Critique 147: Association of alcohol intake with different types of breast cancer — 11 September 2014

Falk RT, Maas P, Schairer C, Chatterjee N, Mabie JE, Cunningham C, Buys SS, Isaacs C, Ziegler RG.  Alcohol and Risk of Breast Cancer in Postmenopausal Women: An Analysis of Etiological Heterogeneity by Multiple Tumor Characteristics.  Am J Epidemiol 2014; advance access; DOI: 10.1093/aje/kwu189.

Authors’ Abstract

Alcohol consumption is an established risk factor for breast cancer.  Whether associations vary by specific tumor characteristics independent of other characteristics is unclear. 

We evaluated the association between alcohol consumption and breast cancer risk in the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial cohort (54,562 women aged 55–74 years recruited at 10 US screening centers between 1993 and 2001; median follow-up, 8.9 years; 1,905 invasive breast cancer cases).  Hazard ratios and 95% confidence intervals for subtypes defined by histological type and estrogen receptor (ER)/progesterone receptor (PR) status were calculated with standard Cox models.  A novel 2-stage Cox model assessed heterogeneity in risk for individual tumor characteristics while adjusting for others.  

Significant trends across categories of alcohol consumption were observed, with hazard ratios for those consuming 7 or more drinks per week versus never drinkers as follows: for estrogen receptor–positive (ER+) cancer, 1.48 (95% confidence interval (CI): 1.19, 1.83); for progesterone receptor–positive (PR+) cancer, 1.64 (95% CI: 1.31, 2.06); for ER+/PR+ cancer, 1.63 (95% CI: 1.30, 2.05); and for mixed ductal/lobular cancer, 2.51 (95% CI: 1.20, 5.24).  For ER+ and PR+ cancers, trends were significant for ductal and mixed ductal/lobular types.  PR status explained the positive association with ER status (for ER status, Pheterogeneity = 0.70 after adjustment for PR status).  

Alcohol consumption was not associated with all breast cancer subtypes.  Future work should emphasize large collaborative studies, precise definition of subtypes, and adjustment for correlated tumor characteristics.

Forum Comments

Background:  Prospective epidemiologic studies have rather consistently shown a slight increase in the risk of breast cancer for women who consume alcohol.  Usually, the estimated risk for women consuming no more than 1 drink/day is in the range of 5-10% over that of non-drinkers (Collaborative Group on Hormonal Factors in Breast Cancer; Ellison et al).  Associated lifestyle factors (such as dietary folate intake, concomitant hormone replacement therapy, and binge drinking) tend to modify the risk.  Further, adjusting for under-reporting of alcohol intake on self-report has also been shown to attenuate any increase in risk from “moderate drinking” (Klatsky et al, 2014).  For example, in the Klatsky report, women reporting <1 or 1-2 drinks/day who were considered to be likely under-reporters of their alcohol intake (based on other evidence of alcohol misuse in their medical records) had twice the risk of breast cancer as those not considered under-reporters.  Overall, the risk of breast cancer in women reporting ≤ 2 drinks/day who were not considered under-reporters was the same as that of non-drinkers.  Klatsky et al also found that for total cancer, an increased risk associated with moderate drinking was only among people who were likely under-reporters of their intake.  Support of their method for identifying under-reporters has also been found for the relation of alcohol to the risk of hypertension and the risk of liver dysfunction (Klatsky et al, 2006), with abnormalities being found only among heavier drinkers.  

Amount of alcohol consumed and risk of breast cancer:  Reviewer Stockley commented on the differential effects according to dosage of alcohol: “The first purported positive association between alcohol consumption and breast cancer was reported in 1977 (Williams & Horm), and since then approximately 100 epidemiological studies have been published, which consistently support such as association (e.g., Longnecker, Key et al).   However, while the association is consistent and considered confirmed for consumers of three or more drinks per day (Rosenberg et al; Longnecker; van den Brandt et al; Swanson et al; Suzuki et al; Zhang & Holman), for consumers of one to two drinks per day, the data is less consistent or erratic (Flatt et al).  The highest risk has been associated with consumption of six or more drinks per day (Chen et al).  It has also been suggested that consumption patterns may modify risk (Morch et al, Chen et al), such that the consumption of four to five drinks at a session may increase/double risk by 50% compared to only one drink consumed per session.  Paradoxically, alcohol dependence does not increase the risk of breast cancer over that seen in light drinkers (Kuper et al).”

