that induce morphologic (4, 5) and biochemical (6) endometrial alterations relevant to uterine receptivity.

FERTILITY AND STERILITY VOL. 70, NO. 2, AUGUST 1998 Copyright 1998 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. 12 Increasing uterine receptivity by decreasing estradiol levels during the preimplantation period in high responders with the use of a follicle-stimulating hormone step-down regimen Carlos Simón, M.D.,* Juan J. Garcia Velasco, M.D.,* Diana Valbuena, M.D., Jose A. Peinado, M.D., Carlos Moreno, M.D.,* José Remohí M.D.,* and Antonio Pellicer, M.D.* Valencia University, Valencia; Instituto Valenciano de Infertilidad Madrid, Madrid; and Laboratorios Serono, Madrid, Spain Received December 22, 1997; revised and accepted March 6, 1998. Supported by FISss 95/ 1352 and 96/1263 grants from the Spanish government, Madrid, Spain. Presented in part at the 48th Annual Meeting of the European Society of Human Reproduction and Embryology, Edinburgh, United Kingdom, June 22 25, 1997. Reprint requests: Carlos Simón, M.D., Instituto Valenciano de Infertilidad, Guardia Civil, 23, 46020 Valencia, Spain (FAX: 346-3694735; E-mail: IVI@FUTURNET.ES). * Instituto Valenciano de Infertilidad and Department of Pediatrics, Obstetrics and Gynecology, Valencia University. Instituto Valenciano de Infertilidad Madrid. Laboratorios Serono. 0015-0282/98/$19.00 PII S0015-0282(98)00140-X Objective: To analyze the effect on uterine receptivity of a decrease in E 2 levels during the preimplantation period with the use of a step-down regimen in high responders undergoing IVF. Design: Prospective controlled clinical study. Setting: The Instituto Valenciano de Infertilidad. Patient(s): High responders in whom at least one previous IVF attempt failed in which 3 4 good-quality embryos were transferred and E 2 levels were 3,000 pg/ml on the day of hcg administration. Intervention(s): Gonadotropins were administered according to two different protocols. Blood samples were collected and IVF was performed. Main Outcome Measure(s): Serum E 2 levels and reproductive outcome of IVF. Result(s): Estradiol levels on the day of hcg administration and throughout the preimplantation period and the number of oocytes collected were significantly lower with the use of the step-down regimen than during the previous failed in which the standard protocol was used. The fertilization rate was similar and the number of good-quality embryos transferred was comparable. However, the implantation and pregnancy rates were significantly improved in patients who underwent the step-down regimen compared with those who received the standard protocol. Conclusion(s): With the use of a step-down regimen with FSH in high responders, our clinical results demonstrate that uterine receptivity can be improved when E 2 levels are decreased during the preimplantation period. (Fertil Steril 1998; 70:234 9. 1998 by American Society for Reproductive Medicine.) Key Words: High responders, estradiol, step-down Since the introduction of IVF, there has been substantial improvement in ovulation induction, oocyte retrieval, fertilization capability, and embryonic development. However, the end point, which is to improve implantation rates and pregnancy rates (PRs) after the transfer of 3 4 embryos, remains elusive. Indeed, up to 90% of apparently healthy embryos that are transferred to the uterus are destined to vanish (1). We know from basic studies that the success of embryonic implantation relies on a perfect dialogue between good-quality embryos and a receptive endometrium. Although the importance of embryo quality in IVF has been demonstrated thoroughly and embryo quality has improved (2), the management of the uterine factor remains vague. Uterine receptivity is deteriorated during controlled ovarian hyperstimulation (COH) for IVF compared with hormone replacement therapy (HRT) and natural s (3). This is not surprising considering that the aim of ovulation induction drugs is to recruit more oocytes; as a consequence, supraphysiologic levels of steroid hormones are produced that induce morphologic (4, 5) and biochemical (6) endometrial alterations relevant to uterine receptivity. 234

