North Carolina Center for Reproductive Medicine, Cary, North Carolina

Transcription

1 FERTILITY AND STERILITY VOL. 75, NO. 3, MARCH 2001 Copyright 2001 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Metformin treatment of patients with polycystic ovary syndrome undergoing in vitro fertilization improves outcomes and is associated with modulation of the insulin-like growth factors Laurel A. Stadtmauer, M.D., Ph.D., Sameh K. Toma M.D., Robert M. Riehl, Ph.D., and Luther M. Talbert, M.D. North Carolina Center for Reproductive Medicine, Cary, North Carolina Received April 27, 2000; revised and accepted November 9, Presented in part at the 55th Annual Meeting of the American Society of Reproductive Medicine in Toronto, Ontario, Canada, September 25 30, Reprint requests: Laurel A. Stadtmauer, M.D., Ph.D., North Carolina Center for Reproductive Medicine, 400 Ashville Ave, Suite #200, Cary, North Carolina (FAX: ; /01/$20.00 PII S (00) Objective: To determine if metformin therapy improves in vitro fertilization (IVF) outcomes in patients with clomiphene-resistant polycystic ovarian syndrome (PCOS). Design: Retrospective data analysis of selective groups of patients. Setting: A private IVF unit. Patient(s): Forty-six women with clomiphene citrate resistant PCOS underwent 60 cycles of IVF embryo transfer with intracytoplasmic sperm injection. Intervention(s): In half of the cycles, patients received metformin (1000 to 1500 mg) daily, starting the cycle prior to gonadotropin treatment. Main Outcome Measure(s): Total number of follicles; serum estradiol (E 2 ) on the day of hcg administration and the cycle s E 2 maximum; total number of oocytes, mature oocytes, embryos, fertilization, and pregnancy rates; and follicular fluid levels of insulin-like growth factors (IGF-I, IGF-II) and IGF-binding proteins (IGFBP-1, IGFBP-3). Result(s): In patients treated with metformin, the total number of follicles on the day of hcg treatment was decreased ( vs ) with no change in follicles 14 mm in diameter ( vs ). Metformin treatment did not affect the mean number of oocytes retrieved ( vs ). However, the mean number of mature oocytes ( vs ) and embryos cleaved ( vs ) were increased after metformin treatment. Fertilization rates (64% vs. 43%) and clinical pregnancy rates (70% vs.30%) were also increased. Metformin led to modulation of preovulatory of follicular fluid IGF levels with increases of IGF-I (140 8 vs ng/mL) and decreased of IGFBP-1 (133 8 vs.153 9ng/mL). Conclusion(s): Metformin use appears to improve IVF outcomes in patients with clomiphene citrate resistant PCOS. (Fertil Steril 2001;75: by American Society for Reproductive Medicine.) Key Words: PCOS, metformin, IVF, FSH stimulation Polycystic ovary syndrome (PCOS) is characterized by menstrual disturbances, oligo-ovulation or anovulation and hyperandrogenism (1, 2). These patients typically have elevated free testosterone levels, LH levels, and LH/FSH ratios as well as micropolycystic ovaries. Insulin resistance and obesity are also common features of the syndrome (3, 4). Metformin, a biguanide insulin-sensitizing agent, has been shown to reduce hyperinsulinemia and hyperandrogenemia associated with polycystic ovary syndrome and aid in ovulation (5 9). Metformin reduces plasma levels of LH as well as ovarian levels of androgens. Metformin administration has been shown to improve ovarian response to exogenous gonadotropin in women with clomiphene-resistant PCOS (10). Patients with PCOS undergoing in vitro fertilization (IVF) usually are high responders to gonadotropins, producing large numbers of follicles with high serum estradiol levels and increasing the risk of ovarian hyperstimulation. However, a greater percentage of these oocytes retrieved are immature. This is presumably due to the high androgen environment in the ovary. 505

