International Journal of

Environmental Research
and Public Health

Article
Association between GnRH Receptor Polymorphisms and
Luteinizing Hormone Levels for Low Ovarian Reserve
Infertile Women
Shun-Long Weng 1,2,3 , Shu-Ling Tzeng 4,5 , Chun-I Lee 6,7,8 , Chung-Hsien Liu 6,7 , Chun-Chia Huang 8 ,
Shun-Fa Yang 4,5 , Maw-Sheng Lee 4,6,8 and Tsung-Hsien Lee 4,6,8, *

1 Department of Obstetrics and Gynecology, Hsinchu MacKay Memorial Hospital, Hsinchu 30071, Taiwan;
[email protected]
2 Department of Medicine, MacKay Medical College, New Taipei 25245, Taiwan
3 Mackay Junior College of Medicine, Nursing and Management College, Taipei 11260, Taiwan
4 Institute of Medicine, Chung Shan Medical University, Taichung 40203, Taiwan; [email protected] (S.-L.T.);
[email protected] (S.-F.Y.); [email protected] (M.-S.L.)
5 Department of Medical Research, Chung Shan Medical University Hospital, Taichung 40203, Taiwan
6 Department of Obstetrics and Gynecology, Chung Shan Medical University Hospital, Taichung 40203, Taiwan;
[email protected] (C.-I.L.); [email protected] (C.-H.L.)
7 Department of Medicine, School of Medicine, Chung Shan Medical University, Taichung 40203, Taiwan
8 Division of Infertility Clinic, Lee Women’s Hospital, Taichung 40602, Taiwan; [email protected]
* Correspondence: [email protected]



Abstract: The choice of ovarian stimulation protocols in assisted reproduction technology (ART)


cycles for low ovarian reserve patients is challenging. Our previous report indicated that the
Citation: Weng, S.-L.; Tzeng, S.-L.;
gonadotrophin-releasing (GnRH) agonist (GnRHa) protocol is better than the GnRH antagonist
Lee, C.-I; Liu, C.-H.; Huang, C.-C.;
(GnRHant) protocol for young age poor responders. Here, we recruited 269 patients with anti-
Yang, S.-F.; Lee, M.-S.; Lee, T.-H.
Müllerian hormone (AMH) < 1.2 ng/mL undergoing their first ART cycles for this nested case-control
Association between GnRH Receptor
study. We investigated the genetic variants of the relevant genes, including follicular stimulating
Polymorphisms and Luteinizing
Hormone Levels for Low Ovarian
hormone receptor (FSHR; rs6166), AMH (rs10407022), GnRH (rs6185), and GnRH receptor (GnRHR;
Reserve Infertile Women. Int. J. rs3756159) in patients <35 years (n = 86) and patients ≥35 years of age (n = 183). Only the genotype
Environ. Res. Public Health 2021, 18, of GnRHR (rs3756159) is distributed differently in young (CC 39.5%, CT/TT 60.5%) versus advanced
7006. https://doi.org/10.3390/ (CC 24.0%, CT/TT 76.0%) age groups (recessive model, p = 0.0091). Furthermore, the baseline
ijerph18137006 luteinizing hormone (LH) levels (3.60 (2.45 to 5.40) vs. 4.40 (2.91 to 6.48)) are different between CC
and CT/TT genotype of GnRHR (rs3756159). In conclusion, the genetic variants of GnRHR (rs3756159)
Academic Editor: Rosaria Meccariello could modulate the release of LH in the pituitary gland and might then affect the outcome of ovarian
stimulation by GnRHant or GnRHa protocols for patients with low AMH levels.
Received: 19 May 2021
Accepted: 26 June 2021
Keywords: GnRH receptor; anti-Müllerian hormone; assisted reproduction technology; single nu-
Published: 30 June 2021
cleotide polymorphism; poor responders; POSEIDON criteria; GnRH antagonist; GnRH agonist

Publisher’s Note: MDPI stays neutral

with regard to jurisdictional claims in
published maps and institutional affil-
1. Introduction
iations.
In assisted reproduction technology (ART) for infertile couples, management for pa-
tients with an inadequate ovarian response is challenging [1–4]. Although investigators
propose two international definitions, the Bologna [5] and POSEIDON [6] criteria, for poor
Copyright: © 2021 by the authors.
responders, the choice of ovarian stimulation protocols for these patients is still controver-
Licensee MDPI, Basel, Switzerland.
sial [7,8]. Most reports recommended the protocol for the use of gonadotropin-releasing
This article is an open access article
hormone (GnRH) antagonist (GnRHant) or the mild stimulation protocol (MSP) for the
distributed under the terms and poor responders, instead of the long protocol for using GnRH agonist (GnRHa) [9]. The pri-
conditions of the Creative Commons mary reason for this recommendation is the cost-effective consideration, which is adjusted
Attribution (CC BY) license (https:// by the cost per oocyte retrieved and the number of exogenous gonadotropin injections.
creativecommons.org/licenses/by/ However, because it remains controversial, the GnRHa protocol is less recommended for
4.0/). poor responders.

