A Randomized Double-Blinded Placebo-Controlled Trial on the Effect of Dehydroepiandrosterone for 16 Weeks on Ovarian Response Markers in Women with Primary Ovarian Insufficiency

Article (PDF Available)inThe Journal of Clinical Endocrinology and Metabolism 98(1) · November 2012with43 Reads
DOI: 10.1210/jc.2012-3071 · Source: PubMed
Abstract
Context:Preliminary reports have shown encouraging effects of dehydroepiandrosterone (DHEA) in women with poor ovarian reserve undergoing assisted reproduction and primary ovarian insufficiency (POI), although data from randomized controlled trials are limited. The present study assesses the effect of DHEA on ovarian response markers in women with POI.Objective:The objective of the study was to evaluate whether DHEA for 16 wk would improve ovarian response markers in women with POI.Design:This was a randomized, double-blinded, placebo-controlled study.Setting:The study was conducted at a tertiary reproductive unit.Patients:Twenty-two women with unexplained POI participated in the study.Interventions:Eligible subjects were randomized into the DHEA group (n = 10), who received DHEA (LiveWell, 25 mg three times a day), or the placebo group (n = 12), who received placebo for 16 wk according to a computer-generated randomization list. Ovarian response markers included serum anti-Mullerian hormone (AMH), FSH levels, and antral follicle count (AFC) as well as follicles of 10 mm or greater in diameter, and hormonal profiles were measured at 4-wk intervals until 4 wk after completion of treatment. Any returns of menses and side effects from treatment were recorded.Main Outcome Measures:The primary outcome was serum AMH level.Results:No significant change in serum AMH and FSH levels had been detected throughout the study. AFC and ovarian volume were significantly higher at wk 12 and 20, respectively, in the DHEA group. Significantly more women having at least one follicle of 10 mm or greater at wk 12, 16, and 20 were found in the DHEA group. Serum testosterone and DHEA sulfate levels along with higher estradiol levels were significantly higher in the DHEA group.Conclusion:This randomized, double-blinded, placebo-controlled trial found higher AFC and ovarian volume at wk 12 and 20, respectively, in the DHEA group, although there were no significant changes in serum AMH and FSH levels. Further trials using a longer duration of DHEA should be considered to evaluate the long-term effect of DHEA in women with POI.

Figures

Figure
Figure
Figure
Figure

Full-text (PDF)

Available from: Tracy Wing Yee Yeung, Dec 04, 2014
ARandomizedDouble-BlindedPlacebo-Controlled
Trial on the Effect of Dehydroepiandrosterone for 16
Weeks on Ovarian Response Markers in Women with
Primary Ovarian Insufficiency
Tracy Wing Yee Yeung, Raymond Hang Wun Li, Vivian Chi Yan Lee,
Pak Chung Ho, and Ernest Hung Yu Ng
Department of Obstetrics and Gynecology, The University of Hong Kong, Hong Kong Special
Administrative Region, People’s Republic of China
Context: Preliminary reports have shown encouraging effects of dehydroepiandrosterone (DHEA)
in women with poor ovarian reserve undergoing assisted reproduction and primary ovarian in-
sufficiency (POI), although data from randomized controlled trials are limited. The present study
assesses the effect of DHEA on ovarian response markers in women with POI.
Objective: The objective of the study was to evaluate whether DHEA for 16 wk would improve
ovarian response markers in women with POI.
Design: This was a randomized, double-blinded, placebo-controlled study.
Setting: The study was conducted at a tertiary reproductive unit.
Patients: Twenty-two women with unexplained POI participated in the study.
Interventions: Eligible subjects were randomized into the DHEA group (n !10), who received
DHEA (LiveWell, 25 mg three times a day), or the placebo group (n !12), who received placebo for
16 wk according to a computer-generated randomization list. Ovarian response markers included
serum anti-Mullerian hormone (AMH), FSH levels, and antral follicle count (AFC) as well as follicles
of 10 mm or greater in diameter, and hormonal profiles were measured at 4-wk intervals until 4
wk after completion of treatment. Any returns of menses and side effects from treatment were
recorded.
Main Outcome Measures: The primary outcome was serum AMH level.
Results: No significant change in serum AMH and FSH levels had been detected throughout the
study. AFC and ovarian volume were significantly higher at wk 12 and 20, respectively, in the DHEA
group. Significantly more women having at least one follicle of 10 mm or greater at wk 12, 16, and
20 were found in the DHEA group. Serum testosterone and DHEA sulfate levels along with higher
estradiol levels were significantly higher in the DHEA group.
Conclusion: This randomized, double-blinded, placebo-controlled trial found higher AFC and ovar-
ian volume at wk 12 and 20, respectively, in the DHEA group, although there were no significant
changes in serum AMH and FSH levels. Further trials using a longer duration of DHEA should be
considered to evaluate the long-term effect of DHEA in women with POI. (J Clin Endocrinol Metab
98: 00000000, 2013)
ISSN Print 0021-972X ISSN Online 1945-7197
Printed in U.S.A.
Copyright © 2013 by The Endocrine Society
doi: 10.1210/jc.2012-3071 Received August 13, 2012. Accepted October 11, 2012.
