The Effect of Intra-Ovarian Androgen Priming on Ovarian Reserve
Parameters in Bologna Poor Responders
Research question: What are the effects of long-term androgen priming in Bologna criteria
poor ovarian reserve (POR) patients undergoing IVF.
Design: This open-label pilot study was conducted at IVFMC, My Duc Hospital, Ho Chi
Minh City, Vietnam. It included consecutive patients aged 18–41 years who fulfilled Bologna
criteria for POR undergoing intra-ovarian androgen priming and ultra-long down-regulation
with a gonadotropin-releasing hormone agonist (GnRHa), followed by stimulation with
gonadotropins and GnRH antagonist co-treatment for IVF (n=30). Priming consisted of lowdose recombinant human chorionic gonadotropin (rhCG) 260 IU every second day plus
letrozole 2.5 mg/day, both for 8 weeks; priming stopped on the first day of ovarian
stimulation. The primary endpoint was serum anti-Müllerian hormone (AMH) concentrations
8 weeks after priming. Secondary endpoints included AFC (2-10 mm), serum hCG,
testosterone, and progesterone levels.
Results: Circulating testosterone, progesterone, estradiol and hCG levels remained
unchanged during androgen priming; the mean AMH level decreased steadily from 0.49
ng/mL (baseline) to 0.33 ng/mL (8 weeks). AFC was 4–5 throughout the study. A mean of
1.10.9 good transferable embryos were obtained; embryo transfer was performed in 15
patients; no ongoing pregnancies were obtained.
Conclusions: Long-term intra-ovarian androgen priming in the current set-up had no
significant effect on hormone levels, AFC and recruitable follicles after ovarian stimulation
in Bologna POR patients undergoing IVF. Further studies are needed to explore other
androgen priming protocols and the clinical value of priming regimens in IVF.
Key words: ovarian reserve; poor response; in vitro fertilization; intra-ovarian androgen
priming; Bologna criteria
In poor ovarian reserve patients who underwent ultra-long GnRHa down-regulation an intraovarian androgen priming regimen for 8 weeks before stimulation had no benefit in terms of
ovarian reserve parameters and steroid levels.
AFC antral follicle count
AMH Anti-Müllerian hormone
AR androgen receptor
ART assisted reproductive technology
BMI body mass index
ESHRE European Society of Human Reproduction and Embryology
ET embryo transfer
FF follicular fluid
FSH follicle-stimulating hormone
FSHR follicle-stimulating hormone receptor
GnRH gonadotropin-releasing hormone
GnRHa gonadotropin-releasing hormone agonist
hCG human chorionic gonadotropin
IVF in vitro fertilization
LH luteinizing hormone
OPU oocyte pick-up
POR poor ovarian response
rhCG recombinant human chorionic gonadotropin
rLH recombinant luteinizing hormone
SD standard deviation
The management of patients with impaired or poor ovarian response (POR) undergoing
infertility treatment has challenged reproductive specialists for decades. Different
classifications of POR, including the Bologna Criteria (Ferraretti et al. 2011) and more
recently the POSEIDON criteria (Humaidan et al. 2016), have helped standardize the
definition of the poor responder population. However, this subset of in vitro fertilization (IVF)
patients still remains a therapeutic challenge, experiencing disappointing overall outcomes in
assisted reproductive technology (ART) due to the retrieval of a very low number of oocytes
after exogenous stimulation with gonadotropins (Papathanasiou et al. 2016). In general,
patients with POR have a low ovarian reserve characterized by few antral follicles and low
circulating anti-Müllerian hormone (AMH) levels.
Androgens play a pivotal role in folliculogenesis, facilitating follicle-stimulating
hormone (FSH) receptor (FSHR) expression on granulosa cells via the androgen receptor
(AR). In turn, increased FSHR expression increases follicular responsiveness to FSH (either
endogenous or exogenous) (Weil et al. 1999). In addition, androgens act synergistically with
insulin-like growth factor to stimulate growth of the follicle (Vendola et al. 1999).