Int. J. Environ. Res. Public Health 2021, 18, 7006. https://doi.org/10.3390/ijerph18137006 https://www.mdpi.com/journal/ijerph
Int. J. Environ. Res. Public Health 2021, 18, 7006 2 of 11

The number of retrieved oocytes and patients’ age are the key predictors for a success-
ful pregnancy and live birth in ART cycles [1]. The aneuploidy rates of oocytes/embryos
are intimately correlated with maternal age. This means that oocyte quantity and quality
(euploidy) are the critical factors for successful ART cycles. Our previous study, however,
indicated that for young patients (<35 years of age) with low anti-Mullerian hormone
(AMH) levels, the GnRHa protocol is better than the GnRHant protocol in terms of embryos
available for transfer and pregnancy rate [10]. A randomized clinical trial also reported
that the GnRHant protocol is correlated with a higher cancellation rate than that of the
GnRHa protocol [11]. Those results suggested that the follicular and the subsequent em-
bryo development after controlled ovarian stimulation might differ between GnRHant and
GnRHa protocols for poor responders.
The response (number of retrieved oocytes) to ART protocols could be predicted by
ovarian reserve markers, such as AMH or antral follicle count (AFC) [12,13]. However,
some researchers observed unexpected poor ovarian responders after controlled ovarian
stimulation in ART cycles, such as POSEIDON group 1 and group 2 patients. The single
nucleotide polymorphism (SNP) in the hormones and hormone receptors related to fol-
licular growth and development might cause such inadequate ovarian response [14]. For
example, SNPs of AMH [14,15], follicular stimulating hormone (FSH) [16,17], and FSH
receptor (FSHR) [16–21] have been surveyed to explain the ovarian response subsequent to
controlled ovarian stimulation in ART cycles [14,17].
We previously reported a better pregnancy outcome by the GnRHa than the GnRHant
protocol in POSEIDON group 3 patients [10]. Therefore, we raised the hypothesis that,
in addition to the SNPs of AMH (rs10407022) and FSHR (rs6166), the SNPs of GnRH
(rs6185) [22] or GnRH receptor (GnRHR; rs3756159) [23], may account for the difference of
embryo development and pregnancy outcome for patients with low AMH levels (POSEI-
DON group 3 and group 4). The present study results revealed that the distribution of SNP
of GnRHR (rs3756159) varied between POSEIDON group 3 and group 4 patients, and those
SNPs are associated with varied baseline LH levels in ART cycles.

2. Materials and Methods

2.1. Study Design and Patient Selection
The infertile couples who underwent their first ART treatment cycle from January
2014 to December 2015 were recruited for this prospective nested case-control study. The in-
clusion criteria were as follows: (1) women age <45 years old; (2) serum AMH <1.2 ng/mL
before ART treatment; and (3) no histories of ovarian surgery or pelvic radiation treatment.
We drew a venous blood sample for DNA extraction and subsequent analysis for the
chosen SNPs. The Institutional Review Board of Chung Shan Medical University Hospital
approved the study protocol (CS13194 and CS2-14033). A written informed consent was
obtained from each participant. All the recruited women for this analysis were Han Chinese
people. Clinical trial register number: ISRCTN12768989.
We attempted to study the SNPs of related hormone molecules for ovarian responses
in the GnRHa and GnRHant protocols. The methods for chosen SNPs were in line with
our previous report [24], which was based on the searches in dbSNP (http://www.ncbi.
nlm.nih.gov/snp, accessed 3 November 2013) and the international HapMap project
(http://hapmap.ncbi.nlm.nih.gov, accessed 3 November 2013). Consequently, we sur-
veyed the SNPs of AMH 146 T > G (rs10407022), FSHR Asn680Ser (rs6166), GnRH (rs6185),
and GnRHR (rs3756159). The AMH 146 T > G is at the coding region of AMH and causes
amnio acid substitution. FSHR Asn680Ser (rs6166) is the most common reported SNP that
related to ovarian responses in ART cycles [16–21]. GnRH (rs6185) and GnRHR (rs3756159)
are chosen to discriminate the different activities of GnRHa and GnRHanta in ART treat-
ment. GnRH (rs6185) is located at the coding region and results in amnio acid substitution.
GnRHR (rs3756159) is located at position 68305073 on chromosome 4 in the 5’ untranslated
region of the GnRHR gene.
Int. J. Environ. Res. Public Health 2021, 18, 7006 3 of 11

2.2. ART Treatment Protocol and Hormone Analysis

During the study period from January 2014 to December 2015, we recruited those
patients undergoing the same GnRHa stimulation protocol to avoid bias in the association
between the chosen SNPs and ART outcomes. The ovarian stimulation procedure is
the same as previously described [24]. The GnRHa protocol comprises daily injections
of 0.5 mg of leuprolide acetate (Lupron; Takeda Pharmaceutics, Konstanz, Germany)
from the mid-luteal phase (cycle day 21) of the previous cycle. After that, recombinant
FSH (Gonal-F, Merck-Serono, Darmstadt, Germany) or highly purified FSH (Menopur;
Ferring Pharmaceuticals) was administered daily for follicular growth. We used 10,000 IU
human chorionic gonadotropin (Profasi, Serono, Norwell, MA, USA) to trigger final oocyte
maturation, and ovum pick-up was performed 36 to 38 h later.
The baseline AMH, FSH, luteinizing hormone (LH), and estradiol (E2) levels were
measured on day 2 to 3 of the menstruation cycles before the controlled ovarian stimulation.
On day 2 to 3 of the hyper-stimulation cycle prior to gonadotropin injection, FSH, LH, and
E2 levels were assessed again. On the day of hCG trigger, measurement of E2, LH, and
progesterone (P4) levels was performed. The AMH/MIS ELISA kit (Immunotech/Beckman
Coulter Inc., Marseille Cedex, France) was used in duplicate to assess serum AMH levels.
A specific immunometric assay kit (Access; Beckman Coulter Inc., Fullerton, CA, USA) was
utilized to measure serum FSH, LH, E2, and P4 levels. The sensitivity, intra-assay coefficient
of variation (CV), and interassay CV for the FSH measurement were 0.2 mIU/mL, 4.3%, and
5.6%, respectively. The sensitivity, intra-assay CV, and interassay CV for the LH evaluation
were 0.2 mIU/mL, 5.4%, and 6.4%, respectively.

