Tuesday, September 22, 2015

Seeing red over breast cancer-wine reports: Why studies still get it wrong

A recent headline announcing the results of another study examining the role of alcohol and risk of breast cancer proclaimed that even a glass of wine a day “can significantly increase a woman’s cancer risk.” The connection between drinking and breast cancer has been a troublesome one, the thorn among the rosés, if you’ll pardon the metaphor. It’s not that I mean to make light of all this well-intentioned alarm bell ringing, but the way I see it research on wine and breast cancer just keeps repeating the same mistakes. When it comes to the question of wine and breast cancer, most studies still get it wrong.
That’s one reason why I can’t help but be a little bit cynical every October, when pink ribbons start sprouting everywhere, from lapels to football jerseys to cereal boxes. Increased awareness is laudable, but I wonder if these efforts do more to make us feel good than they do to actually make a difference for women with breast cancer. Case in point: the Komen Foundation, with its trademarked “for the cure” campaigns, reportedly spends less than a fifth of funds on research support to find a cure.
Advocates for breast cancer victims can’t be blamed for the misunderstandings about wine and cancer, and it’s hard to criticize the researchers - whose intentions are noble – but perspective is lacking. The Komen foundation wouldn’t even accept funds from a charitable wine auction I was involved in a few years ago when we approached the local chapter with the idea of making them a beneficiary. So we have this crazy situation where a breast cancer foundation won’t accept wine money, which may be a good thing because they don’t allocate much of it for research anyway.
Here’s the fundamental problem: When you do a study to identify which lifestyle factors are related to breast cancer (or anything else), of necessity you rely on self-reported questionnaires. With the question of drinking, there is a well-known under-reporting tendency so what subjects say they drink and what they actually drink are often not the same. This results in heavy drinking being mis-categorized as moderate drinking, with the implication that moderate drinking is riskier than it is.
When it comes to type of drinking, for example wine vs. beer or spirits, it becomes even murkier. What is well known about wine drinking and other disease conditions is that a pattern of a glass or two of wine with dinner on a daily basis is associated with lower risk than for nondrinkers. But very few people in a given North American population actually drink this way. What the studies do is ask about drinking preferences; if someone says they prefer wine with dinner, they may only do it occasionally, and sometimes have no alcoholic beverage, sometimes other types of drinks, and in varying amounts depending on circumstances. This makes it virtually impossible to know what the actual role of wine might be, even accounting for the under-reporting bias. No matter how large the study, or how long-term, it’s the same problem only amplified.
It’s clear that heavy drinking of any type is associated with higher risk, but what’s not as clear is how this extrapolates to low or moderate drinking. Generally, studies infer that a drink a day increases breast cancer risk by 10%, 2 drinks 20%, and on up. What this means (often misunderstood) is this is a percentage of baseline risk; so if you are at average lifetime risk of about 8 or 9 percent, a 10% increase is not 18 or 19%, but somewhere around 10 percent overall. (10% of 9% is 0.9%, so a 10% increase = 9.9%)
To put this in perspective it is helpful to consider that heart disease dwarfs breast cancer as a cause of death in women, and there is no question that moderate drinking (especially wine) associates with a lower risk. The same is true for diabetes, osteoporosis, and Alzheimer’s. For most women, stopping drinking to reduce breast cancer risk would result in an increased overall risk of premature death and disability.
But even that doesn’t tell the whole story. I am not convinced that moderate wine drinking contributes to even a low risk of breast cancer for the average woman. The definitive way to answer the question would be to study a population of women who only drink red wine, and in moderation as an integral part of their lifestyle. There are not very many of these studies because there are not many such populations anymore, but there is some good data. The best of these studies was done in Bordeaux a few years ago (1), and the conclusion was that wine drinkers had a substantially lower rate of breast cancer than nondrinkers.
Why should this be the case? For alcohol and breast cancer, it took years for a plausible cause-effect relationship to be put forth, and the best hypothesis is that it changes estrogen metabolism. Higher estrogen translates to greater risk of certain types of breast cancers. These types, known as estrogen-receptor positive (ER+), are often treated with drugs called aromatase inhibitors, which ramp down estrogen production. Red wine, as it turns out, is a natural aromatase inhibitor.
So forgive me if I don’t sport a pink ribbon, but I wish we had more efforts like the former Cleavage Creek Winery, which used profits to directly fund research at institutions like the Fred Hutchinson Cancer Research Center and Bastyr Naturopathic University in Seattle, bypassing “Big Pink.”  Founded by the late California winemaker Budge Brown after losing his wife of 48 years to breast cancer, Cleavage Creek donated more than $90,000 before the winery closed. So I say please do have a glass of red wine and toast to all those who are doing so much to save and improve women’s lives.

1.  Bessaoud F, Daurès JP. Patterns of alcohol (especially wine) consumption and breast cancer risk: a case-control study among a population in Southern France. Ann Epidemiol. 2008 Jun;18(6):467-75.