Abbreviations: AFC, Antral follicle count; AMH, anti-Mullerian hormone; BMD, bone min-
eral density; CV, coefficient of variation; DHEA, dehydroepiandrosterone; DHEA-S, DHEA-
sulfate; E2, estradiol; POI, primary ovarian insufficiency.
ORIGINAL ARTICLE
Endocrine Research
J Clin Endocrinol Metab, January 2013, 98(1):00000000 jcem.endojournals.org 1
J Clin Endocrin Metab. First published ahead of print November 8, 2012 as doi:10.1210/jc.2012-3071
Copyright (C) 2012 by The Endocrine Society
Primary ovarian insufficiency (POI) is characterized by the
triad of amenorrhea for at least 4 months, sex steroid
deficiency, and two recordings of serum FSH in the meno-
pausal range in a woman aged younger than 40 yr (1). Pre-
viously used terms including premature menopause or pre-
mature ovarian failure should be avoided because the term
premature arbitrarily confined the disorders in which menses
stop before the age of 40 yr (2); and as supposed to meno-
pause in which the cessation of ovarian function is perma-
nent, up to 50% of women with POI has varying and un-
predictable ovarian function and about 5–10% of women
conceive spontaneously after the diagnosis (3). It affects
about 1% of women aged younger than 40 yr, 0.1% of under
the age of 30 yr, and 0.01% under the age of 20 yr (4).
Women with POI not only experience estrogen deficiency
but also suffer from the loss of ovarian androgens due to
atrophy of ovarian cortex (5). They have to face major dif-
ficulties with fertility and conception (6) as well as long-term
health risks associated with estrogen deficiency. Women are
counseled to have long-term hormonal replacement and re-
sort to oocyte/embryo donation, adoption, or remain child-
less owing to the very limited and unpredictable ovarian
function.
Dehydroepiandrosterone (DHEA) is an endogenous
steroid produced mainly in the zona reticularis of adrenal
cortex and ovarian theca cells in women. It has been im-
plicated in ovarian follicular steroidogenesis and is be-
lieved to increase follicular IGF-I, which promotes follicu-
logenesis (7) and potentiates gonadotropin effects (8).
Previous studies have shown preliminary success in using
DHEA to improve ovarian response to gonadotrophins,
leading to spontaneous pregnancies or improved out-
comes in assisted reproductive techniques (8–12).
Mamas and Mamas (39) published their preliminary
results that after DHEA supplementation, women with
POI had reduced serum FSH and successful live births
either achieved spontaneously or after mild ovarian stim-
ulation and intrauterine tuboperitoneal insemination. The
encouraging effects of DHEA in improving ovarian func-
tion, even in women with POI, prompted some centers to
use DHEA in routine management of these women. How-
ever, such practice was not supported by evidence from
randomized controlled trials. The present study aims at
filling this gap by assessing the effect of DHEA on ovarian
reserve. Although endocrine tests like anti-Mullerian hor-
mone (AMH), FSH, and ultrasound markers like antral
follicle count (AFC) and ovarian volume have been widely
used and regarded as ovarian reserve markers, none of
these tests can directly measure the number of nongrowing
primordial follicles in ovaries, which constitute the genu-
ine ovarian reserve. These markers should therefore be
more appropriately referred as ovarian response markers.
Materials and Methods
Study design and protocol
Women who presented with secondary amenorrhea and were
diagnosed with primary POI were recruited from the Depart-
ment of Obstetrics and Gynaecology, University of Hong Kong.
Inclusion criteria included the following: 1) age younger than
40 yr; 2) secondary amenorrhea; 3) serum FSH level greater than
30 IU/liter on two occasions of at least 6 wk apart; 4) normal
karyotype of 46XX; and 5) negative FMR1 gene mutation. Pa-
tients were excluded if they had any of the following: 1) a history
of ovarian cystectomy or oophorectomy; 2) having received cy-
totoxic chemotherapy; 3) having received pelvic irradiation; 4) a
diagnosis of autoimmune disease; 5) a history of taking testos-
terone or DHEA supplement; or 6) pregnancy. The study was
approved by the Institutional Review Board of the University of
Hong Kong/Hospital Authority Hong Kong West Cluster and
was registered under Hong Kong Clinical Trial Center (HKCTR-
1148). All women were fully counseled and written consents
were obtained before their participation.
Women who were taking hormonal replacement therapy at the
time of recruitment were asked to withhold them for a minimum of
2months.Baselineinvestigationsforovarianresponsemarkersin-
cluding serum AMH, FSH levels, AFC, and ovarian volume were
performed. Serum estradiol (E2), testosterone, DHEA-sulfate
(DHEA-S), SHBG, IGF-I, and liver enzymes were also measured.
Assignment and masking
Women were randomized according to a computer-generated
randomization list generated by a research nurse not involved in
the subjects’ clinical management. The hospital pharmacy pack-
aged the DHEA and placebo capsules according to the random-
ization list and labeled the drug packs with subject numbers only.
The placebo capsules were identical in appearance with the
DHEA capsules. Physicians, research nurses involved in the trial,
and the study subjects were all blinded to the assignment.