Interestingly, long-term exogenous androgen treatment has been shown to increase the
number of pre-antral and antral follicles and, thus, the number of recruitable follicles in both
animal models and in humans (Spinder et al. 1989, Vendola et al. 1998). Furthermore, intrafollicular concentrations of androgens and expression of ARs on granulosa cells are highly
significantly associated with FSHR expression in humans (Nielsen et al. 2011), suggesting
that circulating androgens in humans may facilitate a greater responsiveness of the ovarian
follicle when exposed to exogenous FSH stimulation.
Short-term intra-ovarian ‘androgen priming’ using a combination of human chorionic
gonadotropin (hCG) and aromatase inhibitors during the early follicular phase was shown to
significantly increase the proportion of cycle day 2 antral follicles reaching 14 mm in
diameter, as well as serum estradiol (E2), and follicular fluid (FF) levels of E2 and
testosterone; however, short time priming had no effect on the number of oocytes retrieved
nor on reproductive outcomes (Lossl et al. 2006, Lossl et al. 2008).
In this study we wanted to explore the effect of a novel long-term intra-ovarian
androgen priming protocol, based on the “two cell-two gonadotropin model” in which
luteinizing hormone (LH) activity regulates theca cell conversion of cholesterol into
androgens, and androgens are converted to E2 in granulosa cells due to the combined actions
of FSH and aromatase. In this new protocol, patients are subjected to long-term (8 weeks)
low-level LH-like activity provided by repeated micro doses of recombinant hCG [rhCG]) to
boost local endogenous androgen production by theca cells. To block the conversion of
androgens into E2 in the granulosa cell, an aromatase inhibitor is administered on a daily
basis, thus, increasing the amount of intra-ovarian and circulating endogenous androgens. We
call this concept “intra ovarian priming”.
The present pilot study investigated the effects of the proposed long-term androgen
priming regimen on ovarian reserve parameters in patients considered to have the lowest
ovarian reserve in IVF, the so-called European Society of Human Reproduction and
Embryology (ESHRE) Bologna criteria POR patient.
Materials and methods
Study design and population
This single center pilot study was conducted at IVFMD, My Duc Hospital, Ho Chi Minh City,
Vietnam between February 2018 and March 2019 (NCT03447184). Consecutive patients
who fulfilled the ESHRE Bologna criteria for POR (Ferraretti et al. 2011), determined within
the previous 2 months, were eligible to participate. Other inclusion criteria were: age 18–41
years; body mass index (BMI) <30 kg/m2
and co-treatment with a gonadotropin-releasing
hormone (GnRH) antagonist during ovarian stimulation. All participants provided informed
consent. Women were excluded if they had any chronic medical conditions (e.g. diabetes,
Crohn’s disease, thyroid disease, hepatitis B or a sexually transmitted disease) or were
participating in another interventional clinical trial.
The study was performed in accordance with the ICH Harmonised Tripartite Guideline for
Good Clinical Practice and the ethical principles of the Declaration of Helsinki. Ethics
approval was obtained from the Ethical Board of My Duc Hospital (approval number
01/18/ĐĐ-BVMĐ, dated 22 January 2018). The trial was registered in clinicaltrials.gov
(NCT03447184), Unique Protocol ID: CS/BVMD/18/01
The primary endpoint was serum AMH levels after 8 weeks of priming. Secondary endpoints
included antral follicle count (AFC) (2-10 mm) after 8 weeks of priming, at ovulation trigger
and at oocyte retrieval; serum hCG, testosterone, and progesterone (P4) levels after 8 weeks
of priming, on the day of ovulation trigger, and oocyte retrieval.
Androgen priming started 8 weeks prior to stimulation for IVF, on day two of menses. From
this day onwards patients received low-dose rhCG (Ovitrelle
) 260 IU subcutaneously (SC)
every second day in addition to letrozole 2.5 mg daily for 8 weeks, stopping on the first stay
of stimulation for IVF with exogenous gonadotropins. Alongside androgen priming, pituitary
down-regulation with a GnRH agonist (GnRHa) started (depot injection of triptorelin 3.75
mg); a second depot dose of triptorelin was given 28 days later. Thus, the total priming
period lasted for 8 weeks (Figure 1).