2 The objective of this study was to see if metformin improves the outcome in clomiphene citrate resistant patients who are undergoing IVF. We hypothesized that metformin may improve the quality of oocytes retrieved from these patients by reducing their hyperinsulinism and by modulating the local insulin and insulin-like growth factor (IGF) levels. We also looked at the effect of metformin on ovarian response to gonadotropins after down-regulation with GnRH-agonist as well as its effect on the oocyte maturity and rates of fertilization and pregnancy. Previous studies suggested that the pathogenesis of PCOS may involve abnormalities of the IGF and ovarian steroidogenesis systems (4). It has been postulated that elevated levels of insulin and IGF-I along with elevated LH, acting on the thecal component in vivo, contribute to the hyperandrogenemia observed clinically in PCOS patients. The in vivo actions of IGF-I and IGF-II are modulated by a system of circulating binding proteins (IGFBPs). Of the six IGFBPs, IGFBP-1 has been found to correlate inversely with the estimates of the free fraction of IGF-I (11, 12). In addition, insulin and IGF-I regulate the production of IGFBP-1 at the level of suppression of the IGFBP-1 synthesis (11 14). This is most likely the mechanism for the decrease of serum and follicular fluid IGFBP-1 concentrations in PCOS (12). Although the levels of total serum IGF-I are normal in PCOS (15 17), the decreased IGFBP-1 concentrations could lead to elevated levels of free IGF-I (figf-i), which may then stimulate ovarian androgen synthesis. The purpose of our study was to review our use of metformin in PCOS patients who are undergoing IVF treatment. We also measured the preovulatory levels of IGF-I, IGF-2, IGFBP-1, and IGFBP-3 in follicular fluid of PCOS patients who were undergoing IVF embryo transfer (IVF- ET) with and without metformin treatment to see if these levels were affected by the metformin treatment. MATERIALS AND METHODS From April 1998 to December 1999, 60 cycles of IVF in PCOS patients (n 46) aged 25 to 39 were reviewed. In 30 of the 60 cycles of IVF-ET and intracytoplasmic sperm injection (ICSI), patients were treated with metformin during their IVF cycle after they had given informed consent. These patients were compared retrospectively to patients who had not received metformin. We did not obtain institutional review board approval, because during that period of time metformin had become a standard treatment for ovulation induction in PCOS patients, and the patients and their physician made the decision of whether to use metformin. The patients did not have additional testing done beyond what was standard for all IVF patients with PCOS. Eight patients had more than one cycle. However, the number of previous attempts was controlled for in each group. All patients had clomiphene-resistant PCOS, as defined by a day 21 progesterone level of less than 2.5 ng/ml and absence of folliculogenesis on the day 14 ultrasound with doses of up to 200 mg of clomiphene citrate. In addition, these patients had polycystic appearing ovaries on ultrasound (10 or more follicles 2 to 10 mm in diameter on each ovary, and increased stroma). They also had hyperandrogenemia as defined by either elevated serum androgen levels (serum T levels 80 ng/ml), or an LH to FSH ratio greater than 2 on initial screening. All patients had normal TSH, prolactin, and day 3 FSH levels. In addition, fasting insulin and glucose levels before metformin treatment were performed as a crude assessment of insulin resistance. Based on these measurements, we were dealing with a heterogeneous group of patients with regard to insulin resistance. However, the decision to treat with metformin was made independent of insulin resistance. In the untreated group, none of the patients had received metformin for at least 3 months prior to the IVF cycle. Metformin-treated patients received it twice a day (500 mg) if they had a body mass index of less than 28 kg/m 2 or three times a day (500 mg) for a body mass index greater than 28kg/m 2 ; the treatment started at day 1 of their cycle prior to commencement of gonadotropins, and continued to the day of the pregnancy test. Patients were down-regulated with GnRH-agonist (Lupron, 0.5 mg/day) following 2 weeks of OCP treatment. Patients were then stimulated with rec-fsh (150 to 300 