Critique 202: Importance of folate intake for reducing breast cancer risk from alcohol consumption, especially for women with a positive family history of breast cancer – 17 July 2017

Kim HJ, Jung S, Eliassen AH, Chen WY, Willett WC, Cho E.   Alcohol consumption and breast cancer risk by family history of breast cancer and folate intake in younger women.  Am J Epidemiol 2017: pre-publication.

Authors’ Abstract

To evaluate the association between alcohol consumption and breast cancer risk in younger women, and by family history of breast cancer and folate intake, we prospectively followed 93,835 U.S. women aged 27-44 years with alcohol consumption data in 1991 in the Nurses’ Health Study II. Alcohol and folate intake was measured by food frequency questionnaire repeated every 4 years. We documented 2,866 incident invasive breast cancer cases between 1991 and 2011.

Alcohol consumption was not associated with breast cancer risk overall (multivariate hazard ratio = 1.07, 95% confidence interval: 0.94, 1.22 for >/= 10 g/d intake vs. nondrinkers). When the association was stratified by family history and folate intake, a positive association between alcohol consumption and breast cancer was found among those with a family history and folate intake of < 400 microg/d (multivariate hazard ratio = 1.82; 95% confidence interval: 1.06, 3.12; P-trend = 0.08). Alcohol intake was not associated with breast cancer in other categories of family history and folate intake (P-interaction = 0.55).

In conclusion, in this population of younger women, higher alcohol consumption was associated with increased risk of breast cancer among those with both family history of breast cancer and lower folate intake.

Forum Comments

Studies of breast cancer in women are important, as breast cancer is a very common type of cancer and is especially feared by most women. Unlike lung cancer, which has a known strong relation with smoking, no single environmental factor is known to have such an effect for breast cancer.

The strongest factor associated with an increase in the risk of a woman developing breast cancer appears to be a positive family history of such a diagnosis in a sibling or mother. Among environmental factors, almost all studies have shown that alcohol consumption relates to increased risk; a slight increase is often seen even among women who report only light drinking, e.g., an average of one drink or less per day.  The present study evaluates how a positive family history of breast cancer in a first-degree relative and folate intake may modify the association between alcohol intake and breast cancer; the analysis is based on data from a large number of young women, aged 27-44 years at baseline, over a follow-up period of 20 years.

The results of this study are especially important as the analyses are based on data from young women in the very large Nurses’ Health Study II, with well-collected and validated exposure and outcome data from more than 90,000 subjects over a long period of time. In this cohort, 2, 866 incident invasive cancers were detected.  While the overall risk of breast cancer did not increase significantly with alcohol consumption, when women were classified as to whether or not there was a family history of breast cancer (present in 6 to 7% of women), a slight increase in risk of cancer was seen among those with family history, but not among the 93-94% of the women without such a history.  Similarly, when subjects were divided into groups according to their folate intake (either < 440 ug/day versus ≥400 ug/day), the increase in breast cancer risk with increasing alcohol intake was seen only among those women with low levels of folate.

To summarize, for women without a family history of breast cancer, there was no significant increase in breast cancer risk from alcohol, although for those with low folate levels, there was a tendency (statistically insignificant higher RR for cancer risk) among women reporting an average of 5 or more grams of alcohol per day than for non-drinkers or consumers of less. With a positive family history of breast cancer plus low folate, however, the RRs for cancer were higher for all categories of alcohol consumption, but the only alcohol category with a statistically significant effect on risk was the group with both low folate and the highest level of alcohol intake (≥ 10 grams/day), where the RR=1.82 (CI 1.06, 3.12).

The new study has many strong features: it was a prospective study focused on younger women, had a large number of subjects in the cohort, used a validated dietary instrument for assessing alcohol and folate intake, had validation (in 98% of cases) of reported invasive breast cancer, and included adjustments in the analysis for most of the other known factors that have been shown to relate to cancer risk. Unfortunately, the investigators did not present results on the pattern of drinking, so binge drinking and consuming alcohol without food (which could lead to higher levels of alcohol in the blood after drinking and could possibly affect cancer risk), were not included in the analyses.  Also, risk estimates according to the specific beverage consumed were not reported.