Stimulation, monitoring and oocyte pick-up
Stimulation with exogenous gonadotropins started after 8 weeks of androgen priming,
irrespective of menses. A fixed dose of recombinant FSH (rFSH; Gonal F
, Merck) was used
for the first 5 days, either 300 IU rFSH alone in patients aged <35 years or 300 IU rFSH and
150 IU rLH (follitropin alfa/lutropin alfa, Pergoveris
, Merck) in patients aged 35 years
(Figure 1). Monitoring was undertaken according to the standard clinical procedure. Cotreatment with a GnRH antagonist started on the fifth day of stimulation. A bolus of 6,500 IU
, Merck) was given for triggering of final oocyte maturation (Figure 1).
Oocyte pick-up (OPU) and embryo transfer were performed based on local policies and
procedures. Luteal phase support was provided by vaginal micronized progesterone, 180 mg
A total of six blood samples (2 ml each) were collected during the study period: at baseline,
during priming after 4 and 8 weeks, on the day of starting stimulation, at ovulation trigger
and at OPU. The first sample at baseline was used to determine FSH, LH, E2, testosterone,
and AMH levels. The other five samples were used to determine FSH, LH, E2, testosterone,
P4, hCG and AMH levels.
Ultrasound examination to count all antral follicles of 2-10 mm in diameter in each ovary was
performed at each of the same time points.
No previous data exist on long-term endogenous androgen priming in IVF patients, and due
to the explorative character of this pilot study a sample size calculation was not feasible;
Instead we decided to recruit a total of 30 ESHRE Bologna criteria POR patients. Data were
analyzed using SPSS version 20 software. Continuous variables are presented as mean
standard deviation (SD) and were compared using Student’s t-test. Categorical data are
expressed as numbers and were compared using the Chi-square test. All tests were two tailed,
and a p-value of <0.05 was considered statistically significant.
A total of 30 patients were enrolled in this pilot study. Patients were lean, the majority had
primary infertility, and the mean number of previous unsuccessful IVF attempts was two
(Table 1). A summary of stimulation details and outcomes is provided in Table 2.
Effects of priming
Mean AMH levels decreased steadily over time (Table 3) and in most patients the AMH
value after 8 weeks of androgen priming was lower than that at baseline (Figure 2). Longterm androgen priming had no significant effects on the AFC (Table 3). FSH levels decreased,
from baseline to 4 weeks, increased slightly at 8 weeks, and then increased markedly at
ovulation trigger, following exogenous stimulation (Table 3). Levels of circulating
testosterone and P4 remained unchanged during the androgen priming period, to increase at
ovulation trigger and OPU (Table 3). No changes were seen in E2 and hCG levels during
androgen priming; as expected E2 levels increased during stimulation, reaching peak levels at
ovulation trigger and decreasing at OPU (Table 3). LH levels decreased from baseline after 4
and 8 weeks during androgen priming, to increase slightly at ovulation trigger and, as
expected, decrease at OPU (Table 3).
Embryo transfer and reproductive outcomes
The mean number of good-quality embryos per patient was 1.050.90. Zero, one, two and
three good quality embryos were available in 7 (23/3%), 8 (26.7%), 6 (20.0%) and 1 (3.3%)
patients, respectively; seven patients (23.3%) had no good-quality embryos available.
Embryo transfer (ET) was performed on day 3 in 15 patients (fresh ET in 5 and frozen ET in
10 patients). The reasons for no ET in the remaining 15 patients included cycle cancellation
due to inadequate response (n=6), no oocytes retrieved (n=2), patient request to delay ET
(n=1) or not to transfer a single poor-quality embryo (n=2), ET not performed during the
study period (n=3), and no embryo (n=1). The characteristics of the seven patients who did
not undergo ET were similar to those in the overall study population. A mean of 1.60.5
embryos were transferred per patient. One clinical pregnancy was achieved, but this ended as
a miscarriage before 12 weeks’ gestation.
In this study we explored a novel concept of long-term “intra-ovarian” androgen priming in
Bologna criteria POR IVF patients. The idea of the protocol was to possibly boost the
residual follicular pool in the POR patient during long-term androgen stimulation, which
might physiologically increase the number of recruitable follicles as well as the follicular
response to exogenous FSH stimulation. We clearly showed that the regimen used in this
study had no effect on steroid levels and follicles prior to and after priming. Thus, the
response to stimulation in this poor ovarian reserve population was similar to what has
previously been reported (Humaidan et al. 2017).