Treatment and monitoring
Each woman received 16 wk of either DHEA (GNC LiveWell,
Pittsburgh, PA) capsule at 25 mg three times a day or matching
placebo capsules. All women were assessed before the start of treat-
ment and were followed up at 4-wk intervals until 4 wk after stop-
ping treatment. Serial transvaginal scans were performed by T.Y.
using a 7-MHz vaginal probe (Voluson 730; GE Healthcare, Mad-
ison, WI) to determine AFC (2–9 mm) in both ovaries and to mea-
sure the volume of the ovaries. The length, height, and width of each
ovary were measured in the sagittal and coronal planes, and then the
ovarian volume was obtained using a formula for the volume of an
ellipsoid (
!
/6 "length "height "width). The total ovarian volume
was the sum of the right and left ovarian volumes. The intraobserver
coefficient of variation (CV) for AFC and ovarian volume was 7%
and less than 10%, respectively.
Blood was collected for serum AMH, FSH, E2, testosterone,
DHEA-S, SHBG, IGF-I, and liver function. Serum samples were
stored at #20 C until assayed as a whole batch. Any return of
menses or side effects were recorded. Serum AMH was measured
using AMH Gen II ELISA (Beckman Coulter, Fullerton, CA);
IGF-I was measured using Quantikine ELISA human IGF-I
(R&D Systems, Minneapolis, MN), and E2, testosterone,
DHEA-S, and SHBG were measured using Beckman Coulter Ac-
cess 2 Immunoassay system.
2Yeung et al. RCT: Effect of DHEA on Ovarian Response in POI J Clin Endocrinol Metab, January 2013, 98(1):0000 0000
The intra-assay CVs were 3.4 –5.4% for AMH, 3.5–4.3% for
IGF-I, less than 12–21% for E2, 1.67–3.93% for testosterone,
1.6 8.3% for DHEA-S and 4.5– 4.8% for SHBG. The inter-
assay CV were 4.0–5.6% for AMH, 7.5– 8.1% for IGF-I, 4.22–
7.08% for testosterone, 3.7–11.3% for DHEA-S and 5.2–5.5%
for SHBG. Detection range was 0.08–22.5 ng/ml for AMH,
0.007– 6 ng/ml for IGF-I, 0.1–16 ng/ml for testosterone, 2–1000
"
g/dl for DHEA-S and 0.33–200 nmol/liter for SHBG.
Statistical analysis
The primary outcome measure was the serum AMH level. We
have chosen serum AMH as our primary outcome measure be-
cause it is produced by granulosa cells of small growing follicles,
whose number and thus the serum concentration should be re-
flective of the ovarian reserve (13). Secondary outcome measures
included serum FSH level, AFC, total ovarian volume, ovarian
follicles of 10 mm or greater and hormonal profiles including
serum E2, testosterone, DHEA-S, SHBG, and IGF-I.
Based on our previous data in patients with newly diagnosed
POI, the mean serum AMH concentration was 2.65 pmol/liter
(0.37 ng/ml) with a SD of 1.8 pmol/liter (0.25 ng/ml) (14). As-
suming a doubling of mean serum AMH levels in the DHEA
group being clinically significant, nine women would be required
in each arm to give a test of significance of 0.05 and a power of
0.8. 22 subjects were recruited to allow for possible dropouts.
Statistical comparisons were carried out with the intention to
treat and per-protocol approaches by Student’s ttest, Mann-Whit-
ney Utest,
#
2
test, and Fisher’s exact test where appropriate using
the Statistical Program for Social Sciences (version 19.0; SPSS Inc.,
Chicago, IL). A two-sided P$0.05 was taken as statistically
significant.
Results
Participant flow
Between June 2010 and June 2011, 63 women with POI
were screened, and 29 fulfilled the selection criteria and
were counseled to participate. Twenty-two women were
recruited and randomized. One woman from the DHEA
group at wk 4 and one woman from the placebo group at
wk 12 withdrew their consents due to personal reasons.
FIG. 1. Consolidated Standards of Reporting Trials 2010 flow diagram.
J Clin Endocrinol Metab, January 2013, 98(1):0000 0000 jcem.endojournals.org 3
One woman from the DHEA group was excluded after 4
wk because she was found to be 8 wk pregnant, i.e. she
conceived before the administration of DHEA (Fig. 1).
Baseline characteristics
No significant difference was found between the two
groups with regard to age, body mass index, duration of
POI [defined as the time between the confirmed diagnosis
(history of amenorrhea $4 months and two FSH results at
menopausal range under the age of 40 yr) and the time of
the baseline blood tests and ultrasound scans] and serum
FSH level at diagnosis (Table 1). A comparable percentage
of the women had received hormone replacement therapy,
which had been withheld for a minimum of 2 months
before the start of the study as a washout period. Baseline
hormonal profiles, AFC, and total ovarian volume were
comparable between the groups, and no subject had fol-
licles 10 mm or larger at the start of study.
Endocrine profile
Median serum AMH levels were undetectable in both
groups throughout the 20-wk study period. It was only
detectable at very low levels in one woman in the DHEA
group (up to 0.11 ng/ml) and two women in the placebo
group (up to 0.14 ng/ml).