Specific comments on present study:  Data for this study were obtained from the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial, in which women 55-74 years of age were recruited at 10 participating centers in the US for randomization into an arm with usual care or one receiving enhanced screening for these cancers.  Screening among the women was for lung, colorectal, and ovarian cancer, with approximately 28,000 women assigned to the intervention group agreeing to the additional screening (Buys et al). 

In most studies, the types of breast cancer tumor most likely to be associated with alcohol consumption are hormone-receptor-positive tumors:  ER+, PR+, and ER+/PR+ tumors.  The present study is based on data from more than 54,000 women; during a follow- up period of 8.9 years, 1,905 invasive breast cancers were diagnosed.   In line with recent recommendations (Hammond et al), the authors considered a tumor to be hormone-receptor-positive “if at least 1% of tumor cells stain positive.”   The authors also evaluated the association with alcohol intake according to subtypes of the tumor (ductal, lobular, mixed). 

The authors used a proportional hazards model to estimate risk of breast cancer associated with alcohol consumption.  As stated by the authors, “To analyze how the effects of alcohol vary by histological subtype or hormone receptor status while controlling for the other, we used a recently developed modification of the Cox proportional hazards model (Chatterjee et al).”  Reviewers Ellison & Zhang considered the model to be appropriate, but had questions about how women were recruited for participation in the trial.  Ellison noted: “Of 54,781 women who entered the trial and provided adequate data for analysis, about 14% reported a positive family history of breast cancer; more than one half had some college education or greater; 91% were white; more than one half were overweight or obese; and two-thirds had taken or were currently on post-menopausal hormone use.  Some of these percentages suggest that the study cohort may not be representative of the general population.”

There were 1.905 cases of invasive breast cancer diagnosed, of which about two-thirds were stage 1 and less than 2 cm in diameter, and 70% were without nodal involvement.  Overall, approximately 76% were ductal carcinoma, 12% lobular carcinoma, and 6% mixed type; 70% were ER+/PR+, 10% ER+/PR-, and 13% ER-/PR-.  With adjustments for most known risk factors (except folate levels, binge drinking, type of beverage, or HER2 status of tumors; and with only one assessment of alcohol intake, at baseline), there was a significant (p=0.04) increase in risk of cancer for consumers of alcohol.  The RR for categories < 0.5, 0.5-<1, 1-< 7, and ≥ 7 drinks per week were 1.15, 1.25, 1.26, and 1.35, respectively, with all but the lowest category being significantly increased in comparison with never-drinkers.  

Increased risks with alcohol were exclusively among women with tumors that were ER+/PR+, and not in women whose tumors were ER- or PR-, or even ER+/PR-.  For analyses according to histologic type of the tumor, for the highest category of alcohol consumption (versus never drinkers), the risk was increased for ductal (the most prominent type), lobular, and mixed ductal/lobular tumors, but a clear dose-response curve was not present for any type. 

Questions about the present analyses:  Several Forum members considered it puzzling, and probably not physiological, that women reporting less than one-half of a drink per week, as well as those reporting ½ to < 1 drink per week, could have an increase in breast cancer risk that was truly related to their alcohol intake.  Under-reporting of their intake, as described by Klatsky et al, may be important here.  Further, inability to adjust for certain other risk factors (folate levels, drinking patterns, potential changes in intake over time, type of beverage, etc.) may also have played a role.  It would be difficult to use the results of this study for the determination of a threshold for an effect of alcohol on breast cancer risk.

Forum member Van Velden commented: “Environmental factors play an important role in breast cancer, and these factors were not taken into consideration in this article.  It is well-known that people on a vegetarian-like diet (alkaline diet) are less prone to develop breast cancer.  The role of adequate folate intake in the prevention of breast cancer is also a well-known fact.  Other factors such as exercise, being overweight, etc., have to be considered as well.  Finally, we assume that, in the future, genetic factors will also be considered in determining breast cancer risk.  Knowledge of a genetic predisposition for breast cancer and interacting with environmental factors should enable physicians to proactively inform their patients on lifestyle factors that may affect their risk of developing breast cancer.”

Reviewer Skovenborg had two comments to add: “First, the observation of the importance of the PR receptor status, and not necessarily the ER receptor status, is in accordance with the results from the WHI study (Rossouw et al).  In that study, for 10.000 women below the age of 60 randomized to Premarin (pure estrogen) compared to placebo, there were 8 fewer cases of breast cancer.”