The stimulation of androgen production by theca cells was performed by micro-dose
administration of hCG, while aromatization into estrogens was prevented by coadministration of the aromatase inhibitor letrozole. In addition, women underwent pituitary
down-regulation using a GnRHa depot preparation for the entire duration of the 8-week
priming period. There were two main reasons for using the ultra-long protocol. Firstly,
because letrozole might induce follicular recruitment and development during the priming
period, and secondly because we wanted to have a synchronous cohort of follicles before
starting stimulation with exogenous FSH.
Although previous studies suggested that the sensitivity of developing follicles to FSH
stimulation might be enhanced by androgens via stimulation of the AR (Lossl et al. 2006,
Lossl et al. 2008), the results of the present study clearly showed that the approach used was
insufficient to cause an endogenous elevation of circulating testosterone, and that the number
of antral follicles available for ovarian stimulation did not increase. Nevertheless, the
circulating androgen level sufficient to act at the ovarian level on pre-antral and small antral
follicles is unknown and cannot be determined from our data. Clinical outcomes in the small
group of women in our study were clearly not satisfactory. However, this is consistent with
earlier studies, showing no significant improvement in implantation or ongoing pregnancy
rates after use of short-term androgen priming regimens (Lossl et al. 2006, Lossl et al. 2008).
There are several potential explanations for our findings. Firstly, the pituitary GnRHa downregulation effectively reduced endogenous FSH and LH levels. Therefore, the rhCG given
only replaced the endogenously produced LH and did not increase the LH drive on theca cell
androgen production, resulting in stable low AMH levels. It would, therefore, appear that the
rhCG dosage of 260 IU administered every second day was too low to cause an augmented
drive for theca cell androgen synthesis despite the presence of the aromatase inhibitor
blocking conversion into estrogens. Equally important, the drive from endogenous FSH to
augment theca cell androgen synthesis was absent because FSH was effectively downregulated, and from a physiological point of view FSH secures synthesis of granulosa cell
paracrine hormones that act synergistically with LH to increase theca cell androgen output
including inhibin-A, inhibin-B and insulin-like growth factor-2 (Hillier et al. 1991, Nahum et
This synergistic effect is pronounced and the presence of both IGF1, inhibin-A and LH
results in a more than 20-times higher secretion of androstenedione from cultured human
theca cells than LH alone (Hillier et al. 1991, Nahum et al. 1995). Thus, in order to improve
the present protocol, it could be suggested that pituitary GnRHa down regulation should be
avoided during the priming period to maintain pituitary function, allowing physiological
levels of circulating FSH and LH. Moreover, the hCG dosage might be increased to provide
stronger theca cell stimulation. A rhCG dosage of 260 IU every second day is approximately
equivalent to a daily dosage of 130 IU. A previous randomized controlled trial (RCT) of
GnRHa down-regulated women undergoing ovarian stimulation with recombinant FSH
titrated the hCG dosage by providing 0, 50, 100 or 150 IU hCG daily (Thuesen et al. 2013).
Steady-state hCG concentrations were achieved after 6 days, being approximately 3, 6 and 11
IU/L after the 50, 100 and 150 IU dosages, respectively. In that study, increasing hCG levels
in the context of a constant FSH stimulation was associated with a dose-dependent increase
in androgen, P4, and E2 levels. Therefore, in the current study, according to the
abovementioned previous findings, it was expected that a daily rhCG dosage of 130 IU would
result in a steady-state concentration of around 8-9 IU/L which theoretically should have
been sufficient to increase the androgen production. In contrast to the RCT described above
(Thuesen et al. 2013), the steady-state concentration of hCG in the current study was about 5
IU/L. A clear explanation for this discrepancy cannot readily be given, but it might relate to
ethnic differences between the Asian and European population of the two studies; however,
these findings indicate that the hCG administration might benefit form an increase.
