Nutrients 2013, 5, 1122-1148; doi:10.3390/nu5041122
Selenium and Prostate Cancer Prevention: Insights from the
Selenium and Vitamin E Cancer Prevention Trial (SELECT)
Holly L. Nicastro 1 and Barbara K. Dunn 2,*
1 Cancer Prevention Fellowship Program, Nutritional Science Research Group, Division of Cancer
Prevention, National Cancer Institute, 9609 Medical Center Dr, Rockville, MD 20850, USA;
2 Chemoprevention Agent Development Research Group, Division of Cancer Prevention, National
Cancer Institute, 9609 Medical Center Dr, Rockville, MD 20850, USA
* Author to whom correspondence should be addressed; E-Mail: firstname.lastname@example.org;
Tel.: +1-301-402-1209; Fax: +1-301-496-8667.
Received: 22 January 2013; in revised form: 11 March 2013 / Accepted: 19 March 2013 /
Published: 3 April 2013
Abstract: The Selenium and Vitamin E Cancer Prevention Trial (SELECT) was conducted
to assess the efficacy of selenium and vitamin E alone, and in combination, on the incidence
of prostate cancer. This randomized, double-blind, placebo-controlled, 2 × 2 factorial
design clinical trial found that neither selenium nor vitamin E reduced the incidence of
prostate cancer after seven years and that vitamin E was associated with a 17% increased
risk of prostate cancer compared to placebo. The null result was surprising given the strong
preclinical and clinical evidence suggesting chemopreventive activity of selenium.
Potential explanations for the null findings include the agent formulation and dose, the
characteristics of the cohort, and the study design. It is likely that only specific
subpopulations may benefit from selenium supplementation; therefore, future studies
should consider the baseline selenium status of the participants, age of the cohort, and
genotype of specific selenoproteins, among other characteristics, in order to determine the
activity of selenium in cancer prevention.
Keywords: selenium; SELECT; prostate cancer; chemoprevention
Nutrients 2013, 5
Prostate cancer is the second leading cause of cancer death among men in the United States (US)
and is the most commonly diagnosed non-cutaneous cancer, with 1 in 6 men expected to be diagnosed
with this disease in their lifetimes . In 2012, an estimated 241,740 new cases were diagnosed in the
US and approximately 28,170 men died of prostate cancer . While 81% of prostate cancers are
diagnosed in the early stage and treated effectively with surgery or radiation, these treatments often
result in poorer quality of life due to side effects like incontinence, impotence, or declining bowel
function [3,4]. Current treatments for advanced prostate cancer are largely palliative. Many known risk
factors for prostate cancer are non-modifiable, including age, race, and genetic factors, whereas
modifiable risk factors associated with prostate cancer include obesity, physical activity, and possibly
dietary factors . Prostate cancer screening is controversial due to overdiagnosis and overtreatment of
non-fatal disease, and the United States Preventive Services Task Force strongly recommends against
prostate-specific antigen (PSA) screening for prostate cancer . Because prostate cancer has a long
natural history, mainly non-modifiable risk factors, and an incidence rate that far exceeds the mortality
rate, a focus on prevention over screening or early detection offers an appealing area of investigation.
Goals for prevention strategies should focus on reducing cancer incidence and delaying cancer
diagnosis until the individual succumbs to other causes .
Because of the established role of androgens in prostate carcinogenesis and the common use of
anti-androgenic therapy for treatment of advanced or recurring prostate cancer, the first large
prevention trials for prostate cancer targeted androgens. The Prostate Cancer Prevention Trial (PCPT),
sponsored by the National Cancer Institute (NCI) and conducted by the Southwest Oncology Group
(SWOG), was the first such trial. The aim of PCPT was to determine whether the 5α-reductase
inhibitor (5ARI) finasteride would reduce the prevalence of prostate cancer after 7 years of
treatment . Because 5α-reductase catalyzes the conversion of testosterone to the more potent
androgen dihydrotestosterone and androgens are promoters of prostate carcinogenesis, investigators
hypothesized that pharmacological inhibition of this enzyme would reduce prostate cancer prevalence.
