Article Poor ovarian response to gonadotrophin stimulation is associated with FSH receptor polymorphism

Transcription

1 RBMOnline - Vol 18 No Reproductive BioMedicine Online; on web 12 February 2009 Article Poor ovarian response to gonadotrophin stimulation is associated with FSH receptor polymorphism Ms Swati Achrekar obtained her MSc in 1997 from the University of Mumbai, India. She is currently a PhD scholar at the National Institute for Research in Reproductive Health (Indian Council of Medical Research), Mumbai, India. She has been working in the area of follicle stimulating hormone receptor and her main focus is on studying the significance of FSH receptor polymorphisms. She is a recipient of the Professor GP Talwar medal awarded by the Indian Society for the study of Reproduction and Fertility, for the year Ms Swati Achrekar Swati K Achrekar 1, Deepak N Modi 1, Sadhana K Desai 2, Vijay S Mangoli 2, Ranjana V Mangoli 2, Smita D Mahale 1,3 1 National Institute for Research in Reproductive Health, J M Street, Parel, Mumbai , India; 2 Fertility Clinic and IVF Centre, 101/102, Shanti Niketan, Vachha Gandhi Road, Gamdevi, Mumbai, India 3 Correspondence: Tel: ; Fax: ; smitamahale@hotmail.com Abstract Similarities in the phenotype observed in women with FSH receptor mutation and in FSH receptor knockout mice have clearly established a critical role of this protein in normal gonadal function. Two common single nucleotide polymorphisms in the exonic region of the FSH receptor gene have been shown to be associated with altered ovarian response in subjects undergoing gonadotrophin treatment. Recent in-vitro studies have shown that the A allele at the 29 position in the 5 untranslated region of the FSH receptor gene is associated with impaired transcriptional activity. Differential expression of the FSH receptor and its function may be one of the factors responsible for altered ovarian response. These observations prompted a study of the association between FSH receptor genotype at the 29 position and ovarian response in women undergoing gonadotrophin treatment. Analysis of the data revealed that the subjects with AA genotype at the 29 position required the highest amount of exogenous FSH for ovulation induction, and oestradiol concentrations before the day of human chorionic gonadotrophin administration were significantly lower (P = 0.015) compared with the GA genotype. The number of pre-ovulatory follicles and retrieved oocytes were lowest in the subjects with AA genotype. These results indicate that the AA genotype at position 29 may be associated with the poor ovarian response. Keywords: FSH, FSH receptor, FSH receptor polymorphism, ovarian response, ovulation induction Introduction Approximately 20% of women undergoing ovarian stimulation in an IVF programme respond poorly to gonadotrophin treatment (Keay et al., 1997). Such patients show low concentrations of serum oestradiol, fewer mature oocytes and reduced pregnancy rates. The basis for low response to gonadotrophin is not well understood. Diminished ovarian reserve and increased maternal age may be associated with poor ovarian response (Kligman and Rosenwaks, 2001). Several parameters such as poor follicular flow (Battaglia et al., 2000), dysfunction of the growth factor network (Salmassi et al., 2005), presence of ovarian antibody (Luborsky et al., 2002) and day 3 serum FSH concentrations (Balasch et al., 1996; Barnhart and Osheroff, 1998) have been proposed as predictors of ovarian response, but have not been proven. In some poor responders, increasing the dose of FSH may not help in achieving an increase in serum oestradiol concentrations (Land et al., 1996). The expression of FSH receptor and its ability to respond to exogenous FSH seems to be a determinant of gonadotrophin treatment. Thus, altered FSH receptor expression and function seems to be a factor and may account for poor ovarian response. A number of naturally occurring inactivating mutations in the FSH receptor gene have been reported in infertile females (Simoni et al., 1997; Themmen and Huhtaniemi, 2000). Specific mutations in the FSH receptor gene are associated with altered receptor expression or trafficking. There is a good correlation between the phenotype and the degree of receptor inactivation. Besides inactivating mutations, Published by Reproductive Healthcare Ltd, Duck End Farm, Dry Drayton, Cambridge CB23 8DB, UK