Despite the negative findings of the present study, we see it as a preliminary study and, thus,
still consider intra-ovarian priming to be an interesting future model for the poor ovarian
reserve patient, which after modifications of the present protocol should induce an increase in
local intraovarian testosterone production and granulosa cell FSHR expression. We believe
that such modifications would be superior to exogenous testosterone supplementation
(Massin et al. 2006, Fabregues et al. 2009, Bosdou et al. 2016), in which the effect of
exogenous testosterone is primarily systemic, with only a fraction of the testosterone
supplemented reaching the theca-granulosa cell compartment and with potentially unwanted
systemic side effects.
This proof-of-concept study provides the first data on the effects of long-term intra- ovarian
androgen priming in patients with POR. However, the results need to be interpreted in light
of several important limitations. Firstly, because this was a pilot study, the number of patients
enrolled was small. Secondly, the study design did not include a control group, making it
difficult to determine whether the results seen could be attributed to the androgen priming
regimen or were caused by other factors. Finally, the single ethnicity population included in
this study could limit the external validity of our findings.
In conclusion, long-term intra-ovarian androgen priming regimen was not associated with
any increase in AMH or AFC in Bologna criteria POR patients undergoing IVF. No ongoing
pregnancies or live births were achieved in the 15 patients who underwent ET. It is suggested
that androgen priming should be performed without GnRHa down-regulation. Thus, the
present approach to androgen priming should not currently be used in patients with POR, but
rather await the results of protocol modifications. Importantly, additional research is needed
in this area before clear recommendations about the value of intra-ovarian androgen priming
in a clinical setting can be made.
This research did not receive any specific grant from funding agencies in the public,
commercial, or not-for-profit sectors.
Conflict of interest
All authors state that they have no conflicts of interest to disclose.
All data generated or analyzed during this study are included in this published article.
Cycle cancelled due to inadequate response 6 (20.0)
No oocyte retrieval 2 (6.7)
Values are mean standard deviation, or number of patients (%).
FSH, follicle-stimulating hormone; MII, metaphase II; rFSH, recombination folliclestimulating hormone.
Values are median (interquartile range).
AFC, antral follicle count; AMH, anti-Müllerian hormone; FSH, follicle-stimulating hormone; hCG, human chorionic gonadotropin; LH, luteinizing hormone;
NA, not applicable.
Figure 1. Intra-ovarian androgen priming and stimulation protocol (8 weeks’ androgen
priming, followed by stimulation (gonadotropin-releasing hormone antagonist protocol) and
trigger. GnRH, gonadotropin-releasing hormone; GnRHa, GnRH agonist; rFSH, recombinant
follicle-stimulating hormone; rhCG, recombinant human chorionic gonadotropin; rLH,
recombinant luteinizing hormone.
Figure 2. Bland-Altman plot showing relationship between anti-Müllerian hormone (AMH)
levels at baseline and end of treatment. Each dot represents one patient. The x-axis is the
mean of two AMH values (at baseline and after 8 weeks of androgen priming) and the y-axis
is the AMH value at 8 weeks minus the AMH value at baseline.
Bosdou, J.K., Venetis, C.A., Dafopoulos, K., Zepiridis, L., Chatzimeletiou, K., Anifandis, G.,
Mitsoli, A., Makedos, A., Messinis, I.E., Tarlatzis, B.C., Kolibianakis, E.M., 2016.
Transdermal testosterone pretreatment in poor responders undergoing ICSI: a
randomized clinical trial. Hum. Reprod. 31, 977-985.
Fabregues, F., Penarrubia, J., Creus, M., Manau, D., Casals, G., Carmona, F., Balasch, J.,
2009. Transdermal testosterone may improve ovarian response to gonadotrophins in
low-responder IVF patients: a randomized, clinical trial. Hum. Reprod. 24, 349-359.
Ferraretti, A.P., La Marca, A., Fauser, B.C., Tarlatzis, B., Nargund, G., Gianaroli, L., ESHRE
working group on Poor Ovarian Response Definition., 2011. ESHRE consensus on
the definition of ‘poor response’ to ovarian stimulation for in vitro fertilization: the
Bologna criteria. Hum. Reprod. 26, 1616-1624.
Hillier, S.G., Yong, E.L., Illingworth, P.J., Baird, D.T., Schwall, R.H., Mason, A.J., 1991.
Effect of recombinant inhibin on androgen synthesis in cultured human thecal cells.