Factors modifying the relation between alcohol consumption and breast cancer risk: A number of large epidemiologic studies have shown that the risk of breast cancer associated with reported light to moderate alcohol consumption is increased for women who binge drink (Chen et al), are users of hormone replacement therapy (Neilsen & Groenbaek), and/or have low levels of folate intake (e.g., Zhang et al 1999, Rohan et al, Sellers et al, Baglietto et al, Stolzenberg-Solomon et al).  Further, Klatsky et al, in their very large Kaiser-Permanente studies, report that women who are found (from an extensive review of their entire medical records indicating greater alcohol consumption at some time) to be likely “under-reporters” of their alcohol intake show a higher risk of breast cancer than those without such evidence in their records.  Specifically for breast cancer incidence, those authors report that for likely under-reporters of alcohol intake, HR for breast cancer was 1.48 (1.23–1.79) at <1 drink per day and 1.39 (1.12–1.74) at 1–2 drinks per day; the corresponding risks for subjects classified as unlikely under-reporters were HRs of 1.05 (0.89–1.24) and 1.06 (0.81–1.37), respectively, for these two levels of alcohol intake.  Thus, accounting for under-reporting of alcohol intake, not included in the present study, may also modify the relation of alcohol to breast cancer.

Specific comments by Forum members: Forum member Finkel commented: “I applaud this paper, mainly for its important clarification of a major confusion, but also for its large number of subjects, the long period of observation, and its simple and direct approach.  We have long been confused by the meaning of murky and conflicted results of studies trying to elucidate the relationship of alcohol to breast cancer.  I have speculated that breast cancer might be best viewed as a group of several diseases, rather than as one.

“Here is the start of two facets of breast-cancer risk being dissected out clearly, and competently. Other facets need this kind of attention. For example, we need to know more about the details of the drinking habits of the subjects and the histopathology of the cancers.  And once again, we present balanced perspective demands that the burdens of breast cancer must be weighed against the cardiovascular and other benefits of moderate drinking, eventually, through studies such as this, with an accurately targeted aim.”

Reviewer Van Velden commented: “This is a well done paper with valuable conclusions. It emphasises that alcohol consumption must not be seen in isolation, but part of a healthy lifestyle of moderation, weight management, Mediterranean-type diet high in folate and alkaline in nature, and an active lifestyle.  Binge drinking can never be considered appropriate.”

Forum member Skovenborg noted: “I agree that the study is well done and of public interest regarding younger women’s fear of the alcohol-breast cancer association. I would like to mention that a similar result was found for postmenopausal women with a family history (Sellers et al).  In a pooled analysis of cohort studies, however, the association between alcohol and breast cancer was not significantly modified by family history (Smith-Warner et al).”

Earlier research on potential mechanisms for a relation of folate to effects of alcohol: Forum member Stockley has provided some additional data on alcohol and folate.  “It has been previously suggested that an adequate consumption of folate may reduce the increased risk of breast cancer associated with alcohol consumption (Zhang et al 1999, Rohan et al, Sellers et al, Baglietto et al, Stolzenberg-Solomon et al, Tjonneland et al).  For example, while alcohol interferes with DNA synthesis and repair, folate is involved in DNA synthesis, repair and methylation.  In animal models, folate supplementation reduces DNA strand breaks in the p53 gene (Kim et al); the P53 protein regulates the cell cycle to prevent genome mutation, and hence functions to suppress tumors.  It can activate DNA repair proteins when it recognizes damaged DNA, hold the cell cycle at the G1/S regulation point on DNA damage recognition to prevent uncontrolled cell division and can initiate apoptosis, the programmed cell death, if the DNA damage proves to be irreparable.  Cancer occurs when the rate of proliferation of mutated cells greatly exceeds the rate of apoptosis.  In breast cancer, the gene has been observed to be mutated in 15 to 50% of tumors (Olivier & Hainaut).

“The concurrent consumption of alcohol and folate (of at least 300 ug/day) has been observed to reduce the relative risk of alcohol-induced breast cancer to 1.05 for women consuming greater than 15 g alcohol/day (or one and a half standard drinks), but was only 0.55 for women consuming greater than 600 ug/day of folate. Indeed, the concurrent consumption of folate-containing vitamin supplements reduces the relative risk to 0.74 for women consuming greater than 15 g alcohol/day compared to those not using vitamins (Zhang et al 1999).  The interaction between alcohol and folate has been observed to be primarily limited to estrogen receptor negative (ER-) breast cancer tumors (Zhu & Williams, Sellers et al, Zhang et al 2005), which is consistent with an interaction of alcohol and folate on breast tissue tumors being mainly through the primary metabolite of alcohol, acetaldehyde, which is directly carcinogenic as well as indirectly carcinogenic via folate depletion, independent of circulating estrogens and estrogen receptor-mediated events.”