2.3. DNA Extraction and Determination of Genotypes

We obtained genomic DNA with a QIAamp DNA blood mini kit (Qiagen, Valencia,
CA, USA) from EDTA anti-coagulated venous blood. Genomic DNA extraction was
performed following the manufacturer’s instructions as in a previous report [25]. We used
Tris-EDTA (TE) buffer to disperse the extracted DNA and then measured the optical density
at 260 nm to determine DNA quantity. The final solution was stored at −20 ◦ C and used
as polymerase chain reaction (PCR) templates. Genotyping of the four studied SNPs was
assessed with the ABI StepOne™ Real-Time PCR System (Applied Biosystems, Foster City,
CA, USA), and allele discrimination was analyzed using SDS version 3.0 software (Applied
Biosystems, Foster City, CA, USA) and the TaqMan assay (Applied Biosystems, Foster City,
CA, USA) [26]. The primers used for each genotype are listed in Table 1.

Table 1. The context sequence of four primers to detect AMH, FSHR, GnRH, and GnRHR SNPs in the study.

Variable Assay ID Context Sequence

GAAGACTTGGACTGGCCTCCAGGCA[G/T]
AMH 146 T > G (rs10407022) C__25599842_10
CCCACAAGAGCCTCTGTGCCTGGTG
AGGGACAAGTATGTAAGTGGAACCA[C/T]
FSHR A2039G (rs6166) C___2676874_10
TGGTGACTCTGGGAGCTGAAGAGCA
CTGGCTGGAGCAGCCTTCCACGCAC[C/G]
GnRH-1 (rs6185) C___1529427_1_
AAGTCAGTAGAATAAGGCCAGCTAG
AACATGAAAGGTATAAAGCCCTCAA[A/G]
GnRHR-1 (rs3756159) C__27477550_10
TGCAGGGTGTGGCTATGAAAGTCGG

2.4. Statistical Analysis

The data supplemental to this prospective study is listed in the Supplementary File
S1: SNP_low_AMH_single.txt A chi-square test was performed to determine the Hardy–
Weinberg equilibrium, including AMH (rs10407022), FSHR (rs6166), GnRH (rs6185), and
GnRHR (rs3756159). Then, a chi-square test examined the associations between POSEIDON
group 3/4 and tested SNPs under the genotypic (AA versus Aa versus aa) and recessive
(AA versus Aa/aa) models.
 and GnRHR (rs3756159). Then, a chi-square test examined the associations between PO-
SEIDON group 3/4 and tested SNPs under the genotypic (AA versus Aa versus aa) and
recessive (AA versus Aa/aa) models.
The Kolmogorov–Smirnov test was used for the demographic characteristics and
Int. J. Environ. Res. Public Health 2021, 18, 7006
other clinical parameters about ovarian responses to determine the distribution of those4 of 11
variables. After that, the continuous variables are presented as medians (interquartile
range (IQR, 25th–75th percentile)), whereas categorical variables are shown as numbers
and percentages. We The usedKolmogorov–Smirnov
the Mann–Whitney test was used for the demographic characteristics and
U test (for continuous variables) or
other clinical parameters about ovarian responses to determine the distribution of those
chi-square test (forvariables. categorical Afteritems)
that, theto compare
continuous the differences
variables are presentedbetween
as mediansgroups withrange
(interquartile
genetic variants under (IQR,the recessive
25th–75th model (AA
percentile)), versus
whereas Aa + aa).
categorical All data
variables were as
are shown analyzed
numbers and
using the IBM SPSS Statistics for Windows, Version 22.0 (IBM Corp., Armonk, NY,orUSA).
percentages. We used the Mann–Whitney U test (for continuous variables) chi-square
p-values < 0.05 weretest (for categorical
considered items) to compare
statistically the differences between groups with genetic variants
significant.
under the recessive model (AA versus Aa + aa). All data were analyzed using the IBM
SPSS Statistics for Windows, Version 22.0 (IBM Corp., Armonk, NY, USA). p-values < 0.05
3. Results were considered statistically significant.
A total of 269 patients with low AMH levels undergoing their first IVF/ICSI cycles
3. Results
were recruited for this nested case-control study in a prospective cohort. The 269 patients
A total of 269 patients with low AMH levels undergoing their first IVF/ICSI cycles
were divided into young (< 35 years
were recruited for thisofnested
age, case-control
POSEIDON group
study 3, n= 86) and
in a prospective advanced
cohort. (≥
The 269 patients
35 years of age, POSEIDON
were dividedgroup 4, n = 183)
into young age groups.
(<35 years of age, POSEIDON group 3, n = 86) and advanced
(≥35 years
The representative results of real-time
of age, POSEIDONPCR groupand TagMan
4, n = assay for each SNP geno-
183) age groups.
type are shown in Figure 1 (for AMH 146 T > G (rs10407022)), Figure 2assay
The representative results of real-time PCR and TagMan (for for
FSHReachA2039G
SNP genotype
are shown in Figure 1 (for AMH 146 T > G (rs10407022)), Figure 2 (for FSHR A2039G
(rs6166)), Figure 3 (for GnRH-1 (rs61850)), and Figure 4 (for GnRHR-1 (rs3756159)).
(rs6166)), Figure 3 (for GnRH-1 (rs61850)), and Figure 4 (for GnRHR-1 (rs3756159)).