Skovenborg continued: “Secondly, the importance of drinking pattern was illustrated by the Danish Nurses’ Cohort Study, as reported by Mørch et al.  During the follow-up period in that study, 457 women were diagnosed with breast cancer.  The relative risk of breast cancer was 2.30 (CI: 1.56-3.39) for reported alcohol intake of 22-27 drinks per week, compared to 1-3 drinks per week.  Among alcohol consumers, weekly alcohol intake increased the risk of breast cancer, with 2% for each additional drink consumed.  Binge drinking of 4-5 drinks on the latest weekday reported increased risk by 55%, compared with consumption of one drink.  A possible threshold in risk estimates was found for consumption of the equivalent of about 3 typical drinks or  more per day.  The authors concluded: ‘For alcohol consumption above the intake most frequently reported, the risk of breast cancer is increased.  The risk is minor for moderate levels but increases for each additional drink consumed during the week.  Weekend consumption and binge drinking imply an additional increase in breast cancer risk.’”

Forum member Goldfinger also found it a bit questionable that very small amounts of alcohol have a true causative effect.  “As much as under- reporting may be at play, the authors may just as well have included the one-social-drink-a-month subject into this low intake group where causation for any real effect would need further explanation.  Again, with respect to breast cancer, beverage type is suggested to play an important role where wine vs spirits play differently in this scenario.  Studies that fail to separate by beverage type are severely limited.”

Considering the net health effects of moderate alcohol intake:  Forum member Stockley provides a good overall statement regarding the net health effects of moderate drinking: “While there is an indisputable association between alcohol consumption and the risk of breast cancer, the mechanisms behind the association require further elucidation. This risk of breast cancer should not, however, be considered in isolation from the risk of other factors for mortality, such as cardiovascular disease, whereby cardiovascular disease is the primary cause of mortality in the industrialised or westernised world.

“Indeed, the light to moderate consumption of alcohol is associated with a significantly reduced risk of mortality from cardiovascular disease and from all causes, for both men and women, irrespective of age and ethnicity.  Thus, it may be advisable for women to enumerate and evaluate their risk factors for cardiovascular disease and for breast cancer before deciding to abstain from alcohol completely.”

References from Forum review

Buys SS, Partridge E, Greene MH, et al, for the PLCO Project Team.  Ovarian cancer screening in the Prostate, Lung, Colorectal and Ovarian (PLCO) cancer screening trial: Findings from the initial screen of a randomized trial.   Am J Obstet Gynecol 2005;193:1630–1639.

Chatterjee N, Sinha S, Diver WR, et al.  Analysis of cohort studies with multivariate and partially observed disease classification data.  Biometrika 2010;97:683–698.

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.

Collaborative Group on Hormonal Factors in Breast Cancer.  Alcohol, tobacco and breast cancer – collaborative reanalysis of individual data from 53 epidemiological studies, including 58 515 women with breast cancer and 95 067 women without the disease.  British Journal of Cancer 2002;87:1234–1245. 

Ellison RC, Zhang Y, McLennan C, Rothman KJ.  Exploring the relation of alcohol consumption to risk of breast cancer.  Am J Epidemiol  2001;154:740-747.

Flatt SW, Thomson CA, Gold EB, et al.  Low to moderate alcohol intake is not associated with increased mortality after breast cancer.  Cancer Epidemiol Biomarkers Prev 2010;19:681-688.

Hammond MEH, Hayes DF, Dowsett M, et al. ASCO-CAP guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer. J Clin Oncol 2010;28:2784–2795.

Key J, Hodgson S, Omar RZ, et al.  Meta-analysis of studies of alcohol and breast cancer with consideration of the methodological issues. Cancer Causes Control 2006;17:759-770.

Klatsky AL (2006), Gunderson EP, Kipp H, Udaltsova N, Friedman GD.  Higher prevalence of systemic hypertension among moderate alcohol drinkers: an exploration of the role of underreporting.  J Stud Alcohol 2006;67:421-428.

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

Kuper H, Ye W, Weiderpass E, Ekbom A, Trichopoulos D, Nyrén O, Adami H-O.  Alcohol and breast cancer risk: the alcoholism paradox. Brit J Cancer 2000;83:949–951.

Longnecker MP.  Alcoholic beverage consumption in relation to risk of breast cancer: meta-analysis and review. Cancer Causes Control 1994;5:73–82.

Mørch LS, et al.  Alcohol drinking, consumption patterns and breast cancer among Danish nurses: a cohort study.  Eur J Public Health 2007;17:624-629.

