SAN DIEGO—The risk of hormone-receptor positive breast cancer is increased by as little as a drink or two a day, according to the largest study to look at the association between the two to date.
A second study shows that variations within two genes coding for the alcohol dehydrogenase enzyme (ADH) that is involved in metabolizing alcohol affect the risk of breast cancer in postmenopausal women.
The studies lend biologic plausibility to “the fairly strong epidemiologic link” between alcohol consumption and breast cancer, suggesting that drinking may affect risk through hormonal and genetic pathways, said Chi-Chen Hong, PhD, Assistant Professor of Oncology at the Roswell Park Cancer Institute.
Dr. Hong was not involved with the research, presented here at the American Association for Cancer Research Annual Meeting.
Also at the meeting, researchers reported that as little as 30 minutes of recreational exercise a week appeared to lower breast cancer risk. Those findings came from an analysis of data on some 6,000 women enrolled in the ethnically diverse Global Epidemiology Study.
Largest Study to Date
Experimental data suggest that alcohol intake increases the risk of breast cancer through its effect on estrogen, but only three major studies have looked at the association between alcohol use and breast cancer according to hormone-receptor status, noted Jasmine Q. Lew, a fourth-year medical student at the University of Chicago. Ms. Lew led the first study as a recipient of the Howard Hughes Medical Center-NIH Research Scholarship at the NCI's Division of Cancer Epidemiology and Genetics.
In that context, the researchers conducted the largest study to date to determine if the relation between alcohol and breast cancer differed by hormone-receptor status in postmenopausal women.
Data were reviewed for 184,418 postmenopausal women, with a mean baseline age of 62, who were enrolled in the prospective NIH-AARP Diet and Health Study. At the study's outset, a food frequency questionnaire was given to determine alcohol and other nutrient intakes, and data on demographics, lifestyle, and medical history were collected.
The women were followed for an average of seven years, during which 5,461 cases of invasive breast cancer were identified. Information on receptor status was available in 2,391 cases: 1,641 tumors were estrogen-receptor and progesterone-receptor positive (ER+/PR+); 366 tumors were negative for both receptors (ER-/PR-); 336 were positive for ER and negative for PR (ER+/PR-); and 48 were negative for ER and positive for PR (ER-/PR+).
Thirty percent of the women reported that they did not drink alcohol, and in those who did, consumption averaged 8.2 g, or less than one drink, per day.
Alcohol Raises Risk
Results showed that the greater the self-reported consumption of alcohol, the greater the risk for any type of breast cancer.
Compared with women who abstained from alcohol, women who reported consuming less than 5 g a day had a nonsignificant 4% increase in breast cancer risk. Women who consumed 10 to 20 g per day had a significant 14% increase in risk, and women who consumed 45 g or more a day had a significant 38% increase in breast cancer risk.
A similar pattern was observed for ER+/PR+ tumor types, which Ms. Lew said account for 70% of breast cancers. Compared with women who abstained from alcohol, women who reported consuming one to two drinks a day were 32% more likely to develop ER+/PR+ invasive breast cancer. Having three or more drinks daily raised the risk of ER+/PR+ tumors by 51%.
She added that drinking alcohol also appeared to raise the risk of ER+/PR- and ER-/PR- tumors, but that there were too few women in these categories to make definitive conclusions.
The relationship between alcohol and breast cancer was not significantly affected by body-mass index, use of hormone-replacement therapy, family history of breast cancer, or folate intake.
The findings support the hypothesis that alcohol interferes with estrogen metabolism, leading to changes in cell metabolism and growth, Ms. Lew said.
Dr. Hong agreed. “Alcohol influences estrogen hormone, so there are increased levels of estrogen and increased production of estrogen. It can also decrease metabolism of androgens. And alcohol can increase the transcription activity of ER-alpha. Normal breast tissue expresses mostly ER-beta receptors; as it progresses to breast cancer, more ER-alpha receptors are expressed.”
She noted that a review article in Nature Clinical Practice Oncology last year (Chen W, Colditz G: 2007;4:415–423) showed that alcohol consumption, parity (having never given birth vs having three or more children), and age over 30 at first birth are all associated with an increased risk of ER-positive breast cancer.
“Very few risk factors have been associated with ER-negative tumors, which are more difficult to treat,” Dr. Hong said.