Mol. Cell. Endocrinol. 75, R1-6.
Humaidan, P., Alviggi, C., Fischer, R., Esteves, S.C., 2016. The novel POSEIDON
stratification of ‘Low prognosis patients in Assisted Reproductive Technology’ and its
proposed marker of successful outcome. F1000Res. 5, 2911.
Humaidan, P., Chin, W., Rogoff, D., D’Hooghe, T., Longobardi, S., Hubbard, J., Schertz, J.,
ESPART Study Investigators., 2017. Efficacy and safety of follitropin alfa/lutropin
alfa in ART: a randomized controlled trial in poor ovarian responders. Hum. Reprod.
Lossl, K., Andersen, A.N., Loft, A., Freiesleben, N.L., Bangsbøll, S., Andersen, C.Y., 2006.
Androgen priming using aromatase inhibitor and hCG during early-follicular-phase
GnRH antagonist down-regulation in modified antagonist protocols. Hum. Reprod. 21,
Lossl, K., Andersen, C.Y., Loft, A., Freiesleben, N.L., Bangsbøll, S., Andersen, A.N., 2008.
Short-term androgen priming by use of aromatase inhibitor and hCG before controlled
ovarian stimulation for IVF. A randomized controlled trial. Hum. Reprod. 23, 1820-
Massin, N., Cedrin-Durnerin, I., Coussieu, C., Galey-Fontaine, J., Wolf, J.P., Hugues, J.N.,
2006. Effects of transdermal testosterone application on the ovarian response to FSH
in poor responders undergoing assisted reproduction technique–a prospective,
randomized, double-blind study. Hum. Reprod. 21, 1204-1211.
Nahum, R., Thong, K.J., Hillier, S.G., 1995. Metabolic regulation of androgen production by
human thecal cells in vitro. Hum. Reprod. 10, 75-81.
Nielsen, M.E., Rasmussen, I.A., Kristensen, S.G., Christensen, S.T., Møllgård, K., Wreford
Andersen, E., Byskov, A.G., Yding Andersen, C., 2011. In human granulosa cells
from small antral follicles, androgen receptor mRNA and androgen levels in follicular
fluid correlate with FSH receptor mRNA. Mol. Hum. Reprod. 17, 63-70.
Papathanasiou, A., Searle, B.J., King, N.M., Bhattacharya, S., 2016. Trends in ‘poor
responder’ research: lessons learned from RCTs in assisted conception. Hum. Reprod.
Update. 22, 306-319.
Spinder, T., Spijkstra, J.J., Van Den Tweel, J.G., Burger, C.W., van Kessel, H., Hompes, P.G.,
Gooren, L.J., 1989. The effects of long term testosterone administration on pulsatile
luteinizing hormone secretion and on ovarian histology in eugonadal female to male
transsexual subjects. J. Clin. Endocrinol. Metab. 69, 151-157.
Thuesen, L.L., Smitz, J., Loft, A., Nyboe Andersen, A., 2013. Endocrine effects of hCG
supplementation to recombinant FSH throughout controlled ovarian stimulation for
IVF: a dose-response study. Clin. Endocrinol. (Oxf). 79, 708-715.
Vendola, K.A., Zhou, J., Adesanya, O.O., Weil, S.J., Bondy, C.A., 1998. Androgens
stimulate early stages of follicular growth in the primate ovary. J. Clin. Invest. 101,
Vendola, K., Zhou, J., Wang, J., Famuyiwa, O.A., Bievre, M., Bondy, C.A., 1999. Androgens
promote oocyte insulin-like growth factor I expression and initiation of follicle
development in the primate ovary. Biol. Reprod. 61, 353-357.
Weil, S., Vendola, K., Zhou, J., Bondy, C.A., 1999. Androgen and follicle-stimulating
hormone interactions in primate ovarian follicle development. J. Clin. Endocrinol.
Metab. 84, 2951-2956.
Dr Vuong is Head of OB/GYN Department, University of Medicine and Pharmacy at HCM
City, Vietnam. She was one of the IVF Letrozole pioneers in Vietnam, starting in 1997. She has
performed about 20,000 ART cycles. Her areas of research activity include ovulation
induction, poor responder, IVM, and frozen embryo transfer.