2 510 a large number of single nucleotide polymorphisms have been identified in the FSH receptor. Most of these polymorphisms are located in the introns: five in the coding region and two in the promoter region (Simoni et al., 2002; Wunsch et al. 2005). Among these, A 978 G and T 2039 C polymorphisms have been well studied and have been referred to in the literature as Thr 307 Ala and Asn 680 Ser polymorphisms respectively. The same nomenclature is being used in this paper with respect to these two polymorphisms. Recently it has been demonstrated that lower expression of the FSH receptor may account for poor ovarian response to gonadotrophin stimulation, suggesting a critical role of FSH receptor in the ovarian response (Cai et al., 2007). Single nucleotide polymorphism (SNP) (G/A) in the 5 untranslated region (UTR) of the FSH receptor gene at the 29 position has been shown to be associated with the altered transcriptional activity (Nakayama et al., 2006). In-vitro studies have demonstrated that the transcriptional activity of the A allele was significantly less than the G allele. Further, serum oestradiol values were significantly lower in patients with AA genotype than in the patients with GG or GA genotypes at the 29 position. Since the expression of the FSH receptor is impaired in poor responders and the polymorphism at 29 position of the FSH receptor gene is associated with transcriptional activity, it was postulated that the AA genotype may be associated with poor responsiveness. This was tested by analysing the SNP at the 29 position of the FSH receptor gene and studying the ovarian response in subjects undergoing ovarian stimulation in an assisted reproductive technology (ART) programme. Materials and methods Study subjects The present study was approved by the institutional ethics committee for clinical research. All subjects included in the present study were of Indian ethnicity and informed consent was obtained from them. A total of 50 normogonadotrophic ovulatory females with infertility because of male or tubal factor were retrospectively analysed and are referred to as ART subjects in the subsequent discussion. The control group consisted of 100 proven fertile healthy females with a history of regular menstrual cycles. Ovulation induction A standard protocol (Peter, 2004) of ovulation induction was employed. The exogenous FSH to be used for attaining optimal ovarian response was decided based on serum oestradiol concentrations and ultrasonography findings. The number of pre-ovulatory follicles, number of retrieved oocytes, rate of fertilization and incidence of clinical pregnancy were recorded for all subjects. Hormone assays To investigate the hormonal profiles in subjects recruited for the ART programme, basal FSH concentrations on day 3 of the menstrual cycle were measured in one of the cycles before ovarian stimulation. Serum concentrations of oestradiol and progesterone were estimated before and on the day of human chorionic gonadotrophin (HCG) administration. Serum concentrations of FSH, LH and oestradiol concentrations were measured by chemiluminescence microparticle assay (Abbott Architect, Abbott Park, IL, USA). Progesterone concentrations were estimated by chemiluminescence immunoassay (Immulite 2500; Siemens, Golden, CO, USA). DNA isolation and polymorphism analysis Genomic DNA was extracted from 250 µl whole blood using a commercial kit (Axygen Scientific, New Delhi, India) according to the manufacturer s instructions. The region of the FSH receptor gene from nucleotides 74 to 162 was amplified by polymerase chain reaction (PCR). Genomic DNA was used as the template and the set of primers 5 -GGA GCT TCT GAG ATC TGT GG-3 (forward primer) and 5 -AAA TGC CAG CCA TGC AGT TG-3 (reverse primer) were used to get an amplified product of 235 base pairs (bp). The G 29 A polymorphism was screened by restriction enzyme analysis using MboII(Promega, Madison, WI, USA) in all the subjects. The recognition site of MboII is GAAGA (8/7). The presence of G at the 29 position of the FSH receptor gene introduces a restriction site, eight bases ahead of the recognition site of this restriction enzyme. This is exploited in the restriction fragment length polymorphism (RFLP) analysis. The PCR amplification was carried out using PCR core system (Promega, Madison, USA). A 25 ng quantity of genomic DNA was used in a 25 µl reaction volume, consisting of dntps (200 mmol/l), buffer (1 concentration), Taq polymerase (1.5 IU), MgCl 2 (1.25 mmol/l) and primers (10 pmol per reaction). PCR was carried out using thermal cycler (M J Research, Waltham, USA). PCR conditions comprised of an initial denaturation step of 95 C for 2 min; followed by 40 cycles each of denaturation at 95 C for 1 min, annealing at 58 C for 50 s and extension at 72 C for 30 s, and a final extension step at 72 C for 10 min. A 5 µl volume of each reaction product was resolved by 10% polyacrylamide gel electrophoresis (PAGE) and silver stained. Restriction digestion was carried out for 20 h at 37 C using 5 IU of the enzyme and 5 µl PCR product in a total volume of 20 µl of 1 reaction buffer (Promega). The products were separated using 10% PAGE and visualized by silver staining. The appropriate mass ladder compatible for PAGE (Promega) was used to determine the size of the products. The specificity and validity of the RFLP were confirmed by direct sequencing of PCR products from different individuals. At least five samples of each variant were selected based on RFLP analysis for sequencing. Fluorescence-based sequencing was done at the sequencing core facility of the National Institute for Research in Reproductive Health. Statistical analysis Statistical analysis of the results was performed by one-way analysis of variance (ANOVA) and chi-squared analysis. The former was used to analyse the differences in clinical parameters

