Study: Does eating meat affect breast cancer risk?

IN-DEPTH REVIEW OF RESEARCH

Study background:

Previous studies on meat in diet and breast cancer risk came to conflicting conclusions. Some studies found a slight increase in breast cancer risk with eating red meat, while others did not. Similarly, several studies found no association between eating poultry and breast cancer risk, while others observed a slight decrease. This study seeks to more clearly determine the impact of eating different types of meat or using different cooking methods on breast cancer risk among women with a family history of breast cancer.

Researchers of this study wanted to know:

how eating meat affects breast cancer risk in a large group of women with a family history of breast cancer.

Populations looked at in this study:

Participants included 42,012 women between ages 35 to 74 who had never been diagnosed with breast cancer, but who were sisters or half-sisters of women with breast cancer. Women were recruited from the United States or Puerto Rico from 2003 to 2009. Potential participants were ineligible if they did not complete a food frequency questionnaire, had extreme body weight (BMI less than 15 or greater than 50), extreme calorie consumption (less than 600 calories per day or greater than 3,500 calories per day), or were pregnant. Data from participant was included only if they had at least 1 year of follow-up in the study.

Study design:

This was a prospective cohort study. Eligible participants completed a 110-item food frequency questionnaire about their diets, including frequency and average serving sizes of food and drinks, and how food was prepared.

• Red meat consumption was based on beef, veal, lamb, pork or game meats.
• Poultry consumption was based on chicken, turkey, Cornish hen, duck, goose, quail, and pheasant or game birds.
• Meat carcinogens were estimated using the Computerized Heterocyclic Amine Resource for Research in Epidemiology of Disease (CHARRED) version 1.7 based on self-reported cooking methods. Dietary iron was estimated from self-reporting of participants’ cooking methods and doneness of meat using the National Cancer Institute Heme iron database.

Participants self-reported breast cancer diagnoses and ER-status at yearly follow-up interviews. Where possible, breast cancer status was confirmed by medical records (82% of participants agreed to the release of their medical records, and 99% of diagnoses were confirmed for estrogen receptor-positive breast cancer and 83% for estrogen receptor-negative breast cancer).

Participants were followed for at least one year, until the end of the study, breast cancer diagnosis or death. All 42,012 participants were evaluated for a total of 275,922 person-years (average 7.9 years of follow up).

At the end of the study, statistical analyses were performed to identify any differences among the women who developed breast cancer regarding the amount and type of meat they consumed, their ER-status, and other dietary and demographic factors. Women who consumed a specific type of meat were ranked into 5 groups based on how much of that meat that they ate (top 20%, next 20% etc.) to compare whether breast cancer cases increase as consumption of that meat increased.

It is important to note that other factors are also correlated with high meat consumption or high red meat consumption (e.g., household income and other eating and health habits). The authors tried to control for confounding factors—variables that they couldn’t account for—by evaluating those that they suspected from prior studies. However, a concern remains that unknown confounding factors may account for the results observed. This is a major limitation of the study.

Study findings:

• 1,536 cases of breast cancer occurred among the 42,012 women with a family history of breast cancer who participated in the Sisters Study. 1,533 women with breast cancer were clearly pre or postmenopausal.
  • 1,232 cases of breast cancer occurred in premenopausal women.
  • 301 cases of breast cancer occurred among postmenopausal women.
     
• Women who ate more red meat had a modestly increased risk of breast cancer.
  • Among women who ate red meat, those who ate greater amounts of red meat had an increased risk of breast cancer.
  • Women who ate the greatest amount of red meat (top 20%) had a 23% increased risk of breast cancer compared to those who ate the least red meat (lowest 20%).
     
• Women who ate more poultry than other participants had a modestly decreased risk of breast cancer.
  • Among women who ate poultry, women who ate greater amounts of poultry had a decreased risk of breast cancer.
  • Women who ate the most poultry (top 20%) had a 15% increased risk of breast cancer compared to those who ate the least poultry (lowest 20%).
     
