This article is more than 5 years old. Information may no longer be current.
High-dose selenium supplementation may increase risk for prostate cancer mortality
Click Here to Manage Email Alerts
Selenium supplements, especially in higher doses, may increase the risk for prostate cancer-related death if administered after diagnosis, according to the results of a collaborative study.
Men who consumed ≥140 mcg/day of supplemental selenium demonstrated a 2.6-fold greater risk for prostate cancer mortality compared with men who did not take supplemental selenium, results showed.
Stacey A. Kenfield
“There is no evidence for benefit, and now we have evidence for harm,” Stacey A. Kenfield, ScD, assistant professor of urology at University of California, San Francisco, told HemOnc Today. “Based on these findings, prostate cancer patients should avoid selenium supplements.”
Prior studies have failed to demonstrate a benefit for selenium supplementation for prevention of prostate cancer. However, the effect of post-diagnosis selenium supplementation on prostate cancer outcomes has not been extensively studied.
Kenfield and colleagues assessed whether selenium supplementation was associated with biochemical recurrence, prostate cancer mortality, cardiovascular disease mortality or overall mortality.
The analysis included 4,459 men enrolled in the Health Professionals Follow-Up Study from 1988 through 2010. All men were initially diagnosed with nonmetastatic prostate cancer.
Median follow-up for mortality was 8.9 years. During that time, 965 participants died. Of these deaths, 226 (23.4%) were attributable to prostate cancer and 267 (27.7%) were attributable to cardiovascular disease.
Researchers reported crude prostate cancer mortality rates of 5.6 per 1,000 person-years among selenium non-users and 10.5 among those who consumed ≥140 mcg/day. Multivariable analysis showed men who consumed 1 mcg/day to 24 mcg/day had a 1.18-fold (95% CI, 0.73-1.91) greater risk for prostate cancer mortality vs. nonusers. That risk increased to 1.33-fold (95% CI, 0.77-2.3) among men who consumed 25 mcg/day to 139 mcg/day, and to 2.6-fold (95% CI, 1.44-4.7) among men who consumed ≥140 mcg/day.
Crude rates of all-cause mortality were 28.2 per 1,000 person-years for nonusers and 23.5 per 1,000 person-years for highest-dose users.
Median follow-up for biochemical recurrence was 7.8 years. During that time, researchers reported 762 recurrences. Crude recurrence rates were 28.4 per 1,000 person years for non-users and 29.3 per 1,000 person years for the highest-dose users.
On average, men in the United States consume 134 mcg/day of selenium, Kenfield and colleagues wrote. That is more than twice the recommended dietary allowance of 55 mcg/day, the average daily dietary intake that is sufficient to meet the nutrient requirements of nearly all healthy individuals.
Therefore, most American men do not need to take supplemental selenium for their health, and clinicians should caution patients about taking high-dose selenium supplements, Kenfield said.
“At this time, there is no strong evidence that any single vitamin or mineral supplement can offer protection against prostate cancer progression,” Kenfield said. “People should not assume that if a product is ‘natural’ then it is safe. Further research evaluating high-dose selenium intake is needed to confirm our results and inform clinical guidelines for prostate cancer survivors.” – by Anthony SanFilippo
Disclosure: The study was supported by grants from the NCI. The researchers report no relevant financial disclosures.
Perspective
Back to Top
Donald L. Trump, MD, FACP
Kenfield and colleagues examined the relationship between selenium supplementation and prostate cancer mortality among participants in the Health Professionals Follow-up Study. This paper emphasizes two important points relative to research in cancer causation and outcome.
First, it demonstrates the power of a carefully constructed, longitudinal study in which detailed data regarding patient behaviors, characteristics and outcomes are rigorously collected over time.
The authors examined data from 4,459 men initially diagnosed with nonmetastatic prostate cancer in the Health Professionals Follow-Up Study from 1988 through 2010. Researchers evaluated whether selenium supplement use after diagnosis was associated with risk for biochemical recurrence, prostate cancer mortality, cardiovascular disease mortality and overall mortality.
With a median follow-up of 8.9 years among more than 4,000 men, 965 died of prostate cancer. Crude rates for prostate cancer deaths per 1,000 person-years were 5.6 among selenium nonusers and 10.5 among men who consumed ≥140 μg/day. Multivariable analyses showed increased risk for prostate cancer mortality among men who consumed 1 to 24 μg/day (RR=1.18; 955 CI, 0.73-1.91), 25 to 139 μg/day (RR=1.33; 95% CI, 0.77-2.3), and ≥140 μg/day (RR=2.6; 95% CI, 1.44-4.7) of supplemental selenium compared with nonusers (P=.001 for trend).
These are intriguing data, and they are consistent with other data that indicate selenium supplementation — while eminently logical — is NOT a good idea. I suspect more than a few of my colleagues have recommended to their patients that selenium supplementation might be helpful (this is what I did for some time!). Three cheers for data, even if derived from a registry — in this case, a very well conducted registry.
This paper also demonstrates retrospective study of associations, even in large numbers of individuals, can provide misleading or even dangerous data
There are numerous examples in which evaluation of the association between characteristic A with disease X generates the hypothesis that having a higher (or lower) level of A is good (or bad) if you have disease X. These examples include beta carotene, vitamin A, vitamin D, selenium, vitamin E … and the list goes on. When these associations provide rationale to recommend an intervention that should be harmless, it often leads the practicing physician to take action. All too often, that action turns out to be worthless — or in the case of selenium, potentially harmful.
A reasonable dictum might be: When there are no large, prospective comparative data to support taking an action, don’t!
Click Here to Manage Email Alerts