Reviewer Stockley added: “The suggestion of a further increased risk of breast cancer by post-menopausal woman who use estrogen-replacement therapy (ERT) and who are also light to moderate consumers of alcohol remains controversial (Suzuki et al, Nielsen & Gronbaek), in particular as ERT decreases the risk of other diseases such as cardiovascular disease.”

Dealing with differing results of studies on folate and breast cancer risk: Reviewer Skovenborg wrote: “I agree with the review from Stockley on folate, alcohol and breast cancer risk.  However, it is disturbing that one of the authors of the present study, Jung, is the leading author of a study from 2016 that found no effect modification of folate intake on breast cancer risk in women with alcohol consumption (Jung et al).  In Denmark the Jung et al study is quoted all the time as proof that a high folate intake offers no protection against breast cancer in women with a moderate intake of alcohol.”

Forum member Waterhouse stated: “I think that the current authors should have done a better job of comparing their results with prior data that was contradictory. In looking at the papers that Skovenborg noted, it seems that there are a couple of explanations that might be valid.  For example, the Jung et al paper was a meta-analysis with very many different methods of assessing folate intake, including datasets where the authors had to create a model for folate based on other factors.  In this case, we might be able to say that their data on folate was weak compared to the present study which was carefully designed to properly assess folate intake.

“In the Sellers et al study, the investigators observed a folate effect in FH- subjects, but not in the FH+ population. In other words, genetics in this case overwhelmed the folate effect.   In the present study, it is not clear if they did the same comparisons, since the tables do not compare differential folate intake levels for specific alcohol intake in family history groupings.  Instead they segregate the populations by folate intake and then look at alcohol intake levels.  I don’t think the HR numbers can be compared between folate intake groupings.  On the other hand, the current data seem to show that high folate intake can reduce the risk of alcohol consumption for those with a family history of breast cancer.”

Clinical implications of the findings of the present study: Forum members considered this to be a very well-done analysis, based on prospectively collected data according to a strict protocol.  There were repeated assessments over time of alcohol consumption and folate intake.  Members conclude that the results from the present study should greatly relieve anxiety about breast cancer for women without a positive family history of breast cancer who choose to consume light-to-moderate amounts of alcohol.  For women with a first-degree relative with breast cancer, the data indicate that if they maintain a high level of folate intake, it may attenuate an increase in risk associated with alcohol intake.

References from Forum review

Baglietto L, English DR, Gertig DM, Hopper JL, Giles GG. Does dietary folate intake modify effect of alcohol consumption on breast cancer risk? Prospective cohort study. Br Med J 2005;331:807.

Chen WY, Rosner B, Hankinson SE, Colditz GA, Willett WC. Moderate alcohol consumption during adult life, drinking patterns, and breast cancer risk.  JAMA 2011;306:1884-1890/

Jung S, Wang M, Anderson K, Baglietto L, Bergkvist L, Bernstein L, et al. Alcohol consumption and breast cancer risk by estrogen receptor status: in a pooled analysis of 20 studies.  Int J Epidemiol 2016;45:916–928.  doi: 10.1093/ije/dyv156.

Kim YI, Shirwadkar S, Choi SW, Puchyr M, Wang Y, Mason JB. Effects of dietary folate on DNA strand breaks within mutation-prone exons of the p53 gene in rat colon.  Gastroenterology 2000;119:151-161.

Klatsky AL, Udaltsova N, Li Y, Baer D, Tran HN, Friedman GD. Moderate alcohol intake and cancer: the role of underreporting.  Cancer Causes Control 2014;25:693-699. DOI 10.1007/s10552-014-0372-8 2014.

Nielsen NR, Grønbaek M. nteractions between intakes of alcohol and postmenopausal hormones on risk of breast cancer. Int J Cancer 2008;122:1109-1113.

Olivier M, Hainaut P. TP53 mutation patterns in breast cancers: searching for clues of environmental carcinogenesis.  Semin Cancer Biol 2001;11:353-360.

Rohan TE, Jain MG, Howe GR, Miller AB. Dietary folate consumption and breast cancer risk. J Natl Cancer Inst 2000;92:266-269.

Sellers TA, Alberts SR, Vierkant RA, Grabrick DM, Cerhan JR, Vachon CM, Olson JE, Kushi LH, Potter JD. High-folate diets and breast cancer survival in a prospective cohort study. Nutr. Cancer 2002;44:139-144.