Moreover, we and other investigators consistently have found even lower implantation rates per embryo replaced in patients undergoing IVF who have a high response to gonadotropins (7 10), although this contrasts with other reports wherein a high response has been associated with no change (11, 12) or increased implantation (13). We previously demonstrated that high serum E 2 levels ( 3,000 pg/ml) on the day of hcg administration in high responders, regardless of the number of oocytes retrieved and the serum progesterone levels, are detrimental to uterine receptivity without affecting embryo quality (9). Indeed, PRs and implantation rates in recipients of embryos derived from the oocytes of high responders are similar to those in normal responders (9). Further, we documented an increase in serum E 2 levels during the preimplantation period in high responders compared with normal responders, suggesting that this abnormal endocrine milieu could be responsible for impaired implantation (10). Following this research line, and in light of the fact that high levels of serum E 2 during the preimplantation period are deleterious to endometrial receptivity, we undertook the present study to determine whether uterine receptivity (as measured by implantation rates and PRs) could be improved in high responders undergoing IVF with the use of a stepdown regimen designed to decrease E 2 levels during the preimplantation period. MATERIALS AND METHODS Study Design This study included a total of 91 IVF s performed in 86 high responders treated at our institution between January 1, 1996, and July 31, 1997. The study was approved by our institution s ethics committee, and institutional review board approval was obtained contingent on volunteer assignment of the patients after the provision of complete information. Blind randomization was ruled out because we had reasonable evidence to believe that one treatment produces better results than the other, and this design was imposed for institutional review board approval. To be included in this study, patients had to have had at least one previous failed IVF attempt in which 3 4 goodquality embryos were transferred and the E 2 level was 3,000 pg/ml on the day that hcg was administered. These patients were classified as high responders, as previously described (14). During their next, the patients were informed about the possibility of being included in the two study groups (standard versus step-down protocol). Twentyfour patients were treated in 29 s with a step-down regimen that previously was used for other clinical situations (15), whereas 62 patients were treated in 62 s with the same standard protocol used in the previous. Written consent was obtained from all participants. Patients and Stimulation Protocol The causes of infertility for patients who entered an IVF program and who received the step-down regimen were as follows: in 13 cases (54.2%) there was male factor infertility, in 4 cases (16.7%) there was tubal infertility, and in 7 cases (29.1%) there was unexplained infertility. Among patients who received the standard regimen, 35 (56.4%) had male factor infertility, (6.4%) had tubal infertility, 3 (4.8%) had endometriosis, and 20 (32.4%) had unexplained infertility. The protocol for ovarian stimulation was initiated with pituitary desensitization by the administration of leuprolide acetate, 1 mg/d SC (Procrin; Abbot S.A., Madrid, Spain), starting in the luteal phase of the previous. Serum E 2 levels of 60 pg/ml (220 pmol/l) and negative findings on vaginal ultrasound (US) scans were used to define ovarian quiescence. For the step-down regimen, on day 1 of ovarian stimulation, 4 ampules of highly purified FSH (NeoFertinorm; Serono Laboratorios, S.A., Madrid, Spain) were administered. On day 2, 3 ampules were administered, and on days 3 and 4, 2 ampules were administered. Starting on day 5, 1 1 ampules of FSH were given on an individual basis according to 2 serum E 2 levels and transvaginal ovarian US scans. For the standard protocol, on days 1 and 2 of ovarian stimulation, 1 ampule of hmg (Pergonal; Serono Laboratorios) was administered together with 3 ampules of highly purified FSH. On days 3, 4, and 5 of ovarian stimulation, 1 ampule of hmg and 1 ampule of FSH were given to each patient. Beginning on day 6, hmg and FSH were administered on an individual basis according to serum E 2 levels and transvaginal ovarian US scans. In both protocols, the criteria for hcg administration (10,000 IU, Profasi; Serono Laboratorios) were the same. The criteria for hcg administration included the presence of 2 follicles measuring at least 19 mm in diameter and serum E 2 levels of 800 pg/ml (2.94 