3 between the study groups, while chi-squared analysis was used to analyse the data on the outcome of clinical pregnancy. All statistical analysis was performed using Statistics Package for Social Sciences (SPSS, USA) software. A P-value of less than 0.05 was considered statistically significant. Results SNP analysis RFLP analysis of the SNP at position 29 of the FSH receptor gene revealed three different patterns (Figure 1A). The undigested fragment indicates the AA genotype and had a size of 235 bp, whereas the digested fragments indicate the GG genotype and resulted in 181 bp and 54 bp fragments. The presence of all three fragments indicates the GA genotype. RFLP results were confirmed by DNA sequencing in randomly selected samples. The DNA sequence of the SNP at position 29 of the FSH receptor is shown in Figure 1B. Frequency distribution The genotypic frequency for the SNP at position 29 of the FSH receptor in the control subjects was 5% for the GG genotype, 1% for the AA genotype and 94% for the GA genotype. In the subjects undergoing ART, the frequency distribution for the polymorphism at this position was 32% for the GG genotype, 14% for the AA genotype and 54% for the GA genotype. The distribution of the SNP at position 29 of the FSH receptor gene differed significantly between the controls and the subjects undergoing the ART programme (P < ) (Figure 2). Clinical and endocrine parameters To study the potential association between the SNP at position 29 of the FSH receptor gene and the ovarian response during ovarian stimulation, clinical, endocrine and ultrasonographic parameters were recorded for the subjects undergoing ART, with different genotypes at position 29 of the FSH receptor gene. The age of the subjects, basal FSH and LH concentrations, peak oestradiol concentrations on the day of HCG administration, and progesterone concentrations before and on the day of HCG administration, did not show any statistically significant differences among the three groups (data not shown). In contrast to the above parameters, the amount of exogenous FSH required for ovulation induction, the peak oestradiol concentrations before the day of HCG administration, the number of pre-ovulatory follicles and the number of retrieved oocytes required for successful treatment were significantly different among the three genotypes (Figures 3 and 4). The maximum amount of the exogenous FSH ( ± 194.2) was required for the subjects with AA genotype at position 29, whereas the subjects with GA genotype required the minimum amount of the exogenous FSH ( ± 141.3) (P = 0.001). The mean amount of exogenous FSH required for ovulation induction in the heterozygous cases was higher ( ± Figure 1. (A) Restriction fragment length polymorphism analysis of the single nucleotide polymorphism at position 29 of FSH receptor. Lanes: 1, 20 base pair (bp) ladder; 2, undigested polymerase chain reaction (PCR) product of 235 bp; 3 5, digested PCR products. Lane 3: two fragments of 181 bp and 54 bp indicate GG genotype. Lane 4: 235 bp fragment indicates AA genotype. Lane 5: three fragments of 235 bp, 181 bp and 54 bp indicate GA genotype. (B) Electropherogram of 3 5 strand showing corresponding complimentary sequence. C, T and N, indicated by an arrow, correspond to G, A and N in the 5 3 strand and represent GG, AA and GA genotypes at position 29 position, respectively. 511

4 Figure 2. Distribution of the single nucleotide polymorphism at position 29 of the FSH receptor gene in controls (n = 100) and assisted reproductive treatment (ART) subjects. The frequency distribution between the two groups was significantly different for all three genotypes (P < ). Figure 3. FSH stimulation and oestradiol concentrations grouped according to the single nucleotide polymorphism at position 29 of the FSH receptor gene. The number of subjects with GG, GA and AA were 16, 27 and 7 respectively. Values are expressed as means ± SEM. Different letters (a and b) indicate statistically significant differences. (A) Exogenous FSH required for ovarian stimulation in assisted reproductive treatment (ART) subjects. Values were significantly different (P = 0.001) among GG and GA genotypes, and GG and AA genotypes. (B) Oestradiol concentrations before human chorionic gonadotrophin (HCG) administration in ART subjects. Values were significantly different (P = 0.015) among GG and AA genotypes. 512 Figure 4. Ultrasonographic findings of the assisted reproductive treatment (ART) subjects grouped according to the single nucleotide polymorphism at position 29 of the FSH receptor gene. The number of subjects with GG, GA and AA were 16, 27 and 7 respectively. Values are expressed as means ± SEM. Different letters (a, b and c) indicate statistically significant differences. (A) Number of preovulatory follicles in ART subjects. Values were significantly different (P = 0.001) among GG, GA and AA genotypes. (B) Number of retrieved oocytes in ART subjects. Values were significantly different (P = 0.003) among GG, GA and AA genotypes.