To determine whether genes coding for ADH may help explain the apparent link between alcohol and breast cancer, Lombardi Comprehensive Cancer Center researchers analyzed DNA samples from 991 postmenopausal women with breast cancer and 1,698 controls matched by age, race, and county of residence.
Variations were found within the DNA sequences rs1042026 in the gene ADH1B and rs1614972 in the gene ADH1C that affected the risk of breast cancer.
Women who had a variant form of ADH1B and drank alcohol were 94% more likely to have breast cancer as those who didn't have the variant and abstained.
“The higher their alcohol consumption, the higher their risk,” said Catalin V. Marian, MD, PhD, a research instructor in the Division of Genetics and Epidemiology in the Oncology Department at Georgetown University.
The variant form of ADH1C appeared to protect against breast cancer, he said, “but protection was lost with increased alcohol consumption.”
Commenting on the study, Dr. Hong said that if confirmed, the findings may help to pinpoint women who may be genetically susceptible to alcohol's damaging effects—“But they don't really help much on a public health level; unless you go in and test everybody, you're still dealing with trying to control exposure as opposed to telling people who can drink.”
Still, genotypes are useful for developing targeted therapies and picking what type of therapy people should get, she said.
AACR Abstracts 4168, 5814, 3083
Talking to Patients
So what should oncologists tell patients, particularly those who bring up the fact that studies have suggested that a few glasses of wine may offer cardioprotection?
“Clinicians should be aware that alcohol increases the relative risk of breast cancer. But it is too early to tell at this point if it is a definitive risk factor,” Ms. Lew said.
Elizabeth A. Platz, ScD, MPH, Associate Professor and Director of the Training Program in Cancer Epidemiology, Prevention, and Control at Johns Hopkins Bloomberg School of Public Health and Stanley Kimmel Cancer Center, said the increased risk of breast cancer associated needs to be balanced against any protective effect against heart disease.
“If the patient has breast cancer in the family, I would imagine you would have to think about cardiovascular risks. There are other ways to modify cardiovascular risk besides having a drink a day. But most risk factors for breast cancer, such as genetics or family history, are non-modifiable,” Dr. Platz said.
For the exercise study, researchers analyzed data on 1,468 breast cancer cases and 4,865 non-cancer controls in the Global Epidemiology Study (GES), which assesses disease risk factors in people recruited from the United States, Tunisia, and Poland.
The GES is linked to a biobank at BioServe Biotechnologies in Beltsville, MD, that houses “more than 600,000 human specimens from 160,000 individuals, allowing us to do all sorts of studies without ever picking up a pipette,” said Teresa A. Lehman, PhD, the company's Chief Technical Officer.
Patients with newly diagnosed breast cancer filled out an extensive questionnaire that asked about their diet, smoking, and exercise habits.
Results showed that women who engaged in recreational exercise 30 to 150 minutes a week were 50% less likely to have breast cancer than women who exercised less than a half-hour per week. African-American women benefited the most: They were 69% less likely to have breast cancer if they exercised 30 to 150 minutes a week than if they exercised less. But exercise had a protective effect in Caucasian-American, Hispanic-American, Tunisian-Arab, and Polish-Caucasian women as well, Dr. Lehman said.
Exercising more than 150 minutes a week did not confer additional benefit, and subgroup analyses showed that the findings held true regardless of menopausal, lymph node, or hormone-receptor status. The odds ratios were adjusted for age, pack-years smoked, and body-mass index.
Dr. Hong said “the evidence for an exercise-breast cancer link is strong. It appears not only to protect against the development of breast cancer, but also appears to prolong the life of women who already have breast cancer.”
A recent study showed that exercise appears to be more protective against ER+/PR- tumors, which are associated with a clinically more aggressive tumor phenotype, than against ER+/PR+ tumors, she said.
Marji McCullough, ScD, RD, Strategic Director of Nutritional Epidemiology and Surveillance Research at the American Cancer Society, said that the findings are consistent with the group's recommendation to engage in regular physical activity as a means of lowering breast cancer risk.
But she said that patients should be advised to work out at least 30 minutes a day, five times a week, to lower their risk.
“Studies have shown that being consistent over a lifetime is particularly beneficial. But it's never too late to start.”