Figure 1. Representative
Figure TaqMan
1. Representative TaqMan AMH for
assay forassay rs10407022 genotyping. The
AMH rs10407022 FAM (blue)The
genotyping. and FAM
VIC (green)
(blue)fluorescence
and VIC
probes detect T and G alleles, respectively. The ROX (red) fluorescence probes are used for calibration.
(green) fluorescence probes detect T and G alleles, respectively. The ROX (red) fluorescence probes
are used for calibration.
3.1. The Distribution of AMH, FSHR, GnRH, and GnRHR
The SNPs in AMH 146 T > G (rs10407022; Table 2), FSHR A2039G (rs6166; Table 3),
and GnRH-1 (rs6185; Table 4), were not correlated with patients with diminished ovarian
reserve phenotype. Only SNP in GnRHR-1 (rs3756159) was distributed differently in young
(CC 39.5%, CT/TT 52 (60.5%)) versus advanced (CC 24.0%, CT/TT 139 (76.0%)) age groups
(recessive model, p = 0.0091; Table 5).
on. Res. Public Health 2021, 18, x 5 of 12
Int. Health
on. Res. Public J. Environ.
2021,Res.
18, Public
x Health 2021, 18, 7006 5 of 12 5 of 11

Figure 2. Representative TaqMan assay for FSHR (rs6166) genotyping. The FAM (blue) and VIC
Representative
Figure 2.(green) TaqMan
fluorescence assay
probes detect FSHR
for A and (rs6166) genotyping.
G alleles, The
respectively. FAM
The (blue)
ROX (red)and VIC (green)
fluorescence fluorescence probes
probes
Figure 2. Representative TaqMan assay for FSHR (rs6166) genotyping. The FAM (blue) and VIC
detect Aisand
used
G for calibration.
alleles, respectively. The ROX (red) fluorescence probes is used for calibration.
(green) fluorescence probes detect A and G alleles, respectively. The ROX (red) fluorescence probes
is used for calibration.

Figure 3. Representative TaqMan assay for GNRH1 (rs6185) genotyping. The FAM (blue) and VIC (green) fluorescence
Figure 3. Representative TaqMan assay for GNRH1 (rs6185) genotyping. The FAM (blue) and VIC
probes detect Gfluorescence
(green) and C alleles,probes
respectively.
detect GThe
andROX (red) fluorescence
C alleles, respectively.probes is used
The ROX (red)for calibration.probes
fluorescence
Figure 3. Representative TaqMan assay for GNRH1 (rs6185) genotyping. The FAM (blue) and VIC
is used for calibration.
(green) fluorescence probes detect G and C alleles, respectively. The ROX (red) fluorescence probes
is used for calibration.
nviron. Res. Int. J. Environ.
Public Res. Public
Health 2021, 18, x Health 2021, 18, 7006 6 of 12 6 of 11

Figure 4. Representative
Figure 4. Representative TaqMan assayTaqMan assay for
for GnRHR GNRHR genotyping.
(rs3756159) (rs3756159) genotyping. The and
The FAM (blue) FAMVIC (blue) and fluorescence
(green)
VIC (green) fluorescence probes detect C and T alleles, respectively. The ROX (red) fluorescence
probes detect C and T alleles, respectively. The ROX (red) fluorescence probes is used for calibration.
probes is used for calibration.
Table 2. The frequencies of AMH 146T > G (rs10407022) SNP among women with serum
3.1. The Distribution
AMH of AMH,
< 1.2 FSHR,
ng/mL GnRH,
in varied ageand GnRHR
groups.
The SNPs in AMH 146 T > G (rs10407022; Table 2), FSHR A2039G (rs6166; Table 3),
AMH
and GnRH-1 (rs6185; 146 T4),
Table >G were notAge
correlated with
< 35 Years (n =patients
86) with
Age ≥diminished
35 Years (n ovarian
= 183) p Value 1
(rs10407022)
reserve phenotype. Only SNP in GnRHR-1 (rs3756159) was distributed differently in
Genotype
young (CC 39.5%, CT/TT n advanced %
52 (60.5%)) versus n
(CC 24.0%, CT/TT % age
139 (76.0%))
TT 37/86
groups (recessive model, p = 0.0091; Table 5). 43.0 62/183 33.9 Reference
TG 37/86 43.0 90/183 49.2 0.1913
GG 12/86 14.0 31/183 16.9 0.2773
Table 2. The frequencies of AMH 146T > G (rs10407022) SNP among women with serum AMH <
Recessive
1.2 ng/mL in varied age groups.
TT 37/86 43.0 62/183 33.9 Reference
TG/GG 49/86 57.0 121/183 66.1 0.1478
AMH 146 T >
AgeAllele
< 35 years (n = 86) Age ≥ 35 years (n = 183) p Value 1
G (rs10407022) T 111/172 64.5 214/366 58.5 Reference
Genotype n G % 61/172 n 35.5 %
152/366 41.5 0.1802
TT 1 37/86 test.
by Chi-square 43.0 62/183 33.9 Reference
TG 37/86 43.0 90/183 49.2 0.1913
GG 3. The frequencies
Table12/86 of FSHR A2039G
14.0 (rs6166) SNP among
31/183 16.9 women with serum AMH < 1.2 ng/mL
0.2773
Recessive in varied age groups.
TT 37/86 62/183 33.9 Reference
FSHR A2039G
TG/GG 49/86 57.0Age < 35 Years (n = 86)
121/183 66.1 ≥ 35 Years0.1478
Age (n = 183) p Value 1
(rs6166)
Allele Genotype n % n %
T 111/172
AA 64.5 30/86 214/366
34.9 58.5
73/183 Reference
39.9 Reference
G 61/172
AG 35.5 48/86 152/366
55.8 41.5
91/183 0.1802
49.7 0.3746
1 by Chi-square test. GG 8/86 9.3 19/183 10.4 0.9593
Recessive
AA 30/86 34.9 73/183 39.9 Reference
AG/GG 56/86 65.1 110/183 60.1 0.4316
Allele
A 108/172 62.8 237/366 64.8 Reference
G 64/172 37.2 129/366 35.2 0.6582
1 by Chi-square test.
Int. J. Environ. Res. Public Health 2021, 18, 7006 7 of 11

Table 4. The frequencies of GnRH-1 (rs6185) SNP among women with serum AMH < 1.2 ng/mL in
varied age groups.