5 184.84) than that required for the subjects with GG genotype (P = 0.001; Figure 3A). When oestradiol concentrations before the day of HCG administration were compared, it was observed that the subjects with AA genotype had significantly (P = 0.015) lower concentrations of oestradiol ( ± 78.05) as compared with subjects with GG genotype ( ± ). Interestingly, in the subjects with GA genotype, oestradiol concentrations were intermediate (2343 ± 209.8) compared with those observed in the subjects with GG and AA genotypes (Figure 3B). Next, the ultrasonographic findings of the subjects undergoing ART were compared. The subjects with AA genotype had a lower number of pre-ovulatory follicles (6.57 ± 0.57); whereas, in the case of the subjects with GG genotype, the number of pre-ovulatory follicles was maximum (14.06 ± 1.10). The subjects with GA genotype had an intermediate number of pre-ovulatory follicles (11.67 ± 0.78). The differences between the three genotypes were statistically significant (P = 0.001; Figure 4A). As expected, the number of retrieved oocytes was lowest in subjects with AA genotype (6.00 ± 1.09); whereas in the subjects with GG genotype, the number of retrieved oocytes was significantly higher (17.88 ± 1.75; P = 0.003). The subjects with GA genotype had an intermediate number of preovulatory follicles (14.22±1.53). Again, differences between the three genotypes were statistically significant (P = 0.003; Figure 4B). In the study group of 50 prospectively selected cases from the ART programme, further analysis of data with respect to clinical parameters revealed that five out of 16 (31%) with GG genotype, 16 out of 27 (59%) with GA genotype, and one out of seven (14%) with AA genotype, were recorded as successful clinical pregnancies. Haplotype analysis The genomic DNA of the subjects undergoing ART in the present study was previously evaluated for the SNP at amino acid positions 307 and 680 of the FSH receptor (Achrekar et al., 2009). In the present study, haplotype analysis of the SNP at nucleotide position 29 and amino acid position 307 of FSH receptor gene was carried out (Table 1). Heterozygosity at both the loci (GA at 29 and Thr Ala at 307) was found to be more prevalent both in controls and ART subjects as compared with other haplotypes. Three of the haplotypes were significantly different in both study groups (P < 0.05). Interestingly, the haplotype AA( 29)-Ala Ala(307) was not observed in either of the study groups. Discussion The correlation of lower expression of the FSH receptor and poor ovarian response to gonadotrophin stimulation (Cai et al., 2007) and the decreased transcription activity of the A allele at position 29 of the FSH receptor gene (Nakayama et al., 2006) prompted this investigation into the association of the polymorphism at position 29 in the 5 UTR of the FSH receptor gene and the ovarian response of the subjects undergoing ovulation induction in an ART programme. First, there was an evaluation of the frequency distribution of the SNP at position 29 of the FSH receptor gene in Indian women, both in the controls as well as in ART subjects. It was observed that the frequency of GA genotypes at position 29 was more prevalent compared with the frequency of GG and AA genotypes, both in controls and in ART subjects (Figure 2). Similar observations have been reported earlier by Nakayama et al. (2006). However, in another study, the frequency of the GA genotype was higher in Indonesian women and the frequency of the GG genotype was higher in German women (Wunsch et al., 2005). In the present study, although the GA genotype was more prevalent in both the groups studied, the frequency distribution varied significantly between the groups. Frequency of the GG genotype was 5% in controls whereas it was 32% in ART subjects. Differences were also observed with respect to the AA genotype, where frequency in the control population was 1% and the frequency in ART subjects was 14%. This kind of comparison of frequency distribution between control and ART subjects has not been previously reported. When the ART subjects were segregated on the basis of the genotypes at position 29 of the FSH receptor gene, it was observed that the ovarian response to FSH stimulation was variable and different amounts of exogenous FSH were required for an optimal response. The amount of