U; Follistim, Organon, West Orange NJ) and the doses were adjusted based on individual response. In both groups, monitoring with ultrasound examination and E 2 levels was initiated on day 6 of goandotropin stimulation. The gonadotropin dosage was dropped if the E 2 levels were 600 pg/ml. At this point, patients were monitored every 1 to 2 days. Coasting, or withholding gonadotropins, was initiated when plasma E 2 levels were 3000 pg/ml. HCG, 10,000 IU (Pregnyl, Organon, West Orange NJ), was administered when four leading follicles were 20 mm mean diameter and E 2 levels were 4500 pg/ml. If E 2 levels rose above 4500 pg/ml, gonadotropin and hcg administration was withheld until levels dropped below 3500 pg/ml. Oocyte retrieval was performed 35 hour following hcg administration. A total of 40 preovulatory follicular fluid samples from the first aspirate (20 from each group) were collected on the day of retrieval and stored at 20 C until assays were run. Assays The concentrations of LH, FSH, testosterone and fasting insulin, and glucose levels were determined by standard methods. Serum E 2 levels were measured on the VIDAS enzyme immunoassay system (bio Mérieux Vitek, Inc., Hazelwood, MO). IGF-I and IGF-II and IGFBP-1 and IGFBP-3 levels were all were done in triplicates. IGF-I and IGF-II were measured by IRMA after acid-ethanol extraction (Diagnostic Systems Laboratories, Webster, TX) and IGFBP-1 and IGFBP-3 levels were measured using two-site immuno- 506 Stadtmauer et al. Metformin treatment and PCOS in IVF Vol. 75, No. 3, March 2001

3 TABLE 1 Clinical and hormonal data before metformin treatment from 46 women with clomiphene citrate resistant polycystic ovary syndrome. Age (years) NS BMI (kg/m 2 ) NS Cycle attempt number NS Fasting insulin ( U/mL) NS T (ng/ml) NS TABLE 2 Results of ovarian stimulation in patients with PCOS undergoing 60 cycles of IVF. No. of cycles FSH (no. of ampules) NS Follicles (total) P.005 Follicles 14 mm NS Maximum E 2 level (pg/ml) P.05 E 2 level day of hcg (pg/ml) NS radiometric assays (Diagnostic Systems Laboratories). Intraassay and interassay CVs were less than 10%. ICSI was performed for all PCOS patients undergoing IVF in this study because our past experience showed poor fertilization rates in a number of PCOS patients unless ICSI was used. Therefore, ICSI enabled us an accurate assessment of oocyte maturity. ICSI was done on all ova determined to be mature by the presence of a polar body. Embryo transfer was performed 3 days after retrieval. The number of embryos transferred was individualized to the patient; three or four embryos was the most common number used. Data Analysis Control and metformin-treated groups were compared. All data are expressed as means SEM. Data was analyzed by the Student t test, the Mann-Whitney U test, or chisquare, as indicated. A value of P.05 was considered statistically significant. Fertilization rates, total pregnancy rates, and clinical pregnancy rates are compared. A clinical pregnancy was confirmed by the presence of a fetal heart motion at 6 weeks after embryo transfer, and a positive pregnancy was confirmed by a serum level of hcg 50 miu/ml at 14 days after transfer. RESULTS Patient Characteristics The patient characteristics of the metformin-treated and untreated groups are summarized in Table 1. Control and metformin groups do not statistically differ with regard to age, body mass index, and the number of cycles attempted. In addition, fasting insulin concentrations and baseline androgen studies did not differ between the groups. The decision to treat with metformin was made independent of the patient s insulin resistance status, depending only on the patient s clinical diagnosis of PCOS (as described in the section on methods). Ovarian simulation in metformin-treated and untreated groups were compared. The amount of gonadotropin used was similar in both groups. The number of follicles greater than or equal to a 14-mm mean diameter, the total number of follicles, and the peak serum E 2 levels and the E 2 levels on day of hcg administration were reviewed. As shown in Table 2, the