• Amount of seafood consumption was not associated with an increase or decrease of breast cancer risk among participants in this study.
  • Women who ate the most seafood (top 20%) had a 2% increased risk of breast cancer which was not statistically significant compared to those who ate the least seafood (lowest 20%).
     
• Data modeling suggested that substituting poultry for red meat in a woman's diet may decrease the risk of breast cancer.
  • Data models suggest that if total meat consumption is the same but poultry is substituted for red meat, the predicted risk of breast cancer may be 28% less. The model used simultaneously reduces red meat consumption and increases poultry consumption.
     
• Breast cancer risk was not associated with particular cooking practices, estimated mutagens from cooking practices (for example charring meat) or iron from red meat.
   
• The impact of meat consumption was seen predominantly among postmenopausal women with breast cancer.
  • Postmenopausal women with the greatest red meat consumption had 41%  increased risk of breast cancer versus those with the least red meat consumption.
  • Postmenopausal women with the greatest poultry consumption had 34%  decreased risk of breast cancer versus those with the least red meat consumption.
  • The small number of premenopausal women with breast cancer may limit the ability to detect an impact in this group.

Limitations:

This study has a number of significant limitations.

• The major limitation is that this is a prospective observational study—although the data provides a correlation, it does not demonstrate that the correlation caused the effect. Multiple factors (socioeconomic differences, weight, alcohol consumption) differ between individuals who consume the greatest amount of red meat and those who consume the least. The authors statistically controlled for factors they suspected might affect the results. However, it is possible that other confounding factors may account for the differences seen.
   
• This type of study does not inherently have controls and it is not a randomized treatment. Individuals self-selected their diet and reported it to researchers. A randomized controlled trial in which participants were provided meals with defined content would be ideal to accurately assess impact and allow researchers to determine if this difference in diet directly causes any differences in breast cancer risk observed.
   
• This study only evaluated women with a family history of breast cancer. This data may or may not be applicable to women in the general population who do not have a family history of breast cancer.
   
• The mutation status of participants was not evaluated. Given the required family history of breast cancer in a sister or half-sister, these participants are more likely to be mutation carriers than women in the general population. However, the frequency of mutation carriers among the women with breast cancer in this study is unknown, which limits the conclusions that can be made.
   
• The diet of participants was evaluated with a single food consumption questionnaire at the beginning of the study. While this was a comprehensive survey, changes in participants’ diet patterns over the course of the study were not evaluated.
   
• Diet and breast cancer diagnoses were self-reported by participants, and where possible, was verified to match their medical records. However, diet was not externally verified or controlled.
   
• Data modeling indicates that substituting poultry for red meat consumption may reduce breast cancer risk even more than the reduction observed among women with greater poultry consumption. While promising, this observation needs to be tested directly in a prospective study to determine if it is accurate.
   
• Study authors note that they had limited ability to evaluate participants’ cooking methods. They used this information to predict mutagen levels among participants based on those cooking practices. The lack of association between suspected mutagens (charred meat, for example) may reflect the limitation of their survey approach.
   
• Relatively few participants (n=301) were premenopausal women who were diagnosed with breast cancer, compared to postmenopausal women who were diagnosed with breast cancer (n=1232). No general association was made regarding menopausal status and meat consumption. However, among postmenopausal women, red meat consumption was associated with increased breast cancer risk and poultry consumption was associated with decreased breast cancer risk. No significant association with meat consumption was observed among premenopausal women. Fewer premenopausal breast cancer cases may limit the ability to observe a small but real impact of meat consumption. Alternatively, diet may not affect premenopausal and postmenopausal women in the same way. Additional data on premenopausal women is needed to clarify this issue

Conclusions:

There may be an association between meat consumption and breast cancer, but it is also possible that other related factors are at play, and this does not reflect a cause-and-effect relationship.

Share your thoughts on this XRAYS article by taking our brief survey.

Posted 9/10/19