exogenous FSH administered to the ART subjects varied from 1450 to 5150 IU to obtain mature follicles. One possible reason for such variability of the exogenous FSH required for a similar response could be due to the differences in the FSH receptor genotype. No variation was observed in other factors such as age, reproductive history and body mass index. Different concentrations of FSH receptor expression in granulosa cells have been attributed to altered ovarian response (Cai et al., 2007). It was observed that the subjects with AA genotype required significantly higher amounts of exogenous FSH for ovulation induction as compared with the subjects with GG and GA Table 1. Frequency distribution of the haplotype (single nucleotide polymorphism at nucleotide positions 29 and 978) of the FSH receptor gene. Single nucleotide polymorphism at 978 corresponds to a change in amino acid of either Thr or Ala at the 307 position of the FSH receptor. Haplotype Number of subjects (%) Controls ART subjects (n = 100) (n = 50) GG( 29)-Thr Thr(307) a 1 14 GG( 29)-Thr Ala(307) a 1 16 GG( 29)-Ala Ala(307) 2 0 GA( 29)-Thr Thr(307) GA( 29)-Thr Ala(307) GA( 29)-Ala Ala(307) AA( 29)-Thr Thr(307) 0 2 AA( 29)-Thr Ala(307) a 1 12 AA( 29)-Ala Ala(307) 0 0 a The frequency distribution between the two groups was significantly different (P < 0.05). ART = assisted reproductive treatment. 513

6 514 genotypes at the 29 position of the FSH receptor gene. These results indicate that the AA genotype is resistant to FSH treatment. Supporting this observation, it was further noted that in the subjects with AA genotypes, the peak serum oestradiol concentrations before the day of HCG administration were significantly lower as compared with the subjects with the GG genotype. The amount of exogenous FSH required for ovulation induction or the concentrations of oestradiol before the day of HCG administration were always intermediate for the subjects with GA genotype (Figure 3). The basal FSH concentrations and the peak oestradiol concentrations on the day of HCG administration did not differ significantly among these genotypes. Similar observations were made by Wunsch et al. (2005). Interestingly, it was found that in the subjects with AA genotype (where the amount of exogenous FSH required was more, and the peak oestradiol concentrations before the day of HCG administration were significantly low), the number of preovulatory follicles and the number of retrieved oocytes were significantly lower when compared with the subjects with GG and GA genotypes (Figure 4). From these results, it is speculated that the AA genotype at position 29 of the FSH receptor gene is probably responsible for poor ovarian response. Such poor responders are likely to show reduced rates of clinical pregnancy (i.e. ultrasound examination at 4 weeks after IVF procedure shows fetal sac in the uterus). Of the 50 ART subjects recruited in this study, 16 were of the GG genotype, 27 were of the GA genotype and seven were of the AA genotype. Among these, five out of 16 subjects (31%) of the GG genotype had a successful clinical pregnancy. Sixteen out of 27 subjects (59%) of the GA genotype had a successful clinical pregnancy, whereas only one subject out of seven (14%) of the AA genotype had a successful clinical pregnancy. These results indicate that the subjects with the AA genotype have a significantly lower rate of successful clinical pregnancy as compared with the GG and GA genotypes (P = 0.047). In proven fertile control subjects the incidence of the AA genotype was only 1%, suggesting a negative correlation between the AA genotype and fertility. Incidentally, subjects with the AA genotype were poor responders as they required the highest amount of exogenous FSH for ovulation induction and their peak oestradiol concentrations before the day of HCG administration were significantly lower. Also, the number of preovulatory follicles and the number of retrieved oocytes in subjects with the AA genotype were significantly lower. Cycle cancellation is a serious problem and a financial burden for the subjects showing poor response during ovarian stimulation. Thus, identification of such poor responders in the standard ART programmes is clinically relevant in improving the treatment, and would reduce the burden of cycle cancellation. The present study, carried out in