GnRH-1 (rs6185) Age < 35 Years (n = 86) Age ≥ 35 Years (n = 183) p Value 1
Genotype n % n %
GG 25/86 29.1 54/183 29.5 Reference
GC 43/86 50.0 93/183 50.8 0.9966
CC 18/86 20.9 36/183 19.7 0.8387
Recessive
GG 25/86 29.1 54/183 29.5 Reference
GC/CC 61/86 70.9 129/183 70.5 0.9414
Allele
G 93/172 54.1 201/366 54.9 Reference
C 79/172 45.9 165/366 45.1 0.8539
1 by Chi-square test.

Table 5. The frequencies of GnRHR-1 (rs3756159) SNP among women with serum AMH < 1.2 ng/mL
in varied age groups.

GnRHR-1 (rs3756159) Age < 35 Years (n = 86) Age ≥ 35 Years (n = 183) p Value 1
Genotype n % n %
CC 34/86 39.5 44/183 24.0 Reference
CT 32/86 37.2 94/183 51.4 0.0071
TT 20/86 23.3 45/183 24.6 0.1166
Recessive
CC 34/86 39.5 34/183 24.0 Reference
CT/TT 52/86 60.5 139/183 76.0 0.0091
Allele
C 100/172 58.1 182/366 49.7 Reference
T 72/172 41.9 184/366 50.3 0.0732
1 by Chi-square test.

3.2. The Clinical Characteristics of Patients with Varied Genotypes of GnRHR SNP (rs3756159)
Table 6 demonstrated the patients’ clinical parameters with the varied genotypes of
GnRHR SNP (rs3756159). Among the clinical parameters related to ART cycles, only the
baseline LH levels (3.60 (2.45 to 5.40) vs. 4.40 (2.91 to 6.48)) are different between the CC
and CT/TT genotypes of GnRHR SNP (rs3756159).

Table 6. The demographic and ovarian stimulation characteristics of infertile woman in GnRHR
(rs3756159) SNP with CC vs. CT/TT genotype. The data are presented with median (25% to 75%).

GnRHR GnRHR
Characteristics of
rs3756159 CC rs3756159 CT/TT p Value
Patients
n = 78 n = 191
Woman age (years) 36.0 (34.0 to 41.0) 38.0 (35.0 to 41.0) 0.0513
Duration of infertility
3.0 (2.0 to 4.0) 3.0 (1.5 to 5.0) 0.5970
(years)
Baseline AMH (ng/mL) 0.60 (0.43 to 0.94) 0.60 (0.28 to 0.90) 0.2619
Baseline FSH (IU/L) 7.60 (4.76 to 9.50) 7.40 (5.35 to 10.38) 0.6122
Baseline LH (IU/L) 3.60 (2.45 to 5.40) 4.40 (2.91 to 6.48) 0.0308
Baseline E2 (ng/mL) 37.0 (25.0 to 59.0) 37.0 (22.0 to 66.5) 0.8942
E2 on Day of trigger
775.5 (485.0 to 1267.0) 823.0 (524.8 to 1208.0) 0.4533
(ng/mL)
P4 on Day of trigger
0.66 (0.37 to 0.90) 0.66 (0.46 to 1.02) 0.3235
(pg/mL)
 Int. J. Environ. Res. Public Health 2021, 18, 7006 8 of 11

Table 6. Cont.

GnRHR GnRHR
Characteristics of
rs3756159 CC rs3756159 CT/TT p Value
Patients
n = 78 n = 191
Day of stimulation (days) 14 (13 to 15) 14 (13 to 15) 0.8692
Number of retrieved
4 (3 to 7) 4 (3 to 6) 0.4028
oocytes
Number of mature
3 (2 to 6) 3 (2 to 5) 0.2512
oocytes
Number of Day3
3 (2 to 5) 3 (2 to 5) 0.3711
embryos
Day3 good embryo rate
70.8 (50.0 to 100.0) 66.7 (50.0 to 100.0) 0.4837
(%)

Figure 5 revealed the baseline LH levels in POSEIDON group 3 and 4 patients divided
by GnRHR SNP (rs3756159) genotypes (CC vs. CT/TT). For POSEIDON group 3 patients,
the baseline LH levels are 3.40 (2.40 to 5.30) in the CC vs. 4.40 (2.85 to 6.46) in the CT/TT
genotypes
Int. J. Environ. Res. Public Health 2021, 18, x (p = 0.095 by Mann–Whitney U test). Furthermore, the baseline LH levels 9 ofare
12
3.65 (2.70 to 5.90) in the CC vs. 4.50 (2.93 to 6.48) in the CT/TT genotypes (p= 0.152 by
Mann–Whitney U test) in POSEIDON group 4 patients.

Figure 5. The baseline LH levels for patients with various GnRHR rs3756159 genotypes. There was
Figure 5. The baseline LH levels for patients with various GnRHR rs3756159 genotypes. There was
nosignificant
no significantdifference
differenceof
ofthese
theseLH
LHlevels
levelsbetween
betweenCC
CCvs.
vs.CT/TT
CT/TTgenotypes
genotypesininyoung
young(POSEIDON
(POSEIDON
group3,3,p p= =
group 0.095)
0.095) or or advanced
advanced ageage (POSEIDON
(POSEIDON group
group p = 0.152)
4, p4,= 0.152) patients
patients by Mann–Whitney
by Mann–Whitney U
U test.
test.

