Downer MK, Kenfield SA, Stampfer MJ, Wilson KM, Dickerman BA, Giovannucci, EL, et al. Alcohol Intake and Risk of Lethal Prostate Cancer in the Health Professionals Follow-Up Study abstract. J Clin Oncology 2019 Pre-publication: : DOI https://doi. org/10.1200/JCO.18. 02462
PURPOSE It is unknown whether alcohol intake is associated with the risk of lethal (metastatic or fatal) prostate cancer. We examine (1) whether alcohol intake among men at risk of prostate cancer is associated with diagnosis of lethal prostate cancer and (2) whether intake among men with nonmetastatic prostate cancer is associated with metastasis or death.
METHODS This prospective cohort study uses the Health Professionals Follow-Up Study (1986 to 2012). Our analysis of alcohol intake among men at risk of prostate cancer included 47,568 cancer-free men. Our analysis of alcohol intake among men with prostate cancer was restricted to 5,182 men diagnosed with nonmetastatic prostate cancer during follow-up. We examine the association of total alcohol, red and white wine, beer, and liquor with lethal prostate cancer and death. Multivariate Cox proportional hazards regression estimated hazard ratios (HRs) and 95% CIs.
RESULTS Alcohol drinkers had a lower risk of lethal prostate cancer (any v none: HR, 0.84 [95% CI, 0.71 to 0.99]) without a dose-response relationship. Total alcohol intake among patients with prostate cancer was not associated with progression to lethal prostate cancer (any v none: HR, 0.99 [95% CI, 0.57 to 1.72]), whereas moderate red wine intake was associated with a lower risk (any v none: HR, 0.50 [95% CI, 0.29 to 0.86]; P trend = .05). Compared with none, 15 to 30 g/d of total alcohol after prostate cancer diagnosis was associated with a lower risk of death (HR, 0.71 [95% CI, 0.50 to 1.00]), as was red wine (any v none: HR, 0.74 [95% CI, 0.57 to 0.97]; P trend = .007).
CONCLUSION Cancer-free men who consumed alcohol had a slightly lower risk of lethal prostate cancer compared with abstainers. Among men with prostate cancer, red wine was associated with a lower risk of progression to lethal disease. These observed associations merit additional study but provide assurance that moderate alcohol consumption is safe for patients with prostate cancer.
While a number of cancers, especially those of the upper aero-digestive tract (e.g., mouth, tongue, larynx) are known to be more frequent among heavy drinkers and alcoholics than among light drinkers or non-drinkers, the association between alcohol consumption and prostate cancer is less clear. Further, an increasing number of studies are showing that consumers of red wine tend to have less cancer than subjects reporting similar consumption of other alcoholic beverages. The present study provides valuable data on these topics as it is from a very large study of men who have been followed for long periods of time in the Health Professionals Follow-up Study, with good assessments of alcoholic beverage intake and near complete outcome assessment (non-metastatic prostate cancer and “lethal prostate cancer”).
The study from which these analyses are based is the Health Professionals Follow-up Study with a total of more than 47,000 men who have been followed for an average of 26 years. Among these subjects, non-metastatic prostate cancer was diagnosed in 5,182 subjects, and their subsequent course was related to their post-diagnosis drinking habits. Lethal prostate cancer (widely metastatic or leading to death) occurred in 869 cases. There were repeated assessments of alcohol (each 4 years), so that recent alcohol intake could be related to the risk of prostate cancer occurring over the next four years. The present analyses compare the effects of total alcohol as well as of beer, white wine, red wine, and liquor.
It is especially important that cancer diagnosis was validated by medical record review in the large majority of subjects (more than 80%), and repeat communications with subjects and review of several listings of deaths permitted essentially complete assessments of mortality. The study also was large enough to evaluate for effects of specific beverages; in all of their analyses, it appears that consumers of red wine have the most favorable outcomes. The two outcomes reported in this paper were non-metastatic prostate cancer and lethal prostate cancer or death.
As for the results of the analyses, among all 47,568 men being followed, there was a significant 16% lower risk of lethal prostate cancer for subjects reporting “any alcohol intake”, but no significant relation between the intake of any specific beverage and cancer. The investigators then related the effects of alcohol consumption after the diagnosis of cancer and found that among the 5,182 men who developed nonmetastatic prostate cancer, there was an inconsistent relation of the consumption of alcohol and of most types of beverages with the development of lethal cancer. In contrast, there was a significantly 50% lower risk among subjects reporting the consumption of red wine (HR 0.50, CI 0.29-0.86), in comparison with non-drinkers. A statistically significant reduction in risk was not seen for white wine or for other beverages.
Forum member Ellison commented on the very high quality of the analyses: “The investigators had extensive data on other lifestyle factors that may have had an effect on prostate cancer risk, including diet, smoking, body mass index, vigorous physical activity, intake of choline and vegetable fat, coffee, lycopene, whole milk, and diabetes. The investigators also adjusted for prostate-specific antigen (PSA) screening. All of these potentially confounding variables were updated at each contact with the subject, and several indices of previous alcohol consumption were used to avoid contaminating their analyses with “sick quitters”. And, to reduce the risk of reverse causation (subjects developing prostate cancer and then modifying their alcohol intake), they varied their exposure windows from 2-4, 4-6, and 6-8 years before the current exposure period (lagged analyses). Given their repeated assessments of alcohol intake over many decades, they were able to also evaluate for changes in cancer risk associated with changes in alcohol consumption. Finally, they were able to evaluate for interactions between amount and frequency of alcohol intake. Overall, this is a superb dataset for evaluating the risk of prostate cancer associated with alcohol-containing beverages. The large size and quality of analysis make the reduction associated with red wine especially meaningful.”