No significant difference was detected in the serial se-
rum FSH levels between the two groups (Fig. 2). The me-
dian serum E2 levels were significantly higher in the
DHEA group at wk 12 at 114 pmol/liter compared with 73
pmol/liter in the control group. The testosterone and
DHEA-S levels were significantly higher in the DHEA
group from wk 4 to wk 16. No significant difference was
detected in the SHBG and IGF-I levels between the two
groups.
Ultrasound findings
Median AFC was significantly higher in the DHEA
group than that in the control group at wk 12 [2.00 (range
1–5) vs. 1.00 (range 0–2), respectively, P!0.034]. Me-
dian total ovarian volume was significantly higher in the
DHEA group than that in the control group at wk 20 (3.79
vs. 2.02 cm
3
, respectively, P!0.033) (Fig. 3). Median
percentage changes of total ovarian volume followed a
similar pattern with an increase starting at wk 16 and
became statistically significant at wk 20 [224% (range
109–529%) vs. 47% (range #56 to 147%), P!0.006]
(Fig. 4).
There was no follicle 10 mm or larger in any subject at
baseline and 4 wk after the intervention. Overall, a higher
proportion of women in the DHEA group had follicles 10
mm or larger, ranging from 11 to 33% from wk 8 onward
(Fig. 5).
Return of menses
Twenty-five percent of the women in the placebo group
and 11.1% in the DHEA group experienced irregular re-
turn of menses, and there was no significant difference
between the groups.
Side effects
No major adverse effects were reported during the study
period. Up to 22% of subjects in the DHEA group com-
TABLE 1. Baseline characteristics, hormonal profile, and ultrasound findings of a randomized comparison between
the DHEA and placebo groups
DHEA group (n !9) Placebo group (n !12) Pvalue
Age (yr) 35.9 %3.26 33.4 %4.74 0.196
a
BMI (kg/m
2
) 21.4 %3.34 21.1 %4.08 0.961
a
Duration of POI (months) 30 (2–81) 43 (6–132) 0.477
b
FSH at diagnosis (IU/liter) 79.2 %35.1 81.8 %33.7 0.783
a
Previous use of HRT 4/9 (44.0%) 11/12 (91.7%) 0.46
c
Baseline serum levels
AMH (ng/ml) 0 (0) 0 (0 0.13) 0.209
b
FSH (IU/liter) 101.9 %49.8 91.6 %32.6 0.678
a
Estradiol (pmol/liter) 89 (73–268) 73 (73–109) 0.320
b
Testosterone (ng/ml) 0.27 %0.12 0.56 %0.37 0.694
a
DHEA-S (
"
g/dl) 160.0 %68.7 157 %107 0.967
a
SHBG (nmol/liter) 47.3 %16.8 50.6 %26.0 0.708
a
IGF-I (ng/ml) 145.4 %56.6 150.0 %60 0.611
a
Baseline USG findings AFC 0 (0–2) 0 (0–2) 0.327
b
Total ovarian volume (cm
3
) 1.50 (0–2.2) 1.51 (0.6–2.9) 0.413
b
Follicles $10 mm 0 0 1.0
c
Data expressed as mean %SD or median (range) or number (percentage) as appropriate. P$0.05 was considered as statistically significant. BMI,
Body mass index; HRT, hormone replacement therapy; USG, ultrasound.
a
Student’s ttest.
b
Mann-Whitney Utest.
c
#
2
test.
4Yeung et al. RCT: Effect of DHEA on Ovarian Response in POI J Clin Endocrinol Metab, January 2013, 98(1):0000 0000
plained of increased acne compared with 8.3% in the pla-
cebo group, and up to 22% in DHEA group and 16.7% in
placebo group had a nonpersisting mild elevation of liver
enzymes. There was no significant difference between two
groups.
Subgroup analyses
Subgroup analyses were performed within the DHEA
group (n !9) to identify the characteristics among those
who responded to DHEA with a return of menses and/or
follicles of 10 mm or larger.
Compared with women who remained amenorrheic
throughout the study period (n !6), women who had a
return of menses (n !3) had a lower serum FSH level at
initial diagnosis (33.3 %9.61 IU/liter vs. 110 %39.5 IU/
liter, P!0.041). They also had significantly lower serum
FSH at wk 8 (59.9 %34.6 IU/liter vs. 117.3 vs. 25.9 IU/
liter, P!0.036) and wk 12 (37.2 %7.44 IU/liter vs. 109.4
FIG. 2. Box-and-whisker plots of hormone concentrations for women randomized into groups using 16 wk of DHEA (shaded box) and placebo
(open box). Boxes indicate 25th and 75th percentiles, with the horizontal line representing the median values. Whiskers span the range between
the fifth and the 95th percentiles of the data. The x-axis represents the time of the blood taking after the DHEA/placebo use. Statistically
significant differences are defined as P$0.05 and are indicated by an asterisk.
J Clin Endocrinol Metab, January 2013, 98(1):0000 0000 jcem.endojournals.org 5
vs. 41.8 IU/liter, P!0.028). In addition, serum estradiol
level was significantly higher at wk 12 [312 pmol/liter
(range 233–390) vs. 89 pmol/liter (range 79–129), P!
0.025]. There were no significant differences in other base-
line characteristics, other hormonal parameters, AFC, and
ovarian volume in the women who responded or not to
DHEA.
All parameters were comparable between women who
had at least one follicle larger than 10 mm (n !4) and
those who had not (n !5).