Men (n = 18,882) over the age of 55 without evidence of prostate cancer detected by digital rectal
exam (DRE) or prostate-specific antigen (PSA) levels were randomized to receive either 5 mg/day
finasteride or placebo for 7 years. Annual DRE and PSA tests were administered and prostate biopsies
were recommended for patients with abnormal results. All men without prostate cancer diagnoses at
the end of the study were also requested to undergo biopsies; 7551 men agreed to this end-of-study
biopsy. After 7 years, the prevalence of prostate cancer was reduced by 24.8% in the finasteride group
compared to the control group. However, this promising result was accompanied by a 27% increase in
the rate of high-grade prostate cancer (defined as having a Gleason score of 7 to 10) in the finasteride
group, dampening enthusiasm for the use of finasteride as a chemopreventive agent. Among the
reasons offered to explain this unexpected outcome included detection bias in the finasteride group.
Detection bias in the finasteride group was thought to be due to increased sampling density because
finasteride reduces the volume of the prostate gland .
The Reduction by Dutasteride of Prostate Cancer Events (REDUCE) study was designed to
determine the effect of dutasteride on incident prostate cancer. Dutasteride inhibits 5α-reductase types 1
and 2 while finasteride only inhibits type 1. In this four-year multicenter, randomized, double-blind,
Nutrients 2013, 5
placebo-controlled, parallel group study, 6729 men were enrolled. In addition to a negative baseline
biopsy, inclusion criteria were based on factors that placed these men at high risk of prostate cancer
including age, slightly elevated serum PSA levels (2.5 to 10.0 ng/mL), or previous prostate biopsies
due to suspected cancer. Participants were randomized to receive either 0.5 mg dutasteride or placebo
daily for 6 months. Free and total PSA levels were measured every six months and biopsies were
performed after 2 and 4 years or when clinically indicated. Dutasteride was associated with a relative
risk reduction of prostate cancer of 22.8%. During the four-year study period, rates of high grade
prostate cancer were similar between the dutasteride and the placebo group, though in years 3 and 4,
there was a small statistically significant increase in rates of tumors with Gleason scores of 8–10 in the
dutasteride group . Due to the concerns about increasing risks of high grade prostate cancer, a US
Food and Drug Administration (FDA) advisory panel voted overwhelmingly not to approve finasteride
or dutasteride for prostate cancer prevention .
Concomitant with the interest in anti-androgens, a totally independent approach to prostate cancer
chemoprevention involved nutritional agents, specifically vitamin E and selenium. Secondary analyses
of other large-scale chemoprevention trials had suggested that these compounds may decrease risk of
prostate cancer [12,13]. Further controlled intervention trials, human observational studies, and
preclinical studies all provided evidence for potential chemopreventive efficacy of these compounds.
Adding to the appeal, both agents are naturally-occurring micronutrients essential to human health that
have antioxidant activities. In this review, we will describe the rationale, results, and implications of
the Selenium and Vitamin E Cancer Prevention Trial (SELECT).
2.1. Dietary Sources and Supplements
Selenium is a nutritionally essential trace mineral. Selenium enters the food chain from the soil in
the form of selenate (SeO42) or selenite (SeO3−2) and is converted in plants to organic forms, largely
L-selenomethionine and to a lesser extent L-selenocysteine [14,15]. Selenium concentrations in foods
can therefore vary widely based on the selenium content of the soil. For example, Ireland, Israel, and
the western US have high soil selenium content, while certain regions of China have very low soil
selenium content . In fact, Keshan disease, a congestive cardiomyopathy, first observed in Keshan
County of Heilongjiang province, Northeast China, was found to be caused by a combination of
dietary deficiency of selenium and the presence of a mutated strain of Coxsackievirus .
The richest dietary sources of selenium are Brazil nuts, meats, fish, eggs, and cereals. Selenium is
also found in lesser amounts in cruciferous vegetables, garlic, and mushrooms [18,19].
Selenium is available in supplement form as selenomethionine or as selenized yeast, yeast grown in
a selenium-rich medium. Commercially available selenized yeast can provide up to 1000 to 2000 μg/g
selenium, over 90% of which is selenomethionine . Certain selenium supplements, particularly
weight loss products or infant formulas, contain sodium selenite or sodium selenate, though these
inorganic forms are not highly bioavailable.
Nutrients 2013, 5
2.2. Selenium Metabolism and Biological Activities
Dietary selenium, as selenomethionine, selenocysteine, selenate, or selenite, is essential for
selenoprotein synthesis. Selenomethionine can be nonspecifically incorporated into proteins in place of
methionine or converted to selenocysteine via a trans-sulfuration pathway. Selenocysteine, either from
the diet or derived from selenomethionine, can be converted to hydrogen selenide, a key metabolite
integral to both selenocysteine insertion into proteins and selenium excretion. Selenate is reduced to
selenite by glutathione, and selenite undergoes further glutathione reduction to hydrogen selenide
(Figure 1) [21–23]. Therefore, all dietary forms of selenium can be used for selenoprotein synthesis
following conversion to hydrogen selenide.
Figure 1. Selenium biology.
Nutrients 2013, 5
Selenophosphate synthetase converts hydrogen selenide to selenophosphate. Selenophosphate then
reacts with L-seryl-tRNA to form L-selenocysteinyl-tRNA, the latter of which is inserted into
selenoproteins where specified by the UGA codon in mRNA . Selenocysteine, considered the 21st
amino acid, is integral to the activity of selenoproteins . Twenty-five human selenoproteins have
been identified, including glutathione peroxidases (GPx), thioredoxin reductases (TR), thyroid hormone
deiodinases, selenophosphate synthetase, and several uncharacterized proteins (reviewed in ).
These selenoproteins play an important role in maintaining redox balance and proper cellular
Hydrogen selenide can also be mono-, di-, or tri-methylated for excretion. Methyl selenol is the
major urinary excretory form of selenium. With larger doses of selenium, dimethyl selenium is exhaled
from the lungs and the trimethylselenonium ion is excreted in the urine . While the methylation
pathway is the main excretion pathway of selenium, selenosugars have also been reported in the
urine . The activities of intracellular selenium metabolites including methyl selenol determine the
clinical efficacy of selenium, including its chemopreventive effects [29–31].
2.3. Nutritional Requirements
The recommended dietary allowance (RDA) for selenium for adult men and women is
55 μg/day . This RDA is based on the daily selenium intake necessary for maximal activity of
GPx-3. However, with an intake of 55 μg/day, not all selenoproteins’ activity levels would be
maximal. Others have suggested that an RDA of 80 μg/day for men is more appropriate for achieving
selenium balance . Deficiency symptoms, including loss of immunocompetency, progression of
viral infections, and reproductive symptoms, became apparent with intake <11 μg/day [34,35].
Symptoms of selenium toxicity, or selenosis, including hair and nail brittleness/loss, gastrointestinal
disturbances, skin rash, garlic breath odor, fatigue, or irritability, appeared with 800 μg/day of
selenium intake. Based on this observation and using an “uncertainty factor” of 2, half of 800 μg/day,
i.e., 400 μg/day is considered the tolerable upper limit .
According to reports using data from the National Health and Nutrition Examination Survey
(NHANES) 2003–2006, a nationally representative cross-sectional survey, the usual intake for
individuals over the age of 19 was 109 ± 1 μg/day from naturally-occurring dietary sources and
126 ± 1 μg/day from naturally occurring sources plus supplements. Less than 1% of adults had intakes
below the estimated average requirements (EAR) and only 0.1% ± 0.4% had intakes above the
tolerable upper limit .
NHANES data from 2003 to 2006 also show that 19% ± 1% of males reported taking a dietary
supplement that contained selenium, including multivitamin-multimineral supplements. Older men
were more likely to take any dietary supplements, including those containing selenium. Of men
51–70 years of age and over the age of 70, 30% ± 2% and 32% ± 2% were taking supplemental
selenium, respectively . Overall, for all individuals, users of any dietary supplement were more
likely to be of non-Hispanic white race, have lower BMI, more education, be less likely to smoke, be
more physically active, and consume more fruits and vegetables. For men, but not women, selenium
supplement users had higher food selenium intake than non-selenium supplement users, suggesting
that those taking supplements were not doing so out of nutritional need. However, selenium