Stated Forum member Van Velden: “I find this a very good study confirming previous observations on the possible protective effects of moderate alcohol, specifically for red wine consumption. It was based on almost 50,000 men over a very long follow-up period, which allows for valid conclusions. The authors do mention that red wine drinkers tended to have a healthier lifestyle, but took that into consideration in their final analysis. It is difficult to explain the mechanism of protection against malignancy, and confounding positive lifestyle factors might also play a role.”
Reviewer Finkel noted: “The results and conclusions of this large and long-term study are presented responsibly by an experienced group. The study addresses relationships of particular importance because of the large number of men carrying the diagnosis of prostate cancer. One caveat not directly addressed: Might the improved survival from lowered risk of cardiovascular disease incident upon moderate drinking have raised the baseline incidence of cancer among those aging survivors?” Other Forum members agreed that this is a possibility, given that moderate red wine consumers have consistently been shown to live longer (primarily from less cardiovascular disease) than non-drinkers as well as when compared with consumers of other beverages.
Potential mechanisms for a reduced risk of lethal prostate cancer from red wine: Forum member Skovenborg wrote: “Very interesting study of the association between moderate alcohol consumption and the risk of lethal prostate cancer; a cancer that in many ways is the male equivalent of female breast cancer. One aspect is the association of these cancers with respectively testosterone and estrogen, and one explanation of the opposite effect regarding alcohol consumption and risk of respectively prostate cancer and breast cancer may be that alcohol may increase sex hormone-binding globulin levels and lower testosterone, while alcohol intake is associated with increased estrogen levels, especially in postmenopausal women.
“Regarding the ever-present risk of confounding, the investigators have done a good job in adjusting for known and potential confounders, and the homogenous study population reduces the risk of confounding by sociodemographic factors. The lower risk of progression to lethal breast cancer associated with moderate red wine consumption is certainly reassuring for male wine drinkers; however, the possible biological pathways quoted by the authors mostly build on results of test tube study of resveratrol and cancer cells.”
In an analysis from the Health Professionals Follow-up Study, the investigators evaluated the relation of alcohol consumption in their more than 45,000 study participants to the development of lethal prostate cancer. They related alcohol first to the overall risk of prostate cancer among all subjects, and then focused on the 5,182 men who developed non-metastatic prostate cancer to judge their risk of developing lethal prostate cancer or mortality. They judged the alcohol intake both prior to the diagnosis and after a diagnosis of prostate cancer. For reported alcohol consumption among subjects prior to the initial diagnosis of prostate cancer, there was a small (16%) but significant reduction in the risk of lethal prostate cancer among subjects reporting any alcohol consumption, but no significant effects of specific beverages on the risk of lethal prostate cancer. However, total mortality was lower for consumers of total alcohol intake and for most categories of intake for all types of beverage.
When they then related post-diagnosis alcohol intake (after the diagnosis of non-metastatic prostate cancer), the investigators found that red wine consumption was associated with a 50% reduction in the risk of developing lethal prostate cancer or mortality. There were no significant relations between total alcohol consumption or between the intake of other specific beverages and lethal disease.
The authors describe a number of hypotheses by which red wine, with its combination of high levels of polyphenols in addition to alcohol, may have lowered the risk of lethal prostate cancer. And there are numerous basic science experiments that have shown how polyphenols impede the development and growth of cancer cells. Presumably, it was not due to the alcohol in red wine, as similar amounts of alcohol in other beverages did not show a protective effect. While the effects of various polyphenols may play a role, at present the mechanisms of such an effect are not known.
Forum reviewers considered this to be a very well-done study and analysis, with repeated assessments of alcohol intake and essentially complete ascertainment of lethal prostate cancer and mortality. Forum members agree with the conclusions of the authors: “Our results indicate that moderate alcohol intake among men with prostate cancer is not associated with a higher risk of progression to lethal disease or overall mortality. The potential benefit of red wine on prostate cancer progression merits additional research.”
* * * * * * *
Comments on this critique by the International Scientific Forum on Alcohol Research have been provided by the following members:
David Van Velden, MD, Dept. of Pathology, Stellenbosch University, Stellenbosch, South Africa
Arne Svilaas, MD, PhD, general practice and lipidology, Oslo University Hospital, Oslo, Norway
Creina Stockley, PhD, MSc Clinical Pharmacology, MBA; Adjunct Senior Lecturer at the University of Adelaide, Australia
Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark
Giovanni de Gaetano, MD, PhD, Department of Epidemiology and Prevention, IRCCS Istituto Neurologico Mediterraneo NEUROMED, Pozzilli, Italy
Harvey Finkel, MD, Hematology/Oncology, Retired (Formerly, Clinical Professor of Medicine, 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