rule out the possibility that the environment will one day be linked to breast cancers. Indeed, considering how rapidly breast cancer rates climb in Asian and Hispanic families that immigrate to the US, nonhereditary factors must clearly play a role in America's high breast cancer incidence.
However, to epidemiologists, the environment encompasses not just pollution, but also what we eat, our social behaviors, and whatever else our lifestyles expose us to, and it has taken several studies, many years and lots of research funding just to nail down common and potent risks -- drinking more than a glass of wine daily, for instance. Exposures that occur less often and carry less risk are harder to detect and often difficult to measure accurately.
In 1995, a group of breast cancer survivors formed Marin Breast Cancer Watch (MBCW) and began searching for scientists who would help them find out why breast cancer rates were so high in the county. Several researchers turned them down flat. In addition to the difficulties presented by looking for new breast cancer causes, some scientists assumed that the high Marin rates would prove unexceptional once additional demographic information besides age was factored in. For example, Marin is overwhelmingly white, and white women are known to have breast cancer rates up to 75 percent higher than Hispanic and Asian women and 25 percent higher than black women. Furthermore, women in Marin tend to delay or forgo child bearing and to have fewer children, both strong breast cancer risks.
In early 1997, Mary Gould, a founding member of MBCW, called UCSF's Margaret Wrensch, who had experience studying difficult-to-measure, non-occupational environmental exposures and their potential relationship to cancer, including an NIH-funded study of power transmission lines and brain tumors (no relationship discovered). "She asked if I would help," Wrensch recalls. "I said I would be happy to come and talk to the group, but I thought one discussion would be the end of it." The meeting spawned a five-year collaboration.
The breast cancer survivors met with Wrensch weekly to review the research literature, explore study designs and discuss community concerns regarding suspected environmental risk factors. Wrensch's own review led her to believe that the high incidence of breast cancer in Marin might not be easily explained by demographics. Many counties are overwhelmingly white, without Marin's high breast cancer rates. Moreover, the number of breast cancer deaths in Marin has remained high, unlike other places where breast cancer mortality, if not incidence, is falling. This argued against the most commonly heard explanation for the county's high breast cancer incidence -- that the increased use of screening mammography had led to earlier detection of nonlethal tumors.
Socioeconomic Status, Adolescence and Breast Cancer Risk
Wrensch and two epidemiologists who soon became research collaborators -- the NCCC's Clarke and Christine Erdmann, then at the Lawrence Berkeley National Laboratory, also were familiar with provocative but little publicized research on socioeconomic status (SES) and breast cancer rates.
Surprisingly, the elevated breast cancer incidence measured in Marin at a countywide level is similar to rates measured in women at the higher rungs of the socioeconomic ladder, regardless of where they live, a conclusion first reported by Clarke's NCCC colleagues Angela Prehn and Dee West in 1998, after they used 1990 census data to compare Marin cancer rates with rates in similarly wealthy census tracts. This connection was confirmed in 2002 by researchers who measured comparable breast cancer rates in more than 100,000 California schoolteachers.
The association between breast cancer and high SES is due in part to the tendency of high-SES women to bear fewer children and to have them later. These and other "reproductive risk factors," including early menarche and late menopause, are believed to raise breast cancer risk by increasing a woman's lifetime exposure to her own estrogen.
Although the start of cancerous changes in a breast cell is thought to begin with damage to genes -- due to random errors in copying or repairing DNA, or environmental exposures to radiation, for instance -- some researchers believe estrogen metabolites also can help start the process. Furthermore, once genes with mutated DNA have taken the helm and sent cells off course, scientists agree that estrogen can act hormonally to promote the subsequent growth and development of tumors.
The connection between SES and reproductive risk factors is no surprise to scientists who say demographics explains the disproportionate number of Marin breast cancers. More provocative are studies suggesting that high socioeconomic status remains an independent risk factor even after this "covariance" with reproductive and other established risk factors is taken into account. If true, the link between SES and breast cancer, while still based on demographics, becomes a complete mystery.
Even if SES is an independent risk factor -- and not all scientists are convinced that it is -- and even if the risk attributed to high socioeconomic status in these other studies is applied to Marin and found to account for the balance of excess breast cancers, it still will not explain them. Having money and a good education does not cause breast cancer; socioeconomic status is merely a proxy for something else, something about lifestyle choices, perhaps, that influences whether one is exposed to undiscovered breast cancer risks.
Wrensch and her new, nonscientist collaborators also were interested in another unexplained mystery, the adolescent "window of vulnerability" to cancer threats. Although breast cancer usually takes decades to evolve from initial DNA mutation to detectable tumor, pubescent girls appear to be especially susceptible to events that start the cellular changes that can lead to breast cancer many years later. Epidemiologists have learned that Japanese women exposed during adolescence to radiation from atomic bomb blasts, as well as women who received radiation treatments for childhood cancer, are particularly vulnerable to breast cancer. So are female mice that are irradiated as they approach sexual maturity. And while it has been well known for decades that breast cancer rates for Asians and other immigrants go up when they immigrate to the US, more recent studies indicate that the increase is more pronounced for girls who arrive before they undergo puberty.
In 1999, with funds secured from the newly established California Breast Cancer Research Program Community Research Collaboration, Wrensch and the MBCW activists, led by physician Georgianna Farren, initiated the Adolescent Risk Factors Study (ARFS). Using cancer registry data, the researchers attempted to contact and interview in person all Marin women diagnosed with breast cancer between 1997 and 1999, and for comparison, a randomly selected group of similar-aged women without cancer. Study participants answered questions about exposures to risks during adolescence and adulthood, growth histories, physical and social environment, and personal habits, including smoking and drinking.
The study revealed no association between how long women lived in Marin as adults or children and the incidence of breast cancer. "I don't think the elevated breast cancer incidence is a Marin-specific issue," Wrensch now believes. "If there was something that had to do with Marin country per se -- the water or the air or the land -- you would expect the women who had breast cancer to have lived there longer than the women without breast cancer. They didn't; they lived there just about the same amount of time."
The study did reveal, however, that ARFS participants with breast cancer were more likely than never-diagnosed women to have grown up in wealthier families, a finding that Wrensch says deserves follow-up in light of the adolescent window of vulnerability and the unexplained association of breast cancer and high SES.
But the study finding most often reported in the press was that the women with cancer drank more on average, as adults, than women who did not have cancer. More recently, a survey by the Marin County Department of Health and Human Services found that 77 percent of Marin women drink. By comparison 53 percent of women drink statewide, and 46 percent nationwide. More than 12 percent of Marin women drink daily, compared with 5 percent nationwide.
Alcohol has become widely regarded as a risk factor for breast cancer in recent years, although the magnitude of the risk measured in major studies has varied somewhat, and drinking has not yet been routinely factored into risk analyses. Alcohol also appears to synergistically increase a woman's chances of getting cancer when combined with other factors now emerging as breast cancer risks, such as a diet low in folate, a vitamin found in many green vegetables and in fortified grains. In two major, independent studies, high folate consumption appeared to lessen the increased risk due to drinking.