Clomiphene citrate versus letrozole: molecular analysis of the endometrium in women with polycystic ovary syndrome

Transcription

1 Clomiphene citrate versus letrozole: molecular analysis of the endometrium in women with polycystic ovary syndrome Kedra L. Wallace, Ph.D., Venessia Johnson, R.N., Victoria Sopelak, Ph.D., and Randall Hines, M.D. Department of Obstetrics & Gynecology, University of Mississippi Medical Center, Jackson, Mississippi Objective: To compare the effect of clomiphene citrate (CC) and letrozole on endometrial receptivity in women with polycystic ovary syndrome (PCOS). Design: A randomized controlled trial. Setting: University teaching hospital. Patient(s): Ten anovulatory women with PCOS and 5 fertile ovulatory women. Intervention(s): Patients received 2.5 mg of letrozole on cycle days 3 7 (5 patients, 1 cycle) or 50 mg of CC on cycle days 5 9 (5 patients, 1 cycle). Main Outcome Measure(s): Serum estrogen (E) and progesterone (P) endometrial protein and messenger RNA (mrna) expression of leukemia inhibitory factor (LIF), dickkhopf homolog 1 (DKK-1), fibroblast growth factor 22 (FGF-22), and endometrial mrna expression of LIF/GP130 receptor (LIFR). Result(s): No statistically significant differences were observed between groups compared with fertile ovulatory women when serum E and P were examined, or between body mass index (BMI), and cycle day at time of biopsy. Letrozole increased mrna expression of LIF, DKK1, LIFR, and FGF-22, whereas CC only increased endometrial mrna expression of LIF. Letrozole mrna expression directly translated into increased protein expression of like genes in the endometrium. The CC protein expression of DKK-1 was significantly decreased compared with controls. Conclusion(s): Letrozole positively influences a number of markers of endometrial receptivity compared with CC. (Fertil Steril Ò 2011;96: Ó2011 by American Society for Reproductive Medicine.) Key Words: Polycystic ovary syndrome, clomiphene citrate, letrozole, leukemia inhibitory factor, angiogenesis Polycystic ovary syndrome (PCOS) is a common cause of infertility that affects up to 6% of reproductive-age women (1). PCOS is associated with irregular menstrual cycles and anovulation, in addition to aberrant changes in insulin and androgen production. Clomiphene citrate (CC) is the most widely used medication for ovulation induction, with ovulation rates of 80% being reported in patients with anovulation. Pregnancy rates (PR) are considerably lower, with estimates ranging from 25% 60% and CC resistance, failure to ovulate after receiving 150 mg/d of CC, occurs in 15% 40% of women with PCOS (2). One possible explanation for the difference between ovulation and PRs could be a deleterious effect of CC on the endometrium (3). Clomiphene citrate is a selective estrogen (E) receptor (ER) modulator that in some patients will have a negative effect on the endometrium. Studies evaluating endometrial thickness reveal variable outcomes, but a decrease in the measured thickness of the endometrium in patients on CC has been reported (4). Letrozole is a newer agent for ovulation induction, first described as a treatment for breast cancer. Letrozole is an aromatase inhibitor and is believed to induce ovulation through transient lowering of E levels, inducing Received February 25, 2011; revised July 9, 2011; accepted July 11, 2011; published online August 18, K.L.W. has nothing to disclose. V.J. has nothing to disclose. V.S. has nothing to disclose. R.H. has nothing to disclose. Reprint requests: Kedra L. Wallace, Ph.D., Department of Obstetrics & Gynecology. University of MS Medical Center, 2500 North State Street, Jackson, MS ( kwallace2@umc.edu). a reflexive increase in FSH. Because letrozole does not bind the ER, letrozole does not appear to have a direct effect on the endometrium. Patients who fail to become pregnant on CC, despite ovulation, may achieve success with the use of letrozole (5). Patients with PCOS who failed to ovulate after CC administration, had an ovulation rate between 54.6% 84.4% after using an aromatase inhibitor (6, 7). There are a number of biochemical pathways and mechanisms regulating implantation and endometrial receptivity. We have previously identified a number of markers of endometrial receptivity that are influenced differently by letrozole and CC (8). In particular leukemia inhibitory factor (LIF), which is decreased in women with PCOS, dickkhopf homolog 1 (DKK-1), and fibroblast growth factor-22 (FGF-22) (9). The DKK-1, an inhibitor of the Wnt signaling, plays an important role in implantation and has recently been found to be up-regulated in PCOS theca cells, suggesting that this pathway is disrupted in PCOS (10). Women with PCOS have an increase in angiogenesis, which is marked by increases in angiogenic factors such as vascular endothelial growth factor and FGFs (11, 12). Fibroblast growth factors are also capable of neovascularization, and have mitogenic effects independent of E, thus providing a potential target of CC and letrozole. With the observation that CC and letrozole work through different mechanisms, and that CC may have a deleterious effect on the endometrium, thus affecting implantation and subsequent pregnancy (13), we studied subjects taking either CC or letrozole to induce ovulation. The aim of this study was designed to compare differences in the endometrial gene expression of LIF, DKK-1, and FGF-22, which have been shown to play a role in implantation /$36.00 Fertility and Sterility â Vol. 96, No. 4, October doi: /j.fertnstert Copyright ª2011 American Society for Reproductive Medicine, Published by Elsevier Inc.

2 MATERIALS AND METHODS Patient Selection, Treatment, and Sample Collection All patients participating in the current study were recruited from the University of MS Medical Center in Jackson, MS. Institutional Review Board approval was obtained. Twenty-three participants were enrolled in this study, but eight were excluded due to failure to ovulate with CC or letrozole. Diagnosis of PCOS was based on the Rotterdam criteria, which included a history of anovulation or oligoovulation (<6 cycles per year, positive withdrawal bleeding to P, and absence of hyperprolactineamia and thyroid dysfunction). Control subjects consisted of women with regular menstrual cycles. Study exclusion was a positive pregnancy test, current use of oral contraceptives (OC) or hormones, a history of CC failure, and not between the ages of 21 and 40 years. Luteal phase endometrial biopsies were obtained from reproductive-age subjects. Biopsies were obtained in an outpatient setting using standard technique. Controls (n ¼ 5; fertile ovulatory women) were compared with women with PCOS who were given either letrozole (n ¼ 5; 2.5 mg on cycle days 3 7) or CC (n ¼ 5; 50 mg on cycle days 3 7). Subjects were given a home urine ovulation detection kit (EZ-LH, Biomerica) and a tutorial on how to use the kit according to the manufacturer s instructions. When women reported with a positive LH surge, they were scheduled for an endometrial biopsy 7 days after the LH surge was detected. Immediately before the biopsy blood was drawn for hormone determination. Estrogen and P levels were measured using an Immunlite 1000 hormone analyzer (Seimens Medical Solutions Diagnostics), the intra- and interassay coefficients of variation (CV) were <3% and <6%, respectively, for E and <6% and <7%, respectively, for P. Biopsied endometrial tissue was immediately placed in RNA Later or 10% buffered formalin solution. Gene Expression RNA was extracted from the endometrium using phenol chloroform (14), 1 mg of total RNA was reverse transcribed into complementary DNA (cdna; Invitrogen). The resulting cdna were subjected to real-time polymerase chain reaction (PCR) with the following gene primers prepared from SA Biosciences: LIF, LIF/glycoprotein130 receptor (LIFR), DKK- 1, and FGF-22. These markers were chosen based on genes shown to be up/down-regulated between the different treatment groups in a human growth factor array (SA Biosciences) (8). Real-time PCR messenger RNA (mrna) semiquantification was performed in 25 ml volume containing the following reagents: 12.5 ml of the iq SYBRgreen Supermix (BioRad Laboratories), 10 mm of respective primer, 1.5 ml cdna, and nuclease-free water. Samples were subjected to an initial step at 95 C for 5 minutes, followed by 40 cycles consisting of 10 seconds at 90 C, 30 seconds at 60 C, followed by 1 cycle at 30 seconds at 95 C. Gel electrophoresis was performed to assess the purity and molecular size of each PCR product. Each reaction was run in triplicate. Mean threshold cycle (Ct) was determined for each transcript normalized to that of b-actin gene in separate reactions. The relative mrna expression levels of the target genes in each sample after normalization were calculated using the 2 DDCt (2 avg. Ct gene of interest avg. Ct b-actin ) calculus method of mrna analysis (15). Immunohistochemical Analysis of Endometrium Endometrial tissue was paraffin embedded and cut to 8 mm for immunohistochemistry. Serial sections of tissue were rehydrated and antigen sites unmasked using Trilogy (CellMarque). After blocking with 5% milk buffer, sections were incubated with the following: goat anti-human DKK-1, LIF (RnD Systems), or FGF-22 (Santa Cruz Biotechnology) overnight at 4 C. After washing, slides were incubated with mouse anti-goat horse radish peroxidase-labeled secondary antibody for 1 hour, followed by detection with diaminobenzidine, and observed and photographed under a phase-contrast microscope. Negative controls were obtained by omitting primary antibodies. Immunohistochemical slides were scored using Image J, which is a javabased image analysis program developed at the US National Institutes of Health that converts digital images to 8, 16, and 32-bit pictures and calculates area and pixel value based on user-defined threshold intensity (16). Two digital images per patient (n ¼ 30) were analyzed for percent area immunostained by Image J software. Statistical Analysis Hormone levels, body mass index (BMI), age, cycle day at time of biopsy, and mrna and protein expression were analyzed using one-way analysis of variance (ANOVA). The c 2 test was used to analyze race between the groups. Data are expressed as mean SD. P values <.05 were considered significant. RESULTS The endometrial tissues obtained from 15 women were all in the luteal phase. There were no significant differences in age, BMI, and cycle date at time of biopsy or hormone measurements between women taking or not taking CC or letrozole (Table 1). An ANOVA was used to compare endometrial expression of LIF, LIFR, DKK-1, and FGF-22 between the different treatment groups. The LIF mrnawas nonsignificantly increased in women in both the CC and letrozole groups compared with women belonging in the control group (n ¼ 15, P¼.891; Fig. 1A). Letrozole increased FGF-22 mrna expression, but failed to reach levels seen in the control population (n ¼ 15, P¼.415; Fig. 1B). The CC mrna expression of FGF-22 was decreased compared with controls and the letrozole treatment group. Letrozole treatment increased DKK-1 mrna expression compared with women in both the control and CC groups (n ¼ 15, P¼.206; Fig. 1C). Amplification of cdna products were detected for each gene examined. The mrna samples were considered to be negative for each respective gene expression TABLE 1 Patient characteristics. Control (N [ 5) Letrozole (N [ 5) Clomiphene citrate (N [ 5) P value Age (y) Race (% African American) 60% 60% 60% 1.00 BMI Cycle day at time of biopsy (d) Circulating P (ng/ml) Circulating estrogen (pg/ml) BMI ¼ body mass index Wallace et al. Endometrial analysis: CC vs. letrozole Vol. 96, No. 4, October 2011

3 FIGURE 1 Effects of clomiphene citrate (CC) or letrozole on the expression of (A) leukemia inhibitory factor (LIF); (B) fibroblast growth factor 22; (C) dickkhopf homolog 1; and (D) LIF receptor messenger RNA expression in the endometrium. Patients per treatment group, N ¼ 5. when there was no gene amplification detected after 40 real-time PCR cycles. There were no negative mrna samples in the present study. Immunostaining for LIF was seen in glandular epithelial cells as well as endometrial stromal cells. However, letrozole did have some immunostaining in the cytoplasm of glandular epithelial cells (Fig. 2C). Letrozole significantly increased LIF immunoexpression compared with endometrium from women in the control and CC groups (P¼.006 and P¼.005, respectively; Fig. 3A). The LIF immunoexpression was also significantly decreased in the CC group compared with controls (P¼.039). This is despite the increase in LIF mrna expression after CC and letrozole treatment (Fig. 1A). To determine whether CC decreased the LIF receptor in the endometrium, LIF receptor mrna expression was analyzed. The CC treatment decreased the mrna expression of the LIF receptor compared with controls or women on letrozole (n ¼ 15, P¼.216; Fig. 1D). The FGF-22 immunostaining was primarily seen in glandular epithelial cells as well as endometrial stromal cells (Fig. 2D F). The FGF-22 immunoexpression was significantly reduced in the CC group when compared with either women taking letrozole or the controls (P¼.003 and P¼.011; Fig. 3B). Letrozole did increase FGF-22 immunoexpression in the endometrium compared with the control group, but the difference was not significant (P¼.114). The DKK-1 immunostaining was found in endometrial stromal cells (Fig. 2G I). Women in the control group had a higher immunoexpression of DKK-1 compared with women in the letrozole group (P¼.107), which reached significance when compared with women in the CC group (P¼.007; Fig. 3C). Letrozole also significantly increased immunostaining of DKK-1 compared with women in the CC group (P¼.015). DISCUSSION Clomiphene citrate is still the first line of therapy for patients with PCOS, despite not being equally effective in all situations for superovulation or induction of ovulation. Letrozole is an aromatase inhibitor that has shown a beneficial role in patients with PCOS who exhibit CC resistance (17). Differences in clinical outcomes between patients taking letrozole and CC may be explained by differential alterations in endometrial gene expression. Both studies in women with infertility and animal studies have shown that low mrna or protein expression of LIF is associated with infertility, thus making LIF a predictor of outcome for implantation (18 21). The LIF expression regulates endometrial receptivity through LIF s anchoring effects on trophoblast and the regulation of trophoblast differentiation (22, 23). Although CC and letrozole increased mrna expression of LIF in the endometrium, only letrozole resulted in a corresponding increase in protein expression of LIF in the endometrium (Figs. 2 and 3). Because LIF receptors are constant throughout the implantation window, we then looked at the expression of LIF receptors and found that CC decreased LIF receptor mrna expression (Fig.1D) (24). This suggests that although CC may increase LIF gene expression, this does not translate to an increase in protein expression, due to a decrease in LIF receptors in the endometrium. Our data are in agreement with a recent microarray study (25) showing that LIF expression was not increased in patients with PCOS treated with CC. Letrozole increased the expression of LIF and LIFR mrna to control levels, which resulted in a significant increase in LIF protein expression. All of which may help explain the increased success rate of implantation and subsequent pregnancy in CC-resistant women. We have previously shown that CC down-regulates vascular endothelial growth factor-a production, which may in part be due to Fertility and Sterility â 1053

4 FIGURE 2 Immunohistochemistry of leukemia inhibitory factor (LIF) (A C), fibroblast growth factor 22 (FGF-22) (D F), and dickkhopf homolog 1 (DKK1) (G I) in endometrial tissue. Arrows indicate glandular epithelial cells (C, F) and endothelial stromal cells (D). Magnification, 100 all representative photomicrographs. Clomid ¼ clomiphene citrate; Letrzl. ¼ letrozole. the antiestrogenic effects of CC, as vascular endothelial growth factor does increase in the presence of E (8, 26). In the present study, CC had a nonsignificant decrease in endometrial mrna and protein expression of FGF-22 compared with controls and women in the letrozole treatment group. The FGF-22, which stimulates proliferation and activation of other cells containing FGF receptors, was increased in women taking letrozole. These data are in keeping with previous reports showing that letrozole is associated with a thicker endometrium leading to improved vascularization, as shown with Doppler studies (4, 7). More recent studies have shown that increases in blood flow to the endometrium help with implantation, thus reinforcing the need for neovascularization (27). In the current study, DKK1, an inhibitor of the Wnt pathway, was increased in women administered letrozole compared with women administered CC. The DKK1 is up-regulated during the window of implantation, and a decrease, as seen in women receiving CC, may lead to implantation failure (28). The DKK1 is regulated by P, in which there was no significant difference in P levels due to CC or letrozole in our study (Table 1). We suggest that CC may be working through a nonparacrine mechanism to decrease DKK1 expression (29). Progesterone was lower in the CC group, which is in agreement with previous studies comparing the effects of CC and letrozole (30, 31). A recent study by Savaris et al. (25) found that women with PCOS are to some degree resistant to P, and that this is independent of the effects of CC. This suggests that letrozole may have, in some way compensated for P resistance, whereas CC does not. It is also possible that some of the differences seen in the present study are due to the timing of the endometrial biopsy in relation to ovulation. We tried to minimize this effect by collecting endometrial biopsies 7 days after the LH surge and ensuring that the patient had measureable amounts of circulating P before she was included in the study. Although the BMI was not statistically significant between the groups, women with PCOS did have a higher BMI compared with non-pcos controls (Table 1). This is also in agreement with other investigators who have found that women with PCOS have a higher BMI compared with women without PCOS (32 34). Our control 1054 Wallace et al. Endometrial analysis: CC vs. letrozole Vol. 96, No. 4, October 2011

5 FIGURE 3 Effects of clomiphene citrate (CC) or letrozole on the protein immunoexpression of leukemia inhibitory factor (A), fibroblast growth factor 22 (B), and dickkhopf homolog 1 (C) in the endometrium. Patients per treatment group, N ¼ 5. *P<.05 and **P<.01 between the indicated groups. subjects also had an average BMI ( ), which is in the overweight range (BMI falls between 25 and 29.9), therefore we cannot determine whether any changes were due to an increase in BMI. However, because endometrial gene expression differed between the CC and letrozole groups we do believe that any detected differences were due to the drug administered and not solely to the patient s BMI. In conclusion, the aim of the present study was to detect differences in the endometrial gene expression of a variety of genes thought to play a role in implantation. From our study we suggest that letrozole increases gene expression of key genes playing a role in implantation, and that letrozole may serve as an alternative to CC in patients with PCOS who fail to conceive with CC. Recent studies have suggested an increase in aromatase enzymatic activity in women with PCOS, which may be one possible explanation for the improved ovulation, implantation, and now endometrial gene expression seen with letrozole administration (4, 35, 36). As more information becomes available, a better understanding of implantation, and in particular the effects of these medications on endometrial receptivity, will be defined. Acknowledgments: The authors are grateful to Linda Robinette, R.N., and Martha R. Walley, M.T., for their assistance in collecting specimens for this study. REFERENCES 1. Heard M, Pierce A, Carson S, Buster J. Pregnancies following use of metformin for ovulation induction in patients with polycystic ovary syndrome. Fertil Steril 2002;77: Beck J, Boothroyd C, Proctor M, Farquhar C, Hughes E. Oral anti-oestrogens and medical adjuncts for subfertility associated with anovulation. Cochrane Database System Rev 2005;1:CD Nakamura Y, Ono M, Yoshida Y, Sugino N, Ueda K, Kato H. Effects of CC on the endometrial thickness and echogenic pattern of the endometrium. Fertil Steril 1997;67: Baruah J, Roy K, Rahman S, Kumar S, Sharma J, Karmakar D. Endometrial effects of letrozole and clomiphene citrate in women with polycystic ovary syndrome using spiral artery Doppler. Arch Gynecol Obstet 2009;279: Holzer H, Casper R, Tulandi T. A new era in ovulation induction. Fertil Steril 2006;85: Al-Omari W, Sulaiman W, Al-Hadithi N. Comparison of two aromatase inhibitors in women with clomiphene-resistant polycystic ovarian syndrome. Intern J Gynecol Obstet 2004;85: Mitwally F, Casper R. Use of an aromatase inhibitor for induction of ovulation in patients with an inadequate response to clomiphene citrate. Fertil Steril 2001;75: Hines R, Wallace K, Sopelak V. Letrozole vs clomiphene citrate: effects on endometrial gene expression. In: Reproductive sciences. Vol. 17. Orlando, Fl:, 2010:S244A. 9. Ledee-Bataille N, Lapree-Delage G, Taupin J, Dubanchet S, Taieb J, Moreau J, et al. Follicular fluid concentration of leukemia inhibitory factor is decreased among women with polycystic ovarian syndrome during assisted reproduction cycles. Hum Reprod 2001;16: Wood J, Ho C, Nelson-Degrave V, McAllister J, Strauss J III. The molecular signature of polycystic ovary syndrome (PCOS) theca cells defined by gene expression profiling. J Reprod Immunol 2004;63: Artini P, Monti M, Matteucci C, Valention V, Cristello F, Genazzani A. Vascular endothelial growth factor and basic fibroblast growth factor in polycystic ovary syndrome during controlled ovarian hyperstimulation. Gynecol Endocrinol 2006;22: Fraser H. Regulation of the ovarian follicular vasculature. Reprod Bio Endocrinol 2006;4:1 9. Fertility and Sterility â 1055

6 13. Casper R, Mitwally F. Review: aromatase inhibitors for ovulation induction. J Clin Endocrinol Metab 2006;91: Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanatephenol-chloroform extraction. Anal Biochem 1987;162: Livak K, Schmittgen T. Analysis of relative gene expression data using real-time quantitative PCR and the 2 DeltaDeltaCt method. Methods 2001;25: Abramoff M, Magelhaes P, Ram S. Image processing with Image J. Biophotonics Inter 2004;11: Elnashar A, Fouad H, Eldosoky M, Saeid N. Letrozole induction of ovulation in women with clomiphene citrate resistant polycystic ovary syndrome may not depend on the period of infertility, the body mass index, or the luteinizing hormone/follicle-stimulating hormone ratio. Fertil Steril 2006;85: Mikolajczyk M, Wirstlein P, Skrzypczak J. The impact of leukemia inhibitory factor in uterine flushing on the reproductive potential of infertile women a prospective study. Am J Reprod Immunol 2007;58: Mikolajczyk M, Wirstlein P, Skrzypczak J. Leukaemia inhibitory factor and interleukin 11 levels in uterine flushings of infertile patients with endometriosis. Hum Reprod 2006;21: Fouladi-Nashta A, Jones C, Nijjar N, Mohamet L, Smith A, Chambers I, et al. Characterization of the uterine phenotype during the peri-implantation period for LIF-null, MF1 strain mice. Dev Bio 2005;281: Kimber S. Leukaemia inhibitory factor in implantation and uterine biology. Reproduction 2005;130: Cheng J, Rodriguez C, Stewart C. Control of uterine receptivity and embryo implantation by steriod hormone regulation of LIF production and LIF receptor activity: towards a molecular understanding of the window of implantation. Rev Endocrinol Metab Disord 2002;3: Marwood M, Visser K, Salamonsen L, Dimitriadis E. Interleukin-11 and leukemia inhibitory factor regulate the adhesion of endometrial epithelial cells: implications in fertility regulation. Endocrinology 2009;150: Cullinan E, Abbondanzo S, Anderson P, Pollard J, Lessey B, Stewart C. Leukemia inhibitory factor (LIF) and LIF receptor expression in human endometrium suggests a potential autocrine/paracrine function in regulating embryo implantation. Proc Natl Acad Sci U S A 1996;93: Savaris R, Groll J, Young S, DeMayo F, Jeong J, Hamilton A, et al. Progesterone resistance in PCOS endometrium: a microarray analysis in clomiphene citrate-treated and artificial menstrual cycles. J Clin Endocrin Metab 2011;96: Taylor R, Lebovic D, Hornung D, Mueller M. Endocrine and paracrine regulation of endometrial angiogenesis. Ann N Y Acad Sci 2006;943: Zhao M, Chang C, Liu Z, Chen L, Chen Q. Treatment with low-dose aspirin increased the level of LIF and integrin B3 expression in mice during the implantation window. Placenta 2010;31: Kao L, Tulac S, Lobo S, Imani B, Yang J, Germeyer A, et al. Global gene profiling in human endometrium during the window of implantation. Endocrinology 2002;143: Tulac S, Overgaard M, Hamilton A, Jumber N, Suchanek E, Giudice L. Dickkopf-1, an inhibitor of Wnt signaling, is regulated by progesterone in human endometrial stromal cells. J Clin Endocrinol Metab 2006;91: Begum M, Ferdous J, Begum A, Quadir E. Comparison of efficacy of aromatase inhibitor and clomiphene citrate in induction of ovulation in polycystic ovarian syndrome. Fertil Steril 2009;92: Elsedeek M, Elmaghraby H. Predictors and characteristics of letrozole induced ovulation in comparison with clomiphene induced ovulation in anovulatory PCOS women. Middle East Fertil Soc J 2011;16: Broekmans F, Knauff E, Valkenburg O, Laven J, Eijkemans M, Fauser B. PCOS according to the Rotterdam consensus criteria: change in prevalence among WHO-II anovulation and association with metabolic factors. Br J Obstet Gynecol 2006;113: Vrbikova J, Hainer V. Obesity and polycystic ovary syndrome. Obesity Facts 2009;2: Pandey S, Pandey S, Maheshwari A, Bhattacharya S. The impact of female obesity on the outcome of fertility treatment. J Reprod Sci 2010;3: Maia H, Casoy J, Filho J. Is aromatase expression in the endometrium the cause of endometriosis and related infertility? Gynecol Endocrinol 2009;25: Wang H, Li Q, Wang T, Yang G, Wang Y, Zhang X, et al. A common polymorphism in the human aromatase gene alters the risk for polycystic ovary syndrome and modifies aromatase activity in vitro. Basic Sci Reprod Med 2011;17: Wallace et al. Endometrial analysis: CC vs. letrozole Vol. 96, No. 4, October 2011