mean ( SD) total number of follicles on ultrasound was increased in the untreated group, with a significant increase in the mean number of follicles less than 14 mm in diameter. Metformin treatment significantly decreased the number of ovarian follicles less than 14 mm on the day of hcg. This led to lower serum peak E 2 levels and lower incidence of coasting with similar E 2 levels on the day of hcg. No pregnancies were achieved when the maximum E 2 was greater than 7500 pg/ml (data not shown), supporting the benefit of lowering the serum peak E 2 levels. The same step-down protocol for adjusting the gonadotropin dosage was used in both groups of patients (see the section on methods). TABLE 3 Results of IVF outcome in patients with PCOS treated with or without metformin. No. of cycles Total no. of oocytes NS No. of mature oocytes P.05 No. of embryos 4 cells P.001 No. of oocytes fertilized a (%) 43 (264/611) 64 (421/657) P.05 No of pregnancies b (%) 40 (12/30) 78 (24/30) NS No of clinical pregnancies c (%) 30 (9/30) 70 (21/30) P.05 Note: Values are means SE unless otherwise indicated. Values in parentheses are the actual numbers obtained. NS is not significant. a Number of oocytes fertilized/total no. of oocytes retrieved. b Number of positive pregnancy tests/number of cycles times 100. c Number of clinical pregnancies/no. of cycles times 100. FERTILITY & STERILITY 507

4 TABLE 4 Folllicular fluid insulin growth factor levels in 40 cycles from control and metformin-treated PCOS patients. Control Table 3 summarizes the embryo outcomes and pregnancy rates in the two groups of patients. The mean number ( SD) of oocytes retrieved did not differ between the untreated and metformin-treated groups, 20 and 22 respectively. However, metformin treatment was associated with an increased mean number of mature oocytes as defined by the presence of a polar body (18.4 vs. 13), and increased number of fourcell embryos on day 3 of transfer (12 vs. 6). Gonadotropin stimulation appeared to lead to improvements in mature oocyte number and numbers of cleaved embryos per cycle. In addition, both fertilization rates and pregnancy rates were improved in the metformin-treated group, with similar numbers of embryos transferred in each group (untreated, ; treated, ). Fertilization rates increased from 43% in control patients to 64% in metformin-treated patients and clinical pregnancy rates increased from 30% to 70%. We measured levels of IGFs and IGFBPs in the preovulatory follicular fluid from the first aspirate of PCOS patients treated with or without metformin to see whether metformin treatment affected these levels. Follicular fluid levels of IGF-I increased in the metformin-treated patients along with a decrease in IGFBP-1. This is consistent with increased levels of free IGFs, as IGFBP-1 has been found to correlate inversely with free fraction of IGF-I. IGFBP-3 and total IGF-II were unchanged. Although total IGF-II was not affected, free IGF-II levels in follicular fluid may be increased because of changes in the binding proteins. DISCUSSION IGF-I (ng/ml) P.001 IGF-II (ng/ml) NS IGFBP-1 (ng/ml) P.05 IGFBP-3 (ng/ml) NS PCOS is characterized by hyperandrogenemia and hyperinsulinemia; metformin treatment has been found to reduce both (5 9). The purpose of this study was to see whether these conditions improved among patients with clomiphene citrate resistant PCOS, who frequently have poorer outcomes in IVF that do women without PCOS. This preliminary study supports the use of metformin in PCOS patients and shows a beneficial effect on the number of mature oocytes and the rates of fertilization, embryos cleaved, and pregnancy. Metformin was also associated with changes in ovarian stimulation. There appears to be a shift in follicle size, reducing the number of small cohort follicles. In addition there was a slower rise of E 2 levels, lower peak E 2 levels, and fewer days of coasting. Our patients were clomiphene-resistant, anovulatory patients; therefore, their PCOS was considered more severe. In addition, their body mass index showed many were obese. These patients have a high incidence of insulin resistance. However, detailed study of insulin tolerance in these patients was not done. Based on the initial fasting insulin and glucose studies, we were dealing with a heterogeneous population of patients, with and without insulin resistance. The mean fasting insulin levels and total testosterone levels were similar in both groups. Further studies do need to be done on the effect of metformin treatment on androgen levels both in serum and follicular fluid. Also, studies need to be done that measure hyperinsulinemia more precisely to determine whether the metformin treatment affected insulin levels in these patients. The decision to treat patients with metformin was made independent of the demonstrated insulin resistance so that we could look at its potential benefit in a heterogeneous group of women with clinical PCOS. Serum and follicular fluid IGFBP-1 concentrations are decreased in PCOS, presumably because of hyperinsulinism and suppression of IGFBP-1 synthesis (13, 14). Therefore, one would expect metformin to increase IGFBP-1 and decrease free IGF-I, leading to decreased free IGF-I levels. We found the opposite. It is possible that metformin may be affecting IGFBP-1 levels through a different mechanism than that of reducing the hyperinsulinemia. Metformin has been shown to reduce free testosterone levels, and in our study it reduced the total number of small ovarian follicles detected on ultrasound. Free IGF-I levels have been shown to vary inversely with the total number of ovarian follicles as well as DHEAS and free T levels (17). Therefore, this may be the mechanism for increasing free IGF-I in our metformin-treated patients. However, the role of IGFs in folliculogenesis and its modulation by metformin treatment is beyond the scope of this paper. The physiological relevance of elevated insulin or free IGF-I and IGF-II in the follicular fluid and the ovarian IGFBP-1 regulation is uncertain in these PCOS patients. These patients were down-regulated and treated with gonadotropins, which may have changed the growth factor levels. We are looking at the IGFBP-1 levels in the preovulatory dominant follicles after stimulation for IVF. Changes in the circulating or follicular fluid IGFBP-1 in women undergoing controlled ovarian hyperstimulation may have influenced the IGF-II mediated events in the follicle in addition to IGF-I. Indeed, IGF-II mediated effects on FSH may be more important than IGF-I (16). Ovarian antral follicle growth and steroidogenesis is believed to be mediated by IGF-II. The study by El-Roeiy et al. (18) supported IGF-II as a mediator of FSH action on human preantral follicles. Higher free 508 Stadtmauer et al. Metformin treatment and PCOS in IVF Vol. 75, No. 3, March 2001

5 IGF-I and -II in follicular fluid of metformin treated patients in conjunction with FSH may be acting locally to produce a higher estrogen environment leading to a higher number of mature eggs and a low number of small androgenic follicles. An increase in the local IGF-I level in metformin-treated patients may lead to the inhibition of IGFBP-1, which binds and modulates the actions of IGFs (12). Insulin, IGF-II, and IGF-I all have inhibitory effects of the IGFBP-1 expression with IGF-I having the largest effect. Higher inhibitory IGFBP levels in control patients may lead to an arrest of follicular development and a larger number of small follicles. Metformin may lower the androgen level in the follicular fluid, affecting local levels of IGFs, and also may produce improved ovarian stimulation. A lesser total number of follicles, but a greater proportion of mature follicles was found on the day of hcg triggering. In addition, the lower peak E 2 levels associated with metformin treatment may have a beneficial effect on the endometrium, as high E 2 levels have been shown to be deleterious to implantation (19 21). In conclusion, in this preliminary study metformin use appears beneficial in severe PCO patients who are undergoing IVF-ET. Local levels of free IGF-I and- II seem to be affected. Therefore, increased free IGFs may be acting in conjunction with FSH in improving the ovarian stimulation, oocyte maturation, and total number of advanced embryos obtained per cycle. This also had an impact on the pregnancy rate. References 1. Van Santbrink EJ, Hop WC, Fauser BC. Classification of normogonadotropic infertility: polycystic ovaries diagnosed by ultrasound vs. endocrine characteristics of polycystic ovary syndrome. Fertil Steril 1997;67: Franks S. Polycystic ovary syndrome: a changing perspective. Clin Endocrinol 1989;31: Utiger RD. Insulin and the polycystic ovary syndrome. N Engl J Med 1996;335: Franks S, Gilling-Smith C, Watson H, Willis D. Insulin action in the normal and polycystic ovary. Endocrinol Metab Clin North Am 1999; 28: Valazquez E, Acosta A, Mendoza SG. Menstrual cyclicity after metformin therapy in polycystic ovary syndrome. Obstet Gynecol 1997; 90: Nestler JE, Jakubowicz DJ, Evans WS, Pasquali R. Effects of metformin on spontaneous and clomiphine-induced ovulation in the polycystic ovary syndrome. N Engl J Med 1998;338: Ehrmann DA, Cavaghan MK, Imperial J, Sturis J, Rosenfield RL, Polonsky KS. Effects of metformin on insulin secretion, insulin action, and ovarian steroidogenesis in women with polycystic ovary syndrome. J Clin Endocrinol Metab 1997;82: Morin-Papunen LC, Koivunen RM, Ruokonen A, Martikainen HK. Metformin therapy improves the menstrual pattern with minimal endocrine and metabolic effects in women with polycystic ovary syndrome. Fertil Steril 1998;69: Sattar N, Hopkinson ZE, Greer IA. Insulin-sensitising agents in polycystic-ovary syndrome. Lancet 1998;351: De Leo V, La Marca A, Ditto A, Morgante G, Cianci A. Effects of metformin on gonadotropin-induced ovulation in women with polycystic ovary syndrome. Fertil Steril 1999;72: Buyalos RP, Pekonen F, Halme JK, Judd HL, Rutanen EM. The relationship between circulating androgens, obesity, and hyperinsulinemia on serum insulin-like growth factor binding protein-1 in the polycystic ovarian syndrome. Am J Obstet Gynecol 1995;172: Lee PD, Guidice LC Conover CA, Powell DR Insulin-like growth factor binding protein-1: recent findings and new directions Proc Soc Exp Biol Med 1997;216: Suikkari AM, Koivisto VA Rutanen EM, Yki-Jarvinen H, Karonen SI, Seppala M. Insulin regulates the serum levels of low molecular weight insulin-like growth factor-binding protein. J Clin Endocrinol Metab 1988; 66: Homberg R, Orvieto R, Bar-Hava I, Ben-Rafael Z. Serum levels of insulin-like growth factor-i (IGF-I) and IGF binding protein-1 (IG- FBP-1) and insulin and the response to human menopausal gonadotrophins in women with polycystic ovary syndrome. Human Reprod 1996;11: Poretsky L, Cataldo NA, Rosenwaks Z, Giudice LC The insulin-related ovarian regulatory system in health and disease. Endocr Rev 1999;20: Guidice LC. Growth factor action on ovarian function in polycystic ovary syndrome Endocrinol Metab Clin North Am 1999;28: Thierry Van Dessel HJ, Lee PD, Faessen G, Fauser BC, Guidice LC. Elevated serum levels of free insulin-like growth factor I in polycystic ovary syndrome J Clin Endocrinol Metab 1999;84: El-Roeiy A, Chen X, Roberts VJ, Le Roith D, Roberts CT Jr, Yen SSC. Expression of the genes encoding the insulin-like growth factors (IGF-I and II), and the IGF and insulin receptors, and IGF-binding proteins 1-6 and localization of their gene products in normal and polycystic ovary syndrome ovaries. J Clin Endocrinol Metab 1994;78: Pellicer A, Valbuena D, Cano F, Remohi J, Simon C. Lower implantation rates in high responders: evidence for an altered endocrine milieu during the preimplantation period. Fertil Steril 1996;65: Diamond MP, Buchholz T, Boyers SP, Lavy G, Shapiro BS, DeCherney AH. Super high estradiol response to gonadotropin stimulation in patients undergoing in vitro fertilization. J In Vitro Fert Embryo Transf 1989;6: Dor J, Seidman DS, Ben-Shlomo I, Levran D, Karasik A, Mashiach S. The prognostic importance of the number of oocytes retrieved and estradiol levels in poor and normal responders in in vitro fertilization (IVF) treatment. J Assist Reprod Genet 1992;9: FERTILITY & STERILITY 509