Sellers TA, Kushi LH, Cerhan JR, Vierkant RA, et al. Dietary folate intake, alcohol, and risk of breast cancer in a prospective study of postmenopausal women. Epidemiol 2001;12:420-428.

Smith-Warner SA, Spiegelman D, Yaun SS, et al. Alcohol and breast cancer in women: a pooled analysis of cohort studies. JAMA 1998;279:535-540.

Stolzenberg-Solomon RZ, Chang SC, Leitzmann MF, Johnson KA, Johnson C, Buys SS, Hoover RN, Ziegler RG. Folate intake, alcohol use, and postmenopausal breast cancer risk in the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial.  Am J Clin Nutr 2006;83:895-904.

Suzuki R, Ye W, Rylander-Rudqvist T, Saji S, Colditz GA, Wolk A. Alcohol and postmenopausal breast cancer risk defined by estrogen and progesterone receptor status: a prospective cohort study.  J Natl Cancer Inst 2005;97:1601-1608.

Tjonneland A, Christensen J, Olsen A, Stripp C, Nissen SB, Overvad K, Thomsen BL. Folate intake, alcohol and risk of breast cancer among postmenopausal women in Denmark. Eur J Clin Nutr 2006;60:280-286.

Zhang S, Hunter DJ, Hankinson SE, Giovannucci EL, Rosner BA, Colditz GA, Speizer FE, Willett WC. A prospective study of folate intake and the risk of breast cancer.  JAMA 1999;281:1632-1637.

Zhang SM, Hankinson SE, Hunter DJ, Giovannucci EL, Colditz GA, Willett WC. Folate intake and risk of breast cancer characterized by hormone receptor status. Cancer Epidemiol Biomarkers Prev 2005;14:2004-2008.

Zhu K, Williams SM. Methyl-deficient diets, methylated ER genes and breast cancer: an hypothesized association.  Cancer Causes Control 1998;9:615-620.

Forum Summary

The strongest factor associated with the risk of a woman developing breast cancer appears to be a positive family history of such a diagnosis in a sibling or mother. Among environmental factors, almost all studies have shown that alcohol consumption relates to increased risk; a slight increase is often seen even among women who report only light drinking, e.g., an average of less than one drink per day.  The present study evaluates how a positive family history of breast cancer in a first-degree relative and folate intake may modify the association between alcohol intake and breast cancer; the analysis is based on data from a large number of young women, aged 27-44 years at baseline, over a follow-up period of 20 years.

Forum members considered that this was a very well-done analysis from the large Nurses’ Health Study II, which followed young women for two decades. Alcohol consumption in total drinks/week was based on repeated assessments (but the pattern of drinking, binge versus regular drinking, was not evaluated).  Repeated dietary questionnaires were used to estimate folate input.  More than 2,800 incident invasive breast cancers were detected, essentially all being validated by a review of medical records.

The main results of this study were that without a positive family history of breast cancer, there was no significant increase in cancer risk for any level of alcohol intake. With a positive family history, however, there was a tendency for higher risk with increasing alcohol intake; this increase was not statistically significant for those with high folate, but a significant positive association (HR=1.82, 95% CI 1.06, 3.12; P-trend = 0.08) with cancer was seen for subjects with the highest level of alcohol intake plus low levels of folate.

Members conclude that the results from the present study should greatly relieve anxiety about breast cancer for women without a positive family history of breast cancer who choose to consume light-to-moderate amounts of alcohol. For women with a first-degree relative with breast cancer, the data indicate that if they maintain a high level of folate intake, it may attenuate an increase in risk associated with alcohol intake.

*                *                      *                      *                      *                      *

Comments on this paper were provided by the following members of the International Scientific Forum on Alcohol Research:

Andrew L. Waterhouse, PhD, Department of Viticulture and Enology, University of California, Davis, USA

David Van Velden, MD, Dept. of Pathology, Stellenbosch University, Stellenbosch, South Africa

Creina Stockley, PhD, MSc Clinical Pharmacology, MBA; Health and Regulatory Information Manager, Australian Wine Research Institute, Glen Osmond, South Australia, Australia

Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark

Harvey Finkel, MD, Hematology/Oncology, Boston University Medical Center, Boston, MA, USA

R. Curtis Ellison, MD, Professor of Medicine, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA