Download Serum Levels of Dehydroepiandrosterone and

Vol.
2, 33-35,
January/February
1993
Cancer
Epidemiology,
Biomarkers
& Prevention
I I
Serum Levels of Dehydroepiandrosterone
and
Dehydroepiandrosterone
Sulfate and the Risk
of Developing
Gastric Cancer1
Gary B. Gordon,2 Kathy J. Helzlsouer,
Anthony J. Alberg, and George W. Comstock
Introduction
Gastric
Oncology
Center
[C. B. C., K. J. H.] and Department
of Pharmacology
and Molecular
Sciences
[C. B. C.], The Johns Hopkins
University
School
of Medicine,
and Department
of Epidemiology
[K. 1. H., A. I. A.,
C. W. C.], The Johns Hopkins
University
School
of Hygiene
and Public
Health,
Baltimore,
Maryland
21205
cancer
Although the incidence
of gastric cancer varies widely
between countries it is nonetheless
a leading cause of
cancer deaths worldwide.
Migration
studies indicate
that dietary choices are an important
exogenous
factor.
The United States has a very low incidence
of gastric
cancer, suggesting that exogenous
etiological
agents
are at a minimum
and providing
a favorable
setting for
detecting
important
endogenous
etiological
factors.
Dehydroepiandrosterone
and dehydroepiandrosterone
sulfate are endogenous
steroids produced
in the
adrenal gland. Epidemiological
studies show that the
risk of developing
specific cancers is related to the
serum
or urinary
levels
of these
steroids.
In addition,
dehydroepiandrosterone
prevents a variety of
spontaneous
and chemically
induced tumors when
administered
to laboratory
animals. To examine the
association
between circulating
levels of
dehydroepiandrosterone
and dehydroepiandrosterone
sulfate and the development
of gastric cancer, we
measured the serum levels of these steroids in 13
individuals
who donated serum to the Washington
County Maryland
serum bank in 1974 and who
subsequently
developed
gastric cancer and in 52
matched controls. Prediagnostic
serum levels of
dehydroepiandrosterone
were 38% lower in cases as
compared
to controls (P = 0.09). The risk of developing
gastric cancer increased with decreasing
levels of both
steroids.
Adjustment
smoking
or the interval
for confounding
between
time to diagnosis
did not alter
results suggest that there may
in the prevention
Received
1
of gastric
factors
blood
such
donation
as
and
the findings.
These
be a role for this steroid
cancer.
4/6/92.
Supported
by Grants
CA36390
and
CA44530
from
the
National
Cancer
Institute,
Department
of Health
and Human
Services.
K. I. H. is a recipient
of Preventive
Oncology
Academic
Award
CAO1 522-01
from the National
Cancer
Institute,
Department
of Health
and Human
Services.
C. W. C. is
a recipient
of Career
Research
Award
HL21670
from the National
Heart
Lung and Blood
Institute,
Department
of Health
and Human
Services.
2 To whom
requests
for reprints
should
be addressed,
at Department
of
Pharmacology
and Molecular
Sciences,
The lohns
Hopkins
School
of
Medicine,
725 N. Wolfe
St., Baltimore,
MD 21205.
one
of
death
from cancer
in 1930,
(2). In 1992
it is estimated
diagnosed
Abstract
remains
the
leading
causes
of
cancer death in the world.
It is particularly
prevalent
in
areas such as Japan and Chile (1). In the United
States
the incidence
of gastric cancer,
the leading
cause of
deaths
(2, 3).
in 23,800
(2).
Men
Several
individuals
will
women
studies
hormones
over
cancer
and account
outnumber
case-control
els of endogenous
has decreased
that gastric
for 13,400
by
almost
have examined
including
10-fold
will be
DHEA5
2:1
the 1evin patients
with gastric cancer. Dehydroepiandrosterone
and its sulfate conjugate
DHEAS are produced
in the adrenal gland
in response
to adnenocorticotropin
(4). Serum levels of
these
hormones
crease
peak
profoundly
incidence
of
atherosclerosis
in young
and
adulthood
continuously
proliferative
rises (4-6).
and
with
then
age
de-
as the
diseases
such
as cancer
and
The risk of developing
specific
cancers and atherosclerosis
has, in general,
been associated with low serum levels of steroids (7- 14). Three casecontrol
studies of gastric cancer have consistently
found
that patients
presenting
with all but the most advanced
stages have a decreased
urinary
excretion
of DHEA(S)
relative
to controls
(15-17).
No prospective
studies
of
the association
between
serum
or urine
levels of DHEA
or DHEAS and the risk of developing
gastric cancer have
been conducted.
We conducted
a nested case-control
study to evaluate the association
between
prediagnostic
serum levels
of DHEA and DHEAS and subsequent
risk for developing
gastric cancer. The study is prospective
in that the serum
was
obtained
Materials
prior
to the
diagnosis
of cancer.
and Methods
Description
of the Washington
Washington
detail
(18).
County
Briefly,
County
Serum
Bank
from
August
Serum
1974,
blood samples and basic demographic
were collected
from 20,305
residents
of
County,
Maryland.
Serum donors were more
better educated,
nonsmokers,
and women.
ranged from 1 1 to 98 years, with the highest
rate among those aged 55 to 64 years.
Selection
cases
of Cases and Controls.
were
‘ The’
al)breviations
de’hyclroepiandrosfe’rone’
identified
use’d
by
Incident
comparing
,ire’: DH EA,
sulf,ste’.
Bank. The
has been described
through
November
information
Washington
likely to be
Participants
participation
gastric
the
de’hydroe’piandroste’rone;
in
of
cancer
Washington
HF-I EAS,
Downloaded from cebp.aacrjournals.org on June 17, 2017. © 1993 American Association for Cancer Research.
34
Serum
Levels
T,ib(e’
1
of
DHEA
Me’(lian
l)HEA
,iiliong
Steroid
h(irFiiOii(’
and
Risk
DHEAS
c an
(,slc uIaR’d
Wilcoxiin
,is )(as(’
sign rink
nie’cli,sn
test.
Gastric
ase’s
-
range’)
Pe’rce’ntage’
difference”
52)
fl
Cancer
(,ind inle’rciuartile
and c ontrols
Controls
7.02
to 10.87)
2.75
) 1 .68 to 4.45)
4.36
I 3. 5(, to 1)89)
2.27
I 1 .29 to 3.42(
)Piiiol/nil)
DIIEAS
)nniol/iiil(
of
k’ve’Is
er
Case’s
(n
1 3)
l)FIEA
S
the
gastric
(units(
,‘
and
-37.9
0.09
-17.5
0.77
) 3.46
.
ontri)I
iiie’clian)/control
nie’clian[
x
100.
County
Cancer
Register
with
the list of donors
to the
serum
bank.
Of the 14 cases of gastric
cancer
(International
Classification
of Diseases
category
1 59) that developed
in this cohort
from
1975 to 1989,
13 adenocarcinomas
were
included,
and one, a leiomyosarcoma,
was
not. Controls
were selected
from
individuals
who were
alive and free from other
cancers,
except
for nonmelanomatous
skin cancers,
at the time
the case was diagnosed.
Four controls
were matched
to each case by age,
sex, race (all cases were white),
and hours since last meal.
ing has been associated
with
increased
levels
of DHEA
and DHEAS
(23) as well as gastric
cancer
(24) and therefore may confound
the association
between
DHEA
and
gastric cancer.
Adjusting
for the effect
of cigarette
smoking history
(even and never
smokers)
did not alter the
observed
associations.
The risk of gastric
cancer
decreased
with increasing
levels of both DHEA and DHEAS,
with the trend
slightly
stronger
for DHEA.
Relative
to the
lowest
tertile,
high DHEA
and DHEAS
levels were associated
with a 5-fold
and 3-fold
decrease,
respectively,
in
the risk of developing
gastric
cancer
(P = 0.06, 0.1 7).
Since these findings
may be influenced
by subclinical disease,
reestimates
were
obtained
from
analyses
limited
to the 10 cases diagnosed
3 on more years after
donating
serum.
The trend
of decreasing
risk of gastric
cancer
with
increasing
levels
of DHEA
and
DHEAS
remained.
Discussion
trend
in odds ratios with increasing
hormone
levels was
assessed
using the likelihood
ratio test from
the conditional
logistic
regression
equation,
with hormone
tertile
entered
as a single quantitative
variable
(22). The median
hormone
value
for each tertile
was used to assign exposure scores
for this quantitative
variable.
The results
of this nested
case-control
study
indicates
that high serum
levels of DHEA
up to 14 years prior
to
diagnosis
are associated
with a decreased
risk of developing
gastric
cancer.
To our knowledge,
there
are no
other
studies
that
have
examined
the relationship
of
prediagnostic
serum
values
of DHEA
or DHEAS
to the
risk of developing
gastric
cancer.
The results are consistent with
case-control
studies
which
demonstrated
that
except
for individuals
with very advanced
gastric
cancer,
patients
with gastric cancer
have lower
urinary
excretion
ofthese
steroidsthan
controls
(15-17).
Although
there
are no reports
of the protective
actions of DHEA
in animal
models
of gastric cancer,
DHEA
has an impressive
spectrum
of activity
in tissue
culture
systems
and in animal
models
of both cancer
and atherosclerosis
(25-29).
In any cohort
study
the possible
effects
of a bias
introduced
by different
losses to follow-up
among
cases
and controls
should
be considered.
However,
the Washington
County
population
has a low rate of outmigration
(1%/year),
and case ascertainment
is estimated
to be
reasonably
complete.
Therefore
this is unlikely
to affect
the observed
results
of this study.
While
the observed
association
may
not be a direct
protective
effect
by
DHEA
or DHEAS
but rather
an effect
of some
other
factor(s)
for which
it is a surrogate
marker,
animal
experimental
studies
suggest
a direct
protective
effect
by
DHEA against cancer.
Despite
the small size ofthis
study,
we have found
a strong,
inverse,
dose-related
association
between
serum
levels of DHEA and the risk of developing
stomach
cancer.
Results
References
Measurement
of Serum
DHEA
DHEAS
were
assayed
with
assay kits (Wien
Laboratories,
to the directions
ofthe
methyletie
chloride
DHEA.
These
kits
to reference
DHEAS
and
DHEAS.
manufacturer,
except
(1:1) was used for
were selected
based
methods
DHEA
and
radioimmunoNJ) according
commercial
Succasunna,
that hexane:
the extraction
on comparisons
for the measurement
of DHEA
of
and
(19).
Statistical
Analysis.
Because
of their
skewed
distnibutions, DHEA and DHEAS
levels were analyzed
as ordered
categories
or with nonparametnic
methods.
The case and
control
hormone
distributions
were compared
using the
median
case-control
difference.
To account
for the
matched
study design,
the Wilcoxon
sign rank test cornparing the case to the average
of its four matched
controls
was used (20). Tertile
cut-off
points
for DHEA and DHEAS
were
based
on the distributions
of the controls.
Using
the lowest
tertile
as the referent
category,
conditional
logistic
regression
was used to obtain
maximum
likelihood estimates
of the matched
odds ratios and approximate
On
95%
average,
confidence
the
cases
intervals
were
(21). The strength
61
years
old
when
of the
they
donated
serum
and over 67 years old when
they were
diagnosed,
with a median
time to diagnosis
of 6.5 years.
Comparisons
of cases
and controls
on the matching
criteria
and other
characteristics
show
no statistically
significant
differences.
More
cases than controls
had a
history
of ever smoking
(matched
odds ratio = 2.6; 95%
confidence
interval
=
0.5 to 14.3). The ratio of males to
females
was 3:1.
Median
DHEA
levels
were
38%
lower
in the cases
than controls
(P = 0.09)
(Table
1). DHEAS
levels
were
1 7.5%
(P = 0.77) lower among
cases than controls.
Smok-
1. Muir,
Cancer
Research
C., Waterhouse,
J., Mack,
Incidence
in Five Continents
on Cancer,
1987.
2. American
Cancer
Society.
Cancer
American
Cancer
Society,
1992.
T.,
Powell,
V. Lyon:
Facts
and
I., and Whelan,
S. (eds.).
International
Agency
for
Figures,
3. Ries,
L. A. C., Hankey,
B. F., Miller,
B. A.,
Edwards,
B. K. (eds.).
Cancer
Statistics
Review,
MD: NCI, NIH Pub. No. 91-2789,
1991.
4. Parker,
L. N. Adrenal
Diego:
Academic
Press,
Androgens
1989.
in Clinical
1992.
Hartman,
1973-1988.
Medicine,
New
A.
pp.
York:
M., and
Bethesda,
31-57.
San
5. Migeon,
C. J., Keller,
A. R., Lawrence,
B., and Shepard,
T. H. Dehydroepiandrosterone
and androsterone
levels
in human
plasma.
Effect of
age and sex; day-to-day
and diurnal
variations.
I. Clin. Endocrinol.
Metab.,
17: 1051-1062,
1957.
Downloaded from cebp.aacrjournals.org on June 17, 2017. © 1993 American Association for Cancer Research.
Cancer
6. Orentreich,
N., Brind,
J. L., Rizer,
R. I.., and Vogelman,
J. H. Age
changes
and
sex differences
in serum
DHEA-sulfate
concentrations
throughout
adulthood.
I. Clin.
Endocrinol.
Metab.,
59: 551-555,
1984.
7. Bulbrook,
R. D., Hayward,
I. L., and
urinary
androgen
and corticoid
Lancet, 2: 395-398,
1971.
8.
Brownsey,
excretion
B., Cameron,
E. H.
A. P. M., and Campbell,
H. Plasma
in patients
131-137,
and
with
1972.
benign
Spicer,
D.,
and
C. C. Relation
subsequent
Griffiths,
K., Cleave,
10.
Zumoff,
Fukishima,
DHEA
Cancer
the
B., Levin,
D.
malignant
breast
J., Rosenfeld,
K. Abnormal
and DHEA-sulfate
disease.
Res., 41:
3360-3363,
Eur.
19.
levels
J. cancer,
8:
C.
B.,
with
mean
Higertova,
J., and
in cancer.
plasma
primary
M.,
Endokri-
Strain,
C.
W.,
concentrations
operable
breast
of
cancer.
1981.
E., Khaw,
K-T.,
dehydroepiandrosterone
Gordon,
Z.,
R. S., Markham,
24-hr
in women
and
Yen,
sulfate,
disease. N. EngI. J. Med.,
12.
Forrest,
sulfate
315:
Bush,
1519-1524,
T.
Comstock,
G. W. Relationship
and dehydroepiandrosterone
opausal breast cancer. Cancer
S. S. C. A prospective
mortality,
and
study
cardiovascular
Holtzclaw,
Helzlsouer,
K.
W.
D.,
and
of dehydroepiandrosterone
and enzymatic
analysis.
SAS Institute,
6.03. Cary,
Inc.
Gordon,
Procedures
NC: SAS Institute,
Breslow,
N. E., and
Vol.
1. Lyon:
International
22.
Statistics
and
Corporation.
Day,
N. E. Statistical
1705-1709,
24.
Moertel,
Medicine,
The
14.
Herrington,
D. M., Gordon,
P. 0., and
terone
nostic
15.
(DHEA)
and DHEA
coronary
angiography.
Kodama,
M.,
Kodama,
C. B., Achuff,
S. C., Trejo, J. F., Weisman,
1. A. Plasma
dehydroepiandros-
sulfate (DHEAS)
in patients
undergoing
diagJ. Am. CoIl. Cardiol.,
16: 862-870,
1990.
1.,
Kobayashi,
S., Kasugi,
1. W.,
and Suga, S. Hormonal
status of gastric cancer. II. Abnormal
of urinary steroids
in gastric cancer patients.
Nutr. Cancer,
Takagi,
H.,
constitution
9: 251-263,
1987.
16. Pankov,
Sex hormone
ingr.),
A. K., Smiller, R. R., Orlovskia,
excretion
in stomach
cancer
24: 46-51,
1978.
L. A., and Tiutiunova,
patients.
Vopr. Onkol.
A. M.
(Len-
Seattle:
Philadelphia:
zyme
26: 355-382,
Regul.,
J. Clin.
and
Research,
1980.
Cytel
Software
SERC and Cytel,
1990.
E. Cigarette
smoking and
women.
N. EngI. J. Med.,
agents.
C.
B.,
Invest.,
Bush,
82:
and E. Frei, III (eds.),
Lea and Febiger,
1982.
and Talalay,
P. Modulation
by dehydroepiandrosterone.
of growth,
Adv.
En-
1987.
26. Schwartz,
A. C., Whitcomb,
Pashko,
L. L. Dehydroepiandrosterone
injury.
Pearson,
Release
in Cancer
on Cancer,
In: J. F. Holland
stomach.
pp. 1760-1774.
C. B., Shantz,
L. M.,
and carcinogenesis
Gordon,
levels
1988.
C. W.
25. Gordon,
differentiation
27.
Computers,
Corporation
Manual.
atherosclerosis
by administration
the hypercholesterolemic
New
H. F., Kwiterovich,
Methods
for Research
Research
Reference
bladder
K. J., and Comstock,
C. W. Serum
levels
and its sulfate
and the risk of developing
Res., 51: 1366-1369,
1991.
for Personal
23. Khaw, K.T., Tazuke, S., and Barrett-Connor,
levels of adrenal androgens
in postmenopausal
38:
of serum
1988.
Agency
ofchemopreventive
cancer
and dehydroDelo.,
2: 105-
B. Measurement
Guide
Epidemiology
EGRET
C.
Inc.,
13. Gordon,
C. B., Helzlsouer,
of dehydroepiandrosterone
cancer.
17-ketosteroids
cancer.
Vrach.
35
sulfate: a comparison
of radioimmunoassay
Steroids, 54: 355-371,
1989.
21.
Cancer
1986.
J., Miller,
S. R., and
of serum levels of dehydroepiandrosterone
sulfate to the risk of developing
postmenRes., 50: 3859-3862,
1990.
L.,
I. Urine
levels of neutral
in patients
with stomach
& Prevention
106, 1971.
20.
1 1 . Barrett-Connor,
of
E. N.,
dehydroepiandrosterone
9. Sonka,
J., Vitkova,
M., Gregorova,
I., Tomsova,
Stas, J. Plasma and urinary
dehydroepiandrosterone
nologie,
62: 61-68,
1973.
and
cancer.
Romanishin,
epiandrosterone
Biomarkers
18. Helzlsouer,
K. J., Comstock,
C. W., and Morris,
J. S. Selenium,
Iycopene,
a-tocopherol,
a-carotene,
retinol,
and subsequent
bladder
cancer. Cancer Res., 49: 6144-6148,
1989.
between
breast
17.
Epidemiology,
Adv.
D.
Nyce,
J. W., Lewbart,
and structural
analogs:
Cancer Res., 51: 391-423,
J. M.,
E.,
and
Weisman,
H.
F.
of dehydroepiandrosterone.
Zealand
white rabbit with
712-720,
M. L., and
a new
class
1988.
Reduction
of
A study in
aortic intimal
1988.
28. Shantz, 1. M., Talalay, P., and Gordon,
C. B. Mechanism
of inhibition
of growth of 313-Li
fibroblasts
and their differentiation
to adipocytes
by
dehydroepiandrosterone
and related
steroids:
role of glucose-6-phosphate dehydrogenase.
Proc. NatI. Acad. Sci. USA, 80: 3852-3856,
1989.
29. Garcea,
Vannini,
M.
R., Daino,
C., Pascale,
I.., Frassetto,
R., Lenzerini,
S., Cozzolino,
L., Simile,
M.
P., Ruggiu,
M., Puddu,
M. E.,
M., and
Feo, F. Reversal by ribo- and deoxyribonucleosides
of dehydroepiandrosterone-induced
inhibition
of enzyme
altered foci in the liver of rats
subjected
to the initiation-selection
process of experimental
carcinogenesis. Carcinogenesis
(Lond.),
9: 931-938,
1988.
Downloaded from cebp.aacrjournals.org on June 17, 2017. © 1993 American Association for Cancer Research.
Serum levels of dehydroepiandrosterone and
dehydroepiandrosterone sulfate and the risk of developing
gastric cancer.
G B Gordon, K J Helzlsouer, A J Alberg, et al.
Cancer Epidemiol Biomarkers Prev 1993;2:33-35.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cebp.aacrjournals.org/content/2/1/33
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cebp.aacrjournals.org on June 17, 2017. © 1993 American Association for Cancer Research.