Free Beer, Wine After Alcohol Warning at Cancer Meeting

Is epidemiology alarmist?

Nick Mulcahy

December 07, 2018

SAN ANTONIO — On the first day here at the world's largest breast cancer meeting, Mary Beth Terry, PhD, an epidemiologist from Columbia University in New York City, gave a late afternoon lecture about alcohol use and the associated risk of breast cancer.

Before delving into multiple studies, Terry challenged the audience at the San Antonio Breast Cancer Symposium: "Hopefully at the end of my brief talk...nobody will be going to get a margarita on the Riverwalk."

"Really, there's no amount of alcohol that...would not be related to an increased risk of breast cancer," she said, reviewing data from a 2016 pooled cohort analysis of 1 million women that looked at risk by intake amount, including "light" drinking.

"Seldom do we get such a nice dose-response relationship in epidemiology," Terry observed.

For many years, epidemiologic studies have found a "consistent but modest association" of alcohol and breast cancer risk, she commented.

"Everybody should know this is a risk factor," announced Terry.

The next day at the symposium, there was an end-of-day poster session, which was billed as providing "refreshments."

At 5:00 pm, many of the 7500 attendees streamed into Hall One of the Henry B. Gonzales Convention Center to look at posters and seek refreshment.

The huge hall included five temporary bars, each operated by two bartenders and all doing a brisk, nonstop business. At 5:23 pm, there was a line 15 people deep at one bar. The alcohol was limited to wines and beers. Wines included pinot grigio (Woodbridge), chardonnay (Sand Dollar), moscato (Canyon Road), chenin blanc (Stellenbosch Hills), and tempranillo (Lost Maples). The beers were heavy on low-calorie varieties (Miller Lite, Bud Light, Coors Light, Michelob Ultra Light) but also included regular lagers (Corona, Shiner Bock).

As with the food, the wine and beer was free and abundant.

Based on a rough look-and-see, Medscape Medical News estimated that 40% of the poster session attendees drank alcohol. "I'm surprised it was that low," laughed Jennifer Ligibel, MD, a medical oncologist at Dana-Farber Cancer Institute in Boston, who attended the meeting and was asked for comment.

Epidemiologists vs Clinicians

Ligibel doesn't object to moderate alcohol consumption. "Alcohol is part of the social fabric," she told Medscape Medical News. "But I don't encourage anyone to start drinking either."

Ligibel did, however, object to some of Terry's talk, which she attended.

True, said Ligibel, there is "clear [epidemiological] evidence that alcohol intake is linked to malignancy" including breast cancer.

But epidemiological findings are not necessarily clinically relevant, she emphasized.

Ligibel explained that there are 50 years of studies linking alcohol and cancers. "With the huge amount of data we have, even small differences [in consumption] are statistically significant," she said.

Ligibel then cited an often-repeated statistic: for the average woman, there is a 12% lifetime risk of breast cancer. "If a woman consumes a drink a day, which is considered a low-level intake, that risk may become about 13% — which is statistically significant," Ligibel explained.

This math amounts to a 9% increase in the relative risk of breast cancer. "It's an increase," but it is a small one and not clinically relevant, she concluded.

Notably, Columbia's Terry reported a similar risk increase during her talk, saying for each daily drink over time, there is an associated 7% to 10% increased breast cancer relative risk.

Drinking more alcohol on a daily basis (> 2 drinks) is a different story — that's a clinically relevant risk, added Ligibel.

So, Terry and Ligibel essentially agreed on the basic facts of risk. But their interpretations of meaning differed greatly.

Ligibel summed up the conflict: The epidemiologist says, "Look I have a really significant P value" and interprets that as an important risk. On the other hand, the clinician sits with a woman in clinic and says the risk, with moderate drinking, is small and relative to other risks.

"The risk related to moderate drinking is about the same if you breastfeed or not. There are a lot things that have this magnitude [of risk] for breast cancer," said Ligibel.

The Dana-Farber oncologist also wanted to avoid being rigid in offering advice about lifestyle choices.

"We don't tell people to have all of their kids by age 20," she said, referring to the breast cancer-risk reduction associated with earlier childbearing, which is another modifiable risk factor.

"It's a balance," she summarized about discussing choices.

Medscape Medical News attempted to interview Terry after her meeting talk, but she had a previous commitment.

What to Tell Patients?

Sandhya Pruthi, MD, an internist, works in a high-risk breast cancer clinic at the Mayo Clinic in Rochester, Minnesota. "I counsel patients who ask me, 'What can I do to prevent breast cancer?' " she said.

Pruthi stresses risk modification, including regular exercise, weight management, and alcohol use awareness. "I recommend having one or less drinks a day," she said. "I don't advise they stop all alcohol."

Pruthi, who was in San Antonio and spoke with Medscape Medical News, also strategically uses a general health questionnaire developed for her breast clinic. Reviewing the form with a patient creates an occasion to discuss alcohol. With drinkers, Pruthi says, "I'm not here to tell you to stop drinking. I'm here to tell you [you] can modify your risk."

Pruthi also believes that discussing a possible mechanism for the alcohol-breast cancer risk is very important.

She tells patients that the metabolism of alcohol generates estrogen, which fuels the growth of many breast cancers. The scientific basis of the recommendation removes any specter of moralism associated with limiting alcohol, she suggested.

At the end of her talk, epidemiologist Terry took a question from a clinician whose patients already have breast cancer.

Standing at a microphone, a female physician asked: "What should we tell our patients when they come to clinic and ask if they can have a drink or two at the parties at the end of the year or if they have a wedding to go to? Should we advise against? Or is it OK to say yes?"

Terry didn't give a yes-or-no answer. Instead, she emphasized the importance of individual susceptibility and the need for metabolic tools to determine that human variance.

After hearing the lecture, Dana-Farber's Ligibel thought about the American Cancer Society guideline on breast cancer survivors, which states they can drink alcohol but not more than one drink a day. "There was nothing in [Terry's] talk that made me feel that this advice needs changing," said Ligibel.

Terry, Ligibel, and Pruthi have disclosed no relevant financial relationships.

2018 San Antonio Breast Cancer Symposium 2018: Abstract ES8-3. Presented December 4.

Follow Medscape senior journalist Nick Mulcahy on Twitter: @MulcahyNick

For more from Medscape Oncology, follow us on Twitter: @MedscapeOnc


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.