Rosenberg L, Metzger LS, Palmer JR.  Alcohol consumption and risk of breast cancer: a review of the epidemiological evidence.  Epidemiol Rev 1993;15:133–144.

Rossouw JE, et al.  Postmenopausal hormone therapy and risk of cardiovascular disease by age and years since menopause.  JAMA 2007;297:1465-1477.

Suzuki R, Iwasaki M, Inoue M, et al.  Alcohol consumption-associated breast cancer incidence and potential effect modifiers: the Japan Public Health Center-based Prospective Study.  Int J Cancer 2010;127:685-695.

Swanson CA, Coates RJ, Malone KE, et al.  Alcohol consumption and breast cancer risk among women under age 45 years. Epidemiol 1997;8:231–237.

Van den Brandt PA, Goldbolm A, van‘t Veer P.  Alcohol and breast cancer: results from the Netherlands Cohort Study. Am J Epidemiol 1995;141:907–915.

Williams RR, Horm JW.  Association of cancer sites with tobacco and alcohol consumption and socioeconomic  status of patients interview study from the Third National Cancer Survey. J Nat Cancer Inst 1977;58:525–547.

Zhang M, Holman CD.  Low-to-moderate alcohol intake and breast cancer risk in Chinese women.  Br J Cancer 2011;105:1089-1095.

Forum Summary

The present study was based on a large number of women participating in a clinical trial (enhanced screening for certain cancers versus routine care) focusing on prostate, lung, colorectal, and ovarian cancer.  During a follow-up period averaging about 9 years, a total of 1,905 women were diagnosed with invasive breast cancer.  The analyses showed that an increase in cancer risk was associated with alcohol intake for estrogen and progestin positive (ER+/PR+) tumors, but not for other histologic types of breast cancer.   The increased risk was predominantly seen among PR+ cancers, as there was no evidence of an increase in risk from alcohol for women with ER+/PR- tumors.  An association with the highest category of alcohol intake was noted for ductal (the most prominent type), lobular, and mixed ductal/lobular tumors, but a clear dose-response curve was not present.  The authors conclude that alcohol consumption is not associated with all breast cancer subtypes, and that future research should include precise definition of subtypes of cancer.

Forum reviewers considered this to be a well-done analysis.  There were some questions about the characteristics of the participants in the study that could affect generalizability of the results.  For example, the cohort of women in this study were more highly educated, more likely to have a positive family history of breast cancer, to be smokers, and perhaps more obese that the general population.  Of more concern, however, was the lack of ability to evaluate a number of other potential risk factors for breast cancer: folate levels and other dietary factors, the pattern of drinking (regular versus binge), potential changes in drinking after the baseline measurement, type of beverage, or the HER2 status of tumors.  The finding in the paper of an increase in breast cancer risk, in comparison with non-drinkers, even for women reporting less than ½ drink per week (RR 1.15, CI 0.97, 1.39) and 0.5 – < 1 drink per week (RR 1.25, CI 1.03, 1.53), suggests that residual confounding must also be considered; it is unlikely that such small amounts of alcohol would have a physiological effect.   

Overall, the present study shows a significant increase in the risk of breast cancer among women in a cancer prevention trial who, at baseline, indicated that they were consuming even less than 1 drink/week.  The increase was exclusively in ER+/PR+ tumors, but not in ER+/PR- or ER-/PR- tumors. 

In future research on this association, information on folate levels and drinking pattern, and especially on type of beverage consumed and on evidence of under-reporting of alcohol intake, will be important in better defining how alcohol consumption relates to the risk of breast cancer.  While the results of this study are important in studying the etiology of breast cancer (as alcohol appears to relate only to hormone receptor positive tumors), the findings may not necessarily help individual women know how their alcohol consumption by itself may relate to their overall risk of the disease.

                                                     *                       *                      *                     

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

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

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

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

Tedd Goldfinger, DO, FACC, Desert Cardiology of Tucson Heart Center, Dept. of Cardiology, University of Arizona School of Medicine, Tucson, Arizona, USA

Erik Skovenborg, MD, Scandinavian Medical Alcohol Board, Practitioner, Aarhus, Denmark

Yuqing Zhang, MD, DSc, Epidemiology, Boston University School of Medicine, Boston, MA, USA

Ulrich Keil, MD, PhD, Institute of Epidemiology and Social Medicine,  University of Münster, Münster, Germany

Dominique Lanzmann-Petithory,MD, PhD, Nutrition/Cardiology, Praticien Hospitalier Hôpital Emile Roux, Paris, France

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