4. Discussion
4. Discussion
In the present study, we found that the young patients with low AMH levels (PO-
In the present study, we found that the young patients with low AMH levels (PO-
SEIDON group 3) are associated with a higher frequency of wild-type CC of GnRHR
SEIDON group 3) are associated with a higher frequency of wild-type CC of GnRHR
(rs3756159). Interestingly, we also noted a lower baseline of serum LH in patients with
(rs3756159). Interestingly, we also noted a lower baseline of serum LH in patients with
CC GnRHR (rs3756159). The results indicated that GnRHR SNP rs3756159 distributed
CC GnRHR (rs3756159). The results indicated that GnRHR SNP rs3756159 distributed
significantly differently between POSEIDON group 3 and group 4 patients and might
significantly differently between POSEIDON group 3 and group 4 patients and might
modulate the ovarian responses using GnRHa or GnRHanta protocols.
modulate the ovarian responses using GnRHa or GnRHanta protocols.
The primary physiological function of GnRH-GnRHR signaling is to release FSH and
LH The
fromprimary physiological
the pituitary gland. Thefunction of GnRH-GnRHR
high baseline serum LH signaling is with
in patients to release FSH
the CT/TT
and LH from
GnRHR the pituitary
(rs3756159) gland.
genotype The high
indicated baseline
that serum(rs3756159)
the GnRHR LH in patients
mightwith the CT/TT
modulate the
GnRHR (rs3756159) genotype indicated that the GnRHR (rs3756159) might
function of GnRH on the target organs or tissues. The GnRHR (rs3756159) is locatedmodulate theat
function of GnRHuntranslated
the 50 upstream on the target organs
region or tissues.
of the GnRHRThe GnRHR
gene. (rs3756159)
How the is located
genetic variants at
affect
the 5′ upstream untranslated region of the GnRHR gene. How the genetic variants affect
the protein structure or function of GnRHR deserves further investigation. Interestingly,
our previous report also showed higher LH levels in POSEIDON group 4 than those in
POSEIDON group 3 for patients with GnRHa or GnRHant protocols [10]. Both reports
Int. J. Environ. Res. Public Health 2021, 18, 7006 9 of 11

the protein structure or function of GnRHR deserves further investigation. Interestingly,

our previous report also showed higher LH levels in POSEIDON group 4 than those in
POSEIDON group 3 for patients with GnRHa or GnRHant protocols [10]. Both reports
indicated that the high LH levels in POSEIDON group 4 patients might correlate with the
genetic variants of GnRHR (rs3756159). Furthermore, such different LH levels might affect
follicular and embryo development [27–29].
The use of LH-containing agents, such as human menopausal gonadotropins (HMG),
could improve the live birth rate for POSEIDON group 3 and 4 patients [27]. Nonetheless,
the supplement of recombinant LH for IVF patients is beneficial for women 36–39 years
of age but not young (<35 years) normal responders in a recent systemic review [28].
Furthermore, the GnRHa protocol is associated with a deeper suppression of LH, the
supplementary of recombinant LH is no benefit for young patients in that meta-analysis [28].
It seems that the benefit of high serum LH levels or supplementing recombinant LH is only
exhibited among women >35 years of age. By contrast, for women <35 years, the GnRHant
protocol is associated with a higher rate of cancellation in our previous report for poor
responders [10] and a large retrospective analysis by Grow et. al. in 2014 for women with a
good prognosis [29].
The GnRHR (rs3756159) features a modulation effect of LH release from the pituitary
gland in the present study. Although the GnRHR (rs3756159) CT/TT genotype is associated
with a higher baseline serum LH levels and more common in POSEIDON group 4 patients,
the benefit of high LH levels is demonstrated for women 36–39 years of age, in other words,
the POSEIDON group 4 patients. These might partially explain why the GnRHa protocol’s
performance was better than the GnRHant protocol for POSEIDON group 3 patients, but
the efficacy of these two protocols is almost equal for POSEIDON group 4 patients in our
previous report [10].
The limitation of the present study includes a relatively small sample size, and we did
not recruit patients with GnRHant protocol. The baseline LH levels are higher in patients
with GnRHR (rs3756159) CT/TT genotype than those with CC genotype either POSEI-
DON group 3 or group 4 patients, but the difference did not reach statistical significance
(Figure 1). Nonetheless, the effect of GnRHR genotype on baseline LH levels is exhibited be-
fore the commence of controlled ovarian stimulation, no matter which stimulation protocol
would be used.
The FSHR SNPs, including A2039G (rs6166), are the most common studied genetic
variants related to ovarian response in ART treatment [16–21]. A recent meta-analysis
indicated that FRHR rs6166 is associated with premature ovarian insufficiency (POI) in an
Asian population [20]. In the present study, although most of the population is younger
than the age of 40 (for the definition of POI), the reference group consists of poor responders
< 35 years of age instead of normal responders. Furthermore, the ovarian dysfunction in
the patients we recruited for this prospective study is not as severe as those in POI patients.
That may explain why FSHR rs6166 is not different between POSEIDON group 3 and group
4 patients in the present study.

5. Conclusions
The genetic variants at GnRHR (rs3756159) distributed differently between POSEI-
DON group 3 and group 4 patients. The CC phenotype is more common in POSEIDON
group 3 patients than in POSEIDON group 4 patients. Furthermore, the CC phenotype
is associated with lower LH levels compared with the CT/TT phenotype. The GnRHR
(rs3756159) may modulate the ovarian responses using GnRHa or GnRHant protocols.
The effect of GnRHR (rs3756159) on ovarian responses and embryo development deserves
further investigation.
Int. J. Environ. Res. Public Health 2021, 18, 7006 10 of 11