Indian women undergoing ovarian stimulation in an ART programme, shows a relationship between subjects with the AA genotype at position 29 of the FSH receptor gene and poor ovarian response. However, studies of larger cohorts need to be undertaken to evaluate the significance of SNP analysis at position 29 of the FSH receptor gene in identifying the poor responders. Most of the previous studies have reported a correlation between ovarian response and polymorphism of FSH receptor gene at the amino acid position 680 (Asn 680 Ser). When patients undergoing ovarian stimulation were segregated based on the FSH receptor genotype at position 680, it was observed that the patients with homozygous Ser had high basal FSH concentrations. Also, the number of FSH ampoules required for ovulation induction was significantly more in patients with homozygous Ser polymorphism (Perez Mayorga et al., 2000; Sudo et al., 2002). Results from a study in healthy volunteers and findings from subjects undergoing IVF treatment suggest that granulosa cells of women with homozygous Ser at position 680 are more resistant to FSH action (Greb et al., 2005). These studies demonstrated that the ovarian response to FSH stimulation depends on the FSH receptor genotype. However, the observations were not the same in other studies (Laven et al., 2003; de Castro et al., 2003; d Alva et al., 2005). Klinkert et al. (2006) have reported that, in their study population, Ser 680 Ser polymorphism of the FSH receptor gene has a role in the prediction of clinical pregnancy in IVF but it may not be used as a marker for predicting the ovarian response. Although most of the reported studies have not observed any significance of the SNP at amino acid position 307, in a previous study, the authors observed that the FSH receptor polymorphism Thr 307 Ala is associated with variable ovarian response and ovarian hyperstimulation syndrome in Indian women (Achrekar et al., 2009). In some studies the SNP at amino acid position 680 was identified as a marker for predicting ovarian response, although the significance of the amino acid change at this position could not be evaluated in in-vitro studies (Sudo et al., 2002; Klinkert et al., 2006). The present study provides supporting evidence for this observation, as the subjects with AA genotype at position 29 did not respond well to FSH treatment, possibly due to low transcriptional activity of the FSH receptor. Thus, there is a good association with the observations made in vitro and in vivo. The present observation that the AA genotype at 29 position is associated with poor ovarian response, and previous findings suggesting that homozygous Ala at amino acid position 307 is associated with ovarian hyperstimulation syndrome (Achrekar et al., 2009), prompted us to undertake haplotype analysis of both these positions in the subjects of this study. It was observed that none of the recruited subjects, who were of AA genotype at 29 position of the FSH receptor gene, showed homozygous Ala at amino acid position 307 as expected (Table 1). In conclusion, the results of the present study suggest that ART subjects with the AA genotype at position 29 of the FSH receptor gene appear to be poor responders to gonadotrophin treatment based on their endocrine and clinical parameters. To the best of the authors knowledge, this is the first report describing the association of the AA genotype at position 29 of the FSH receptor gene and poor ovarian response in ART subjects undergoing ovarian stimulation in an ART programme. This novel finding warrants a need to undertake a multicentre study across various populations before this information is used as a marker to identify poor responders. Acknowledgements The research work related to this publication was supported by grants from the National Institute for Research in Reproductive Health, Mumbai (NIRRH/MS/6/2008), and from the Indian Council of Medical Research (63/116/2001-BMS II), New Delhi, India. The authors would like to thank Dr Anurupa Maitra and Mr Chinnaraj Saravanan (DNA Sequencing core facility, NIRRH) for their help in DNA sequencing. The authors

7 acknowledge Dr D Balaiah and Mr Prashant Tapse (Division of Biostatistics, NIRRH) for their help in statistical analysis. Technical help provided by Ms Jignasha Chauhan (Fertility Clinic and IVF centre) is acknowledged. References Achrekar SK, Modi DN, Desai SK et al Follicle-stimulating hormone receptor polymorphism (Thr 307 Ala) is associated with variable ovarian response and ovarian hyper stimulation syndrome in Indian women. Fertility and Sterility 19, Balasch J, Creus M, Fabregues F et al Inhibin, folliclestimulating hormone, and age as predictors of ovarian response in in vitro fertilization cycles stimulated with gonadotropin-releasing hormone agonist-gonadotropin treatment. American Journal of Obstetrics and Gynecology 175, Barnhart K, Osheroff J 1998 Follicle stimulating hormone as a predictor of fertility. Current Opinion in Obstetrics and Gynecology 10, Battaglia C, Genazzani AD, Regnani G et al Perifollicular Doppler flow and follicular fluid vascular endothelial growth factor concentrations in poor responders. Fertility and Sterility 74, Cai J, Lou HY, Dong MY et al Poor ovarian response to gonadotropin stimulation is associated with low expression of follicle-stimulating hormone receptor in granulosa cells. Fertility and Sterility 87, d Alva CB, Serafini P, Motta E et al Absence of folliclestimulating hormone receptor activating mutations in women with iatrogenic ovarian hyperstimulation syndrome. Fertility and Sterility 83, de Castro F, Ruiz R, Montoro L et al Role of follicle-stimulating hormone receptor Ser680Asn polymorphism in the efficacy of follicle-stimulating hormone. Fertility and Sterility 80, Greb RR, Behre HM, Simoni M 2005 Pharmacogenetics in ovarian stimulation current concepts and future options. Reproductive BioMedicine Online 11, Keay SD, Liversedge NH, Mathur RS, Jenkins JM 1997 Assisted conception following poor ovarian response to gonadotropin stimulation. British Journal of Obstetrics and Gynaecology 104, Kligman I, Rosenwaks Z 2001 Differentiating clinical profiles: predicting good responders, poor responders, and hyper responders. Fertility and Sterility 76, Klinkert ER, te Velde ER, Weima S et al FSH receptor genotype is associated with pregnancy but not with ovarian response in IVF. Reproductive BioMedicine Online 13, Land JA, Yarmolinskaya MI, Dumoulin JC, Evers JL 1996 High-dose of human menopausal gonadotropin stimulation in poor responders does not improve in vitro fertilization outcome. Fertility and Sterility 75, Laven JS, Mulders AG, Suryandari DA et al Folliclestimulating hormone receptor polymorphisms in women with normogonadotropic anovulatory infertility. Fertility and Sterility 8, Luborsky JL, Thiruppati P, Rivany B et al Evidence for different aetiologies of low E2 response to FSH: age-related accelerated luteinization of follicles or presence of ovarian autoantibodies. Human Reproduction 17, Nakayama T, Kuroi N, Sano M et al Mutation of the folliclestimulating hormone receptor gene 5'-untranslated region associated with female hypertension. Hypertension 48, Perez Mayorga M, Gromoll J, Behre HM et al Ovarian response to follicle-stimulating hormone (FSH) stimulation depends on the FSH receptor genotype. Journal of Clinical Endocrinology and Metabolism 85, Peter PR 2004 Ovarian follicular stimulation reigns in ART: The Bourn Hall Experience. In: Rao K (ed.) The Infertility Manual. Jaypee (FOGSI) publication, India, pp Salmassi A, Schmutzler AG, Schaefer S et al Is granulocyte colony-stimulating factor level predictive for human IVF outcome? Human Reproduction 20, Simoni M, Nieschlag E, Gromoll J 2002 Isoforms and single nucleotide polymorphisms of FSH receptor gene: implications for human reproduction. Human Reproduction Update 8, Simoni M, Gromoll J, Nieschlag E 1997 The follicle stimulating hormone receptor: biochemistry, molecular biology, physiology, and pathophysiology. Endocrine Review 18, Sudo S, Kudo M, Wada S et al Genetic and functional analyses of polymorphism in the human FSH receptor gene. Molecular Human Reproduction 8, Themmen APN, Huhtaniemi IT 2000 Mutations of gonadotropins and gonadotropin receptors: elucidating the physiology and pathophysiology of pituitary-gonadal function. Endocrine Review 21, Wunsch A, Ahda Y, Banaz-Yaşar F et al Single-nucleotide polymorphisms in the promoter region influence the expression of the human follicle-stimulating hormone receptor. Fertility and Sterility 84, Declaration: The authors report no financial or commercial conflicts of interest. Received 16 April 2008; refereed 10 June 2008; accepted 24 September