All the above analyses were repeated by per protocol
approach and the results (not shown) were essentially the
same.
Discussion
To the best of our knowledge, this is the first randomized,
double-blinded, placebo-controlled study to detect any
improvement in ovarian response markers in women with
POI after 16 wk of DHEA and monitor the change in these
markers 4 wk after stopping DHEA. We could not find any
significant improvement in serum AMH and FSH levels
throughout the whole study period. AFC and ovarian vol-
ume were significantly higher at wk 12 and 20, respec-
tively, in the DHEA group. More women having at least
one follicle 10 mm or larger at wk 12, 16, and 20 were
found in the DHEA group. Significantly higher serum tes-
tosterone and DHEA-S levels were achieved in the DHEA
group from wk 4 to wk 16, confirming the compliance of
subjects in taking the drugs.
A recent systematic review and meta-analysis demon-
strated that total testosterone concentrations are reduced
in women with spontaneous POI compared with the age-
matched controls (15). It has previously been shown that
accumulation of androgens in the micro milieu of the pri-
mate ovary plays a critical role in early follicular devel-
opment and granulosa cell proliferation (16, 17). Andro-
gens promote recruitment and initiation of primordial
FIG. 4. Percentage differences in ovarian volume compared with
baseline presented at 4-wk intervals. Statistically significant differences
are defined as P$0.05 and are indicated by an asterisk.
FIG. 5. Percentage of women having at least one follicle greater than
10 mm during the 20 wk.
FIG. 3. Box-and-whisker plots of AFC (left panel) and total ovarian volume (right panel) for women randomized into groups using 16 wk of DHEA
(shaded box) and placebo (open box). Boxes indicate 25th and 75th percentiles, with the horizontal line representing the median values. Whiskers
span the range between the fifth and the 95th percentiles of the data. The x-axis represents the time of the ultrasound monitoring after DHEA/
placebo use. Statistically significant differences are defined as P$0.05 and are indicated by an asterisk.
6Yeung et al. RCT: Effect of DHEA on Ovarian Response in POI J Clin Endocrinol Metab, January 2013, 98(1):0000 0000
follicle growth and induce significant increase in the num-
ber of primary, preantral, and antral follicles through the
up-regulation of IGF-I (7, 17); up-regulate FSH receptor
expression in granulosa cells to potentiate the effect of
FSH (7, 16, 18); and exert paracrine regulation on follic-
ular maturation and reduce follicular atresia (7, 18). At the
same time, lack of androgen has been shown to reduce the
number of antral follicles and ovulated oocytes (19) as well
as leading to an accelerated rate of follicular atresia (20).
With the presumed important roles of androgens in
folliculogenesis and reducing follicular atresia, attempts
to restore androgen levels in women with POI in the hope
of improving ovarian functions seem biologically plausi-
ble. There have been increasing publications on improved
treatment outcomes after DHEA supplementation in poor
responders undergoing in vitro fertilization as well as re-
ports on improved serum AMH in women with dimin-
ished ovarian reserve (21) and pregnancies among women
with POI (22). It becomes imperative to further investigate
the potential benefits of DHEA in women with POI.
The baseline AMH level in the present study were un-
expectedly lower than that of the women in our previous
study (14) on which we based our statistical analysis. It
may be explained by the fact that women in the previous
study were newly diagnosed with POI during their inves-
tigations for secondary amenorrhea, whereas women in
the present study have already been diagnosed with POI
for a much longer time (median 30 months, range 2–132
months). AMH is mainly secreted by the granulosa cells of
preantral and early antral follicles, which are constantly
being recruited from a primordial follicle pool (23). In
women with POI, the primordial pools are already se-
verely depleted at the time of diagnosis. Ongoing depletion
of the remaining follicular pool in women diagnosed with
POI for years may further reduce the number of preantral
and early antral follicles leading to undetectable baseline
AMH levels.
In contrast to the reported improvement of serum
AMH in women with poor ovarian reserve (21), our study
was unable to detect any significant improvement of se-
rum AMH levels after 16 wk of DHEA, and serum AMH
remained largely undetectable in both groups. It has been
well accepted that women are born with all the primordial
follicles and additional primordial follicles cannot be pro-
duced. Even with 16 wk of DHEA, the proposed actions
of androgens are not to generate new primordial follicles
de novo but only to increase the proportion of follicles in
the growing pool by increasing the recruitment and initi-
ation of folliculogenesis and reduce those undergoing atre-
sia. The pool of preantral and early antral follicles may still
be too small to secret AMH in a detectable amount using
the current kits despite an improvement after DHEA.
Moreover, we would not be able to detect a doubling of
AMH because the baseline level is unexpectedly undetect-
able, and this would be one of the weaknesses of our study.
However, substantial ovarian function and spontaneous
conception had been reported in some patients who un-
derwent orthotopic transplantation of ovarian tissue after
gonadotoxic treatment whom serum AMH remained very
low or undetectable (24).
Despite the lack of apparent improvement in serum
AMH and FSH levels in the DHEA group, higher AFC and
total ovarian volume had been observed. Up to 30% of
patients in the DHEA group had follicles 10 mm or larger
after 8 wk of DHEA compared with 17% in the placebo
group. Although the variability of AFC and ovarian vol-
ume measurement is usually greater than that of AMH,
and possibly more prone to observer bias, the double-
blind placebo control design in our study can minimize the
potential observer bias in the ultrasound measurements.
It has been suggested that DHEA is involved in regu-
lating follicular development through increased IGF-I (8,
17). It has been shown to increase the number of primary,
preantral, and antral follicles by increasing the follicular
recruitment and initiation together with reduced follicular
atresia. These may result in an increase in the number of
follicles in the growing pools that are susceptible to stim-
ulation by FSH. It has also been shown to potentiate the
effect of FSH by the up-regulation of FSH receptors (7, 16,
18). These could lead to increased ovarian volume and
proportion of women having at least one follicle 10 mm or
larger as observed in the DHEA group. Although our find-
ings did not show any changes in serum IGF-I after DHEA
use, it is possible that the up-regulation of IGF-I acts lo-
cally in primordial and subsequent stages of follicles and
may not be reflected in serum levels. On the other hand, it
is possible that DHEA supplementation can provide an
increased amount of androsteinedione to the theca cells of
the remaining follicles and thus increase the estradiol pro-
duction by the granulosa cells. This could be a mechanism
to improve follicle growth and responsiveness.
On average, women reaches menopause when the pri-
mordial follicular pool falls below 1000 (25). Although
the depleted follicular pool would unlikely be revived by
DHEA, its supplementation may help in maximizing the
growing pool by increased recruitment and follicular ac-
tivities together with reduced follicular atresia. Approxi-
mately 50% of young women with POI experience inter-
mittent and unpredictable ovulation that can continue for
many years (26–30), and spontaneous pregnancies have
been reported in 5–10% of these women (6). Maximizing
the number of recruited follicles that remain in the grow-
ing pool may provide a better chance of spontaneous ovu-
lation and pregnancy during the period of supplementa-
J Clin Endocrinol Metab, January 2013, 98(1):0000 0000 jcem.endojournals.org 7
tion. Interestingly, women who responded to DHEA
supplementation with return of menses had lower FSH at
first diagnosis. Whether this is suggesting women who
have less depleted ovarian reserve at the start with more
primordial follicles remained for DHEA to work upon
would respond better would be worth exploring. How-
ever, this should be interpreted with caution in view of the
small sample size.
In terms of optimal duration of treatment, because it takes
more than 120 d for the primordial follicles to pass through
the primary follicle stage to reach preantral stage, and it takes
65 d to grow from preantral to antral follicles, a longer du-
ration of DHEA supplementation may confer additional
benefits for increased folliculogenesis. Whether a longer du-
ration and/or higher dose of DHEA could recruit more of
the remaining follicles, building up a bigger growing pool
and secrete detectable amounts of serum AMH would re-
quire further investigations.
Another important aspect in managing women with
POI is the replacement of estrogen. Mean serum E2
levels were significantly improved after 2 months of
DHEA use and reduced vaginal dryness was reported.
Although increased serum E2 level was observed,
whether it is sufficient for bone protection would re-
quire long-term follow-up and an objective assessment
of bone mineral density (BMD) by dual-energy x-ray
absorptiometry. Several randomized control trials had
evaluated the effects of DHEA therapy for 26–104
months on BMD in postmenopausal women. There
were evidences of improved BMD at the lumbar spine
(31–35) and hip (33, 34, 36, 37) along with the evidence
of significantly increased plasma bone alkaline phospha-
tase and osteocalcin (38). Its effect on POI patients would
definitely be worth further exploration.
Traditional hormonal replacement therapy increases
serum estradiol but reduces bioavailable testosterone by
increasing SHBG. In contrast, DHEA supplementation in-
creases serum estradiol and testosterone without affecting
serum SHBG. This may have positive effect in sexual func-
tion, although it is beyond the scope of the present study.
We have excluded women who suffered from POI due
to secondary causes like previous ovarian surgery, cyto-
toxic chemotherapy, and pelvic irradiation, FMR1 pre-
mutation, or Turner syndrome. By including only women
with idiopathic POI, we were able to look at the effect of
DHEA on ovarian function in a more homogeneous group
of patients. The main weakness of our study was the rel-
atively small sample size, which was calculated based on
the assumption that DHEA supplementation would lead
to the doubling of serum AMH levels. Moreover, we did
not check serum progesterone level on a weekly basis to
confirm ovulation and did not use ongoing pregnancy or
live birth rate as the primary outcome.
Successful pregnancies after DHEA supplementation
had been reported in case series (22). However, it should
be noted that spontaneous pregnancies occur in 5–10% of
women with POI without any intervention (6). Indeed,
one of our subjects was found to be 8 wk pregnant at the
second visit and had already conceived before the entry to
the study. Without a properly randomized control group,
any beneficial effect of DHEA would remain speculative.
Conclusion
Although there was no significant improvement in se-
rum AMH and FSH levels, higher AFC and total ovarian
volume were found after 16 wk of DHEA in women with
POI. Given the small number of women with POI encoun-
tered in most units, a larger multicentered randomized
trial addressing the effect of a longer duration of DHEA on
hormonal profile, ovulation rate, and ultimately ongoing
pregnancy or live birth rate would be important for both
patients and clinicians. Before there is further evidence
supporting the benefits of DHEA on women with POI,
their empirical use should not be encouraged outside a
clinical trial context.
Acknowledgments
The clinical trial registration number is HKCTR-1148.
Address all correspondence and requests for reprints to:
Tracy Wing Yee Yeung, (M.B.B.S., M.R.C.O.G., F.H.K.C.O.G.,
F.H.K.A.M., Department of Obstetrics and Gynecology, The
University of Hong Kong, Queen Mary Hospital, Pokfulam
Road, Hong Kong. E-mail: tracyyeungwy@gmail.com.
This work was supported by the Hong Kong Obstetrical and
Gynaecological Trust Fund and The Department of Obstetrics
and Gynaecology, University of Hong Kong.
Disclosure Summary: The authors have nothing to disclose.
References
1. de Moraes-Ruehsen M, Jones GS 1967 Premature ovarian failure.
Fertil Steril 18:440
2. De Vos M, Devroey P, Fauser BCJM 2010 Primary ovarian insuf-
ficiency. Lancet 376:911–921
3. Nelson LM 2009 Primary ovarian insufficiency. N Engl J Med 360:
606614
4. Coulam CB, Adamson SC, Annegers JF 1986 Incidence of premature
ovarian failure. Obstet Gynecol 67:604606
5. Elias AN, Pandian MR, Rojas FJ 1997 Serum levels of androstene-
dione, testosterone and dehydroepiandrosterone sulfate in patients
with premature ovarian failure to age-matched menstruating con-
trols. Gynecol Obstet Invest 43:47–48
6. van Kasteren YM, Schoemaker J 1999 Premature ovarian failure: a
systematic review on therapeutic interventions to restore ovarian
function and achieve pregnancy. Hum Reprod Update 5:483–492
8Yeung et al. RCT: Effect of DHEA on Ovarian Response in POI J Clin Endocrinol Metab, January 2013, 98(1):0000 0000
7. Vendola KA, Zhou J, Adesanya OO, Weil SJ, Bondy CA 1998 An-
drogens stimulate early stages of follicular growth in the primate
ovary. J Clin Invest 101:2622–2629
8. Casson PR, Lindsay MS, Pisarska MD, Carson SA, Buster JE 2000
Dehydroepiandrosterone supplementation augments ovarian stim-
ulation in poor responders: a case series. Hum Reprod 15:2129
2132
9. Barad DH, Gleicher N 2005 Increased oocyte production after treat-
ment with dehydroepiandrosterone. Fertil Steril 84:756.e1–756.e3
10. Barad D, Gleicher N 2006 Effect of dehydroepiandrosterone on
oocyte and embryo yields, embryo grade and cell number in IVF.
Hum Reprod 21:2845–2849
11. Barad D, Brill H, Gleicher N 2007 Update on the use of dehydro-
epiandrosterone supplementation among women with diminished
ovarian function. J Assist Reprod Genet 24:629634
12. Gleicher N, Ryan E, Weghofer A, Blanco-Mejia S, Barad DH 2009
Miscarriage rates after dehydroepiandrosterone (DHEA) supple-
mentation in women with diminished ovarian reserve: a case control
study. Reprod Biol Endocrinol 7:108
13. Anderson RA, Nelson SM, Wallace WH 2012 Measuring anti-Mu¨ l-
lerian hormone for the assessment of ovarian reserve: when and for
whom is it indicated? Maturitas 71:28–33
14. Li HW, Anderson RA, Yeung WS, Ho PC, Ng EH 2011 Evaluation
of serum antimullerian hormone and inhibin B concentrations in the
differential diagnosis of secondary oligoamenorrhea. Fertil Steril
96:774–779
15. Janse F, Tanahatoe SJ, Eijkemans MJ, Fauser BC 2012 Testosterone
concentrations, using different assays, in different types of ovarian
insufficiency: a systematic review and meta-analysis. Hum Reprod
Update 18:405–419
16. Weil S, Vendola K, Zhou J, Bondy CA 1999 Androgen and follicle-
stimulating hormone interactions in primate ovarian follicle devel-
opment. J Clin Endocrinol Metab 84:2951–2956
17. Vendola K, Zhou J, Wang J, Bondy CA 1999 Androgens promote
insulin-like growth factor-I and insulin-like growth factor-I receptor
gene expression in the primate ovary. Hum Reprod 14:2328–2332
18. Hillier SG, Tetsuka M 1997 Role of androgens in follicle maturation
and atresia. Baillieres Clin Obstet Gynaecol 11:249–260
19. Wang H, Andoh K, Hagiwara H, Xiaowei L, Kikuchi N, Abe Y,
Yamada K, Fatima R, Mizunuma H 2001 Effect of adrenal and
ovarian androgens on type 4 follicles unresponsive to FSH in im-
mature mice. Endocrinology 142:49304936
20. Hu YC, Wang PH, Yeh S, Wang RS, Xie C, Xu Q, Zhou X, Chao
HT, Tsai MY, Chang C 2004 Subfertility and defective folliculo-
genesis in female mice lacking androgen receptor. Proc Natl Acad Sci
USA 101:11209–11214
21. Gleicher N, Weghofer A, Barad DH 2010 Improvement in dimin-
ished ovarian reserve after dehydroepiandrosterone supplementa-
tion. Reprod Biomed Online 21:360–365
22. Mamas L, Mamas E 2009 Dehydroepiandrosterone supplementa-
tion in assisted reproduction: rationale and results. Curr Opin Ob-
stet Gynecol 21:306–308
23. Weenen C, Laven JS, Von Bergh AR, Cranfield M, Groome NP,
Visser JA, Kramer P, Fauser BC, Themmen AP 2004 Anti-Mu¨ llerian
hormone expression pattern in the human ovary: potential impli-
cations for initial and cyclic follicle recruitment. Mol Hum Reprod
10:77–83
24. Janse F, Donnez J, Anckaert E, de Jong FH, Fauser BC, Dolmans
MM 2011 Limited value of ovarian function markers following
orthotopic transplantation of ovarian tissue after gonadotoxic treat-
ment. J Clin Endocrinol Metab 96:1136–1144
25. Faddy MJ, Gosden RG 1996 Ovary and ovulation: a model con-
forming the decline in follicle numbers to the age of menopause in
women. Hum Reprod 11:1484–1486
26. Rebar RW, Erickson GF, Yen SS 1982 Idiopathic premature ovarian
failure: clinical and endocrine characteristics. Fertil Steril 37:35
27. Rebar RW, Connolly HV 1990 Clinical features of young women
with hypergonadotropic amenorrhea. Fertil Steril 53:804810
28. Nelson L, Anasti J, Kimzey L, Defensor R, Lipetz K, White B,
Shawker TH, Merino MJ 1994 Development of luteinized graafian
follicles in patients with karyotypically normal spontaneous prema-
ture ovarian failure. J Clin Endocrinol Metab 79:1470–1475
29. Conway GS, Kaltsas G, Patel A, Davies MC, Jacobs HS 1996 Char-
acterization of idiopathic premature ovarian failure. Fertil Steril 65:
337–341
30. Taylor AE, Adams JM, Mulder JE, Martin KA, Sluss PM, Crowley
Jr WF 1996 A randomized, controlled trial of estradiol replacement
therapy in women with hypergonadotropic amenorrhea. J Clin En-
docrinol Metab 81:3615–3621
31. Weiss EP, Shah K, Fontana L, Lambert CP, Holloszy JO, Villareal
DT 2009 Dehydroepiandrosterone replacement therapy in older
adults: 1-and 2-y effects on bone. Am J Clin Nutr 89:1459–1467
32. von Mu¨ hlen D, Laughlin GA, Kritz-Silverstein D, Bergstrom J, Bet-
tencourt R 2008 Effect of dehydroepiandrosterone supplementation
on bone mineral density, bone markers, and body composition in
older adults: the DAWN trial. Osteoporos Int 19:699–707
33. Jankowski CM, Gozansky WS, Kittelson JM, Van Pelt RE, Schwartz
RS, Kohrt WM 2008 Increases in bone mineral density in response
to oral dehydroepiandrosterone replacement in older adults appear
to be mediated by serum estrogens. J Clin Endocrinol Metab 93:
4767–4773
34. Jankowski CM, Gozansky WS, Schwartz RS, Dahl DJ, Kittelson JM,
Scott SM, Van Pelt RE, Kohrt WM 2006 Effects of dehydroepi-
androsterone replacement therapy on bone mineral density in older
adults: a randomized, controlled trial. J Clin Endocrinol Metab 91:
2986–2993
35. Villareal DT, Holloszy JO, Kohrt WM 2000 Effects of DHEA re-
placement on bone mineral density and body composition in elderly
women and men. Clin Endocrinol (Oxf) 53:561–568
36. Baulieu EE, Thomas G, Legrain S, Lahlou N, Roger M, Debuire B,
Faucounau V, Girard L, Hervy MP, Latour F, Leaud MC, Mokrane
A, Pitti-Ferrandi H, Trivalle C, de Lacharrie`re O, Nouveau S, Ra-
koto-Arison B, Souberbielle JC, Raison J, Le Bouc Y, Raynaud A,
Girerd X, Forette F 2000 Dehydroepiandrosterone (DHEA), DHEA
sulfate, and aging: contribution of the DHEAge Study to a socio-
biomedical issue. Proc Natl Acad Sci USA 97:42794284
37. Nair KS, Rizza RA, O’Brien P, Dhatariya K, Short KR, Nehra A,
Vittone JL, Klee GG, Basu A, Basu R, Cobelli C, Toffolo G, Dalla
Man C, Tindall DJ, Melton LJ 3rd, Smith GE, Khosla S, Jensen MD
2006 DHEA in elderly women and DHEA or testosterone in elderly
men. N Engl J Med 355:1647–1659
38. Labrie F, Diamond P, Cusan L, Gomez JL, Be´langer A, Candas B
1997 Effect of 12-month dehydroepiandrosterone replacement
therapy on bone, vagina, and endometrium in postmenopausal
women. J Clin Endocrinol Metab 82:3498–3505
39. Mamas L, Mamas E 2009 Premature ovarian failure and dehydro-
epiandrosterone. Fertil Steril 91:644666
J Clin Endocrinol Metab, January 2013, 98(1):0000 0000 jcem.endojournals.org 9
Show more