Supplementary Materials: The data supplemental to this prospective study is available online at
https://www.mdpi.com/article/10.3390/ijerph18137006/s1, File S1: SNP_low_AMH_single.txt.
Author Contributions: Conceptualization, S.-L.W., T.-H.L. and M.-S.L.; methodology, T.-H.L. and
S.-F.Y.; validation, S.-L.T., C.-C.H. and C.-I.L.; formal analysis, S.-L.W. and T.-H.L.; data curation,
C.-C.H.; writing—original draft preparation, C.-I.L.; writing—review and editing, C.-H.L.; project
administration, S.-L.T.; funding acquisition, T.-H.L. All authors have read and agreed to the published
version of the manuscript.
Funding: This research was funded by the Hsinchu MacKay Memorial Hospital, grant number
CSMU-HCMMH-107-02.
Institutional Review Board Statement: The study was conducted according to the guidelines of the
Declaration of Helsinki and approved by the Institutional Review Board of Chung Shan Medical
University Hospital (CS13194 and CS2-14033).
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: The data presented in this study are available in Supplementary Materials.
Acknowledgments: We acknowledge administrative and technical support by Hui-Mei Tsao and
En-Hui Cheng.
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design
of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or
in the decision to publish the results.

References
1. Oudendijk, J.; Yarde, F.; Eijkemans, M.; Broekmans, F.; Broer, S. The poor responder in IVF: Is the prognosis always poor? A
systematic review. Hum. Reprod. Update 2011, 18, 1–11. [CrossRef] [PubMed]
2. Drakopoulos, P.; Bardhi, E.; Boudry, L.; Vaiarelli, A.; Makrigiannakis, A.; Esteves, S.C.; Tournaye, H.; Blockeel, C. Update on the
management of poor ovarian response in IVF: The shift from Bologna criteria to the Poseidon concept. Ther. Adv. Reprod. Heal.
2020, 14, 2633494120941480. [CrossRef]
3. Zhang, Y.; Zhang, C.; Shu, J.; Guo, J.; Chang, H.-M.; Leung, P.C.K.; Sheng, J.-Z.; Huang, H. Adjuvant treatment strategies in
ovarian stimulation for poor responders undergoing IVF: A systematic review and network meta-analysis. Hum. Reprod. Updat.
2020, 26, 247–263. [CrossRef] [PubMed]
4. Haahr, T.; Dosouto, C.; Alviggi, C.; Esteves, S.C.; Humaidan, P. Management Strategies for POSEIDON Groups 3 and 4. Front.
Endocrinol. 2019, 10, 614. [CrossRef] [PubMed]
5. Ferraretti, A.P.; La Marca, A.; Fauser, B.C.J.M.; Tarlatzis, B.; Nargund, G.; Gianaroli, L.; on behalf of the ESHRE working group
on Poor Ovarian Response Definition. ESHRE consensus on the definition of ’poor response’ to ovarian stimulation for in vitro
fertilization: The Bologna criteria. Hum. Reprod. 2011, 26, 1616–1624. [CrossRef] [PubMed]
6. Alviggi, C.; Andersen, C.Y.; Buehler, K.; Conforti, A.; De Placido, G.; Esteves, S.C.; Fischer, R.; Galliano, D.; Polyzos, N.P.;
Sunkara, S.K.; et al. A new more detailed stratification of low responders to ovarian stimulation: From a poor ovarian response
to a low prognosis concept. Fertil. Steril. 2016, 105, 1452–1453. [CrossRef]
7. Pandian, Z.; McTavish, A.R.; Aucott, L.; Hamilton, M.P.; Bhattacharya, S. Interventions for ’poor responders’ to controlled ovarian
hyper stimulation (COH) in in-vitro fertilisation (IVF). Cochrane Database Syst. Rev. 2010, CD004379. [CrossRef]
8. Sunkara, S.K.; Coomarasamy, A.; Faris, R.; Braude, P.; Khalaf, Y. Long gonadotropin-releasing hormone agonist versus short
agonist versus antagonist regimens in poor responders undergoing in vitro fertilization: A randomized controlled trial. Fertil.
Steril. 2014, 101, 147–153. [CrossRef] [PubMed]
9. Revelli, A.; Chiadò, A.; Dalmasso, P.; Stabile, V.; Evangelista, F.; Basso, G.; Benedetto, C. “Mild” vs. “long” protocol for controlled
ovarian hyperstimulation in patients with expected poor ovarian responsiveness undergoing in vitro fertilization (IVF): A large
prospective randomized trial. J. Assist. Reprod. Genet. 2014, 31, 809–815. [CrossRef] [PubMed]
10. Huang, M.-C.; Tzeng, S.-L.; Lee, C.-I.; Chen, H.-H.; Huang, C.-C.; Lee, T.-H.; Lee, M.-S. GnRH agonist long protocol versus
GnRH antagonist protocol for various aged patients with diminished ovarian reserve: A retrospective study. PLoS ONE 2018, 13,
e0207081. [CrossRef]
11. Prapas, Y.; Petousis, S.; Dagklis, T.; Panagiotidis, Y.; Papatheodorou, A.; Assunta, I.; Prapas, N. GnRH antagonist versus long
GnRH agonist protocol in poor IVF responders: A randomized clinical trial. Eur. J. Obstet. Gynecol. Reprod. Biol. 2013, 166, 43–46.
[CrossRef] [PubMed]
12. Nelson, S.M. Biomarkers of ovarian response: Current and future applications. Fertil. Steril. 2013, 99, 963–969. [CrossRef]
[PubMed]
13. La Marca, A.; Sunkara, S.K. Individualization of controlled ovarian stimulation in IVF using ovarian reserve markers: From
theory to practice. Hum. Reprod. Update 2014, 20, 124–140. [CrossRef] [PubMed]
Int. J. Environ. Res. Public Health 2021, 18, 7006 11 of 11

14. Čuš, M.; Vlaisavljević, V.; Repnik, K.; Potočnik, U.; Kovačič, B. Could polymorphisms of some hormonal receptor genes, involved
in folliculogenesis help in predicting patient response to controlled ovarian stimulation? J. Assist. Reprod. Genet. 2019, 36, 47–55.
[CrossRef] [PubMed]
15. Chen, D.; Zhu, X.; Wu, J. Can polymorphisms of AMH/AMHR2 affect ovarian stimulation outcomes? A systematic review and
meta-analysis. J. Ovarian Res. 2020, 13, 103. [CrossRef] [PubMed]
16. Alviggi, C.; Conforti, A.; Santi, D.; Esteves, S.C.; Andersen, C.Y.; Humaidan, P.; Chiodini, P.; De Placido, G.; Simoni, M. Clinical
relevance of genetic variants of gonadotrophins and their receptors in controlled ovarian stimulation: A systematic review and
meta-analysis. Hum. Reprod. Update 2018, 24, 599–614. [CrossRef] [PubMed]
17. Conforti, A.; Vaiarelli, A.; Cimadomo, D.; Bagnulo, F.; Peluso, S.; Carbone, L.; Di Rella, F.; De Placido, G.; Ubaldi, F.M.; Huhtaniemi,
I.; et al. Pharmacogenetics of FSH Action in the Female. Front. Endocrinol. 2019, 10, 398. [CrossRef] [PubMed]
18. Sindiani, A.M.; Batiha, O.; Al-Zoubi, E.; Khadrawi, S.; Alsoukhni, G.; Alkofahi, A.; Alahmad, N.A.; Shaaban, S.; Alshdaifat, E.;
Abu-Halima, M. Association of single-nucleotide polymorphisms in the ESR2 and FSHR genes with poor ovarian response in
infertile Jordanian women. Clin. Exp. Reprod. Med. 2021, 48, 69–79. [CrossRef]
19. Song, D.; Huang, X.-L.; Hong, L.; Yu, J.-M.; Zhang, Z.-F.; Zhang, H.-Q.; Sun, Z.-G.; Du, J. Sequence variants in FSHR and CYP19A1
genes and the ovarian response to controlled ovarian stimulation. Fertil. Steril. 2019, 112, 749–757.e2. [CrossRef]
20. Huang, W.; Cao, Y.; Shi, L. Effects of FSHR polymorphisms on premature ovarian insufficiency in human beings: A meta-analysis.
Reprod. Biol. Endocrinol. 2019, 17, 80. [CrossRef]
21. Van der Gaast, M.H.; Beckers, N.G.; Beier-Hellwig, K.; Beier, H.M.; Macklon, N.S.; Fauser, B.C. Ovarian stimulation for IVF and
endometrial receptivity—The missing link. Reprod BioMed. Online 2002, 5, 36–43. [CrossRef]
22. Valkenburg, O.; Uitterlinden, A.; Piersma, D.; Hofman, A.; Themmen, A.; De Jong, F.; Fauser, B.; Laven, J. Genetic polymorphisms
of GnRH and gonadotrophic hormone receptors affect the phenotype of polycystic ovary syndrome. Hum. Reprod. 2009, 24,
2014–2022. [CrossRef] [PubMed]
23. Chen, W.-Y.; Du, Y.-Q.; Guan, X.; Zhang, H.-Y.; Liu, T. Effect of GnRHR polymorphisms on in vitro fertilization and embryo
transfer in patients with polycystic ovary syndrome. J. Hum. Genet. 2017, 62, 1065–1071. [CrossRef] [PubMed]
24. Wu, C.-H.; Yang, S.-F.; Tsao, H.-M.; Chang, Y.-J.; Lee, T.-H.; Lee, M.-S. Anti-Müllerian Hormone Gene Polymorphism is Associated
with Clinical Pregnancy of Fresh IVF Cycles. Int. J. Environ. Res. Public Health 2019, 16, 841. [CrossRef] [PubMed]
25. Chung, T.-T.; Pan, M.-S.; Kuo, C.-L.; Wong, R.-H.; Lin, C.-W.; Chen, M.-K.; Yang, S.-F. Impact of RECK gene polymorphisms
and environmental factors on oral cancer susceptibility and clinicopathologic characteristics in Taiwan. Carcinogenesis 2011, 32,
1063–1068. [CrossRef] [PubMed]
26. Tsai, H.-T.; Hsin, C.-H.; Hsieh, Y.-H.; Tang, C.-H.; Yang, S.-F.; Lin, C.-W.; Chen, M.-K. Impact of Interleukin-18 Polymorphisms
-607A/C and -137G/C on Oral Cancer Occurrence and Clinical Progression. PLoS ONE 2013, 8, e83572. [CrossRef] [PubMed]
27. Berker, B.; Şükür, Y.E.; Özdemir, E.; Özmen, B.; Sönmezer, M.; Atabekoğlu, C.S.; Aytaç, R. Human Menopausal Gonadotropin
Commenced on Early Follicular Period Increases Live Birth Rates in POSEIDON Group 3 and 4 Poor Responders. Reprod Sci.
2021, 28, 488–494. [CrossRef] [PubMed]
28. Alviggi, C.; Conforti, A.; Esteves, S.C.; Andersen, C.Y.; Bosch, E.; Bühler, K.; Ferraretti, A.P.; De Placido, G.; Mollo, A.;
Fischer, R.; et al. Recombinant luteinizing hormone supplementation in assisted reproductive technology: A systematic review.
Fertil. Steril. 2018, 109, 644–664. [CrossRef]
29. Grow, D.; Kawwass, J.F.; Kulkarni, A.D.; Durant, T.; Jamieson, D.J.; Macaluso, M. GnRH agonist and GnRH antagonist protocols:
Comparison of outcomes among good-prognosis patients using national surveillance data. Reprod. Biomed. Online 2014, 29,
299–304. [CrossRef]