nmol/l). Leuprolide acetate and gonadotropin injections were discontinued the day of hcg administration. Oocyte retrieval was scheduled for 36 38 hours after hcg injection and intravaginal micronized progesterone, 400 mg/d (Progeffik; Effik Laboratory, Madrid, Spain), was administered as luteal support. In Vitro Fertilization/Intracytoplasmic Sperm Injection The standard IVF/intracytoplasmic sperm injection procedure has been described previously (9, 10). Briefly, oocyte-cumulus complexes were evaluated under the dissecting microscope and classified. Oocyte-cumulus complexes were incubated at 37 C under 5% CO 2 in atmospheric air. Embryos were scored on the day of transfer according to their morphology under the dissecting microscope. Four types of embryos were established, ranging from types I IV. Type I embryos were the best and were defined as round and wellshaped blastomeres without fragments. Our policy for ET FERTILITY & STERILITY 235

TABLE 1 Reproductive outcome of the step-down versus the previous failed of IVF. Variable Previous (n 24) Step-down (n 24) P value Patient age (y) 31.1 1.4 31.6 1.2 NS GnRH analogue (d) 20.2 0.6 22.9 0.7 NS Days of treatment 9.7 0.4 10.6 0.1 NS Ampules of gonadotropins 27.2 2.1 22.4 0.8 NS E 2 level (pg/ml) 5,770 650 1,919 477.001 No. of oocytes 24.0 1.9 18.1 2.1.001 Fertilization rate (%) 71.2 74.2 NS No. of embryos transferred 3.8 0.2 3.3 0.2 NS No. of embryos frozen 6.2 1.0 2.5 0.7.002 Pregnancy rate (%) 0 64.2.0001 Implantation rate (%) 0 29.3.0001 Note: Values are expressed as means SE unless otherwise indicated. NS not significant. TABLE 2 Reproductive outcome of the standard protocol versus the previous failed of IVF. Variable Previous (n 62) Standard protocol (n 62) P value Age (y) 33.1 0.4 33.8 0.3 NS GnRH analogue (d) 20.4 1.6 20.5 0.9 NS Days of treatment 10.0 0.4 9.6 0.2 NS Ampules of gonadotropins 36.7 3.7 31.6 1.6 NS E 2 level (pg/ml) 4,035 489 5,271 613 NS No. of oocytes 22.9 0.9 23.1 1.6 NS Fertilization rate (%) 77.9 76.1 NS No. of embryos transferred 3.8 0.1 3.4 0.2 NS No. of embryos frozen 3.0 0.8 3.1 0.5 NS Pregnancy rate (%) 0 24.2.0001 Implantation rate (%) 0 8.5.0001 Note: Values are expressed as means SE unless otherwise indicated. NS not significant. was to select as many type I and type II embryos as possible; the remaining embryos were cryopreserved with 1, 2-propanediol and sucrose. Only patients with freshly transferred embryos were included in the study. Hormone Measurements Blood was drawn daily, starting with hcg administration (day 0) and continuing for 7 days (day 6). The samples were stored at 20 C in aliquots for subsequent E 2 analysis together in the same assay. Serum E 2 was measured using commercially available RIA kits (BioMérieux, Charbonniéres les Bains, France). The interassay and intraassay variability for E 2 at a concentration of 40 pg/ml was 2.8% and 4.3%, respectively. Statistical Analysis Data were expressed as means SEM. Comparisons between different s of the same patient (Tables 1 and 2) were performed with the use of the Student s t-test, except that PRs and implantation rates were compared with the use of the 2 test. Analysis of variance was used to compare the results of step-down versus standard protocols for different patients. Statistical significance was defined as P.05. Statistical analysis was performed with the use of the Statistical Package for the Social Sciences (SPSS Inc., Chicago, IL). RESULTS With the use of the step-down regimen, E 2 levels on the day of hcg administration were significantly lower in patients undergoing IVF compared with their own previous failed (1,919 477 versus 5,770 650 pg/ml, respectively; P.001). As expected, E 2 levels on the day of hcg administration in patients undergoing the same standard protocol as in the previous failed were similar (5,271 613 versus 4,035 489 pg/ml, respectively). Circulating levels of E 2 during the preimplantation period (days 0 6) with the step-down regimen also were decreased compared with the previous (Fig. 1). Estradiol levels were significantly (P.01) lower in terms of absolute values with the step-down regimen compared with the previous. Table 1 compares the outcome of 24 s of IVF with the use of the step-down regimens compared with the previous failed. The number of oocytes collected was lower with the step-down regimen than in the previous (18.1 2.1 versus 24.0 1.9, respectively, P.001). Similar fertilization rates were obtained (74.2% versus 71.2%, respectively), and the number of good-quality embryos transferred was comparable (3.3 0.2 versus 3.8 0.2, respectively). Nevertheless, the implantation rate and PR were 29.3% and 64.2%, respectively. Five s (17.2%) were cancelled because of an 80% decrease in E 2 levels, from 1,000 pg/ml to 200 pg/ml. Table 2 compares the reproductive outcome of 62 patients who underwent the same standard protocol as was used in their previous failed. All the parameters analyzed were similar to those of the previous (because the induction ovulation protocol was similar). However, the implantation rate and PR were 8.5% and 24.2%, respectively. No cancellations were recorded. In patients for whom at least one IVF attempt failed in which 3 4 good-quality embryos were transferred and E 2 levels were 3,000 pg/ml on the day of hcg administration, the step-down regimen and standard protocol were compared in the following (Table 3). Estradiol levels 236 Simón et al. Step-down regimen in high responders Vol. 70, No. 2, August 1998

FIGURE 1 Circulating levels of E 2 during the preimplantation period (days 0 6) in patients undergoing the step-down regimen (dashed line) versus the previous failed (solid line). Bars represent means SE. were significantly lower with the step-down regimen than with the standard protocol on the day of hcg administration (1,919 477 pg/ml versus 5,271 613 pg/ml, respectively; P.001), as well as throughout the preimplantation period. As expected, the number of oocytes collected was TABLE 3 Reproductive outcome of the step-down versus the standard protocol of IVF. Variable Step-down (n 24) Standard protocol (n 62) P value Age (y) 31.6 1.2 33.8 0.3 NS No. of previous failed s 1.2 0.1 1.0 0.0 0.02 GnRH analogue (d) 22.9 0.7 20.5 0.9 NS Days of treatment 10.6 0.1 9.6 0.2 NS Ampules of gonadotropins 22.4 0.8 31.6 1.6 NS E 2 level (pg/ml) 1,919 477 5,271 613.001 No. of oocytes 18.1 2.1 23.1 1.6.001 Fertilization rate (%) 74.2 76.1 NS No. of embryos transferred 3.3 0.2 3.4 0.2 NS No. of embryos frozen 2.5 0.7 3.1 0.5 NS Pregnancy rate (%) 64.2 24.2.0002 Implantation rate (%) 29.3 8.5.02 Hyperstimulated patients (%) 0 12.9.04 Note: Values are expressed as means SE unless otherwise indicated. NS not significant. lower with the step-down regimen than with the standard protocol (18.1 2.1 versus 23.1 1.6, respectively; P.001). Nevertheless, similar fertilization rates were obtained (74.2% versus 76.1%, respectively), and the number of good-quality embryos transferred was comparable (3.3 0.2 versus 3.4 0.2, respectively). However, the implantation rate and PR were significantly improved in patients who underwent the step-down regimen compared with those who were stimulated with the standard protocol (29.3% and 64.2% versus 8.5% and 24.2%, respectively; P.02 and P.0002, respectively). DISCUSSION Comparison of standard IVF and donor oocyte s reveals that COH, which is necessary to produce multiple embryos for IVF, inhibits embryo implantation by decreasing endometrial receptivity (3). This concept is proved further by analysis of implantation rates in patients undergoing IVF who donate some of their oocytes to HRT-treated recipients; with the same oocyte and embryo quality, endometrial receptivity and therefore implantation rates and PRs is better in the HRT-treated recipients (9). Ovarian hyperstimulation inhibits embryo implantation in humans and in mice (16) and rats (17). Controlled ovarian hyperstimulation is associated with supraphysiologic hormone levels compared with natural or HRT-treated s. High E 2 levels, which are known to be FERTILITY & STERILITY 237

interceptive (18), and altered E 2 progesterone ratios, which also are associated with the impairment of endometrial receptivity (19), are the main factors affecting endometrial receptivity in high responders (9, 10). These clinical observations are reinforced by the morphologic changes detected in the secretory endometrium of patients undergoing COH by histology (4) and scanning electron microscopy (5) as well as by the biochemical patterns found in the endometrial fluid in these patients compared with patients undergoing HRT (6). On the basis of these considerations, we designed a series of studies aimed at developing clinical strategies to improve endometrial receptivity while using COH to produce multiple embryos for IVF. In the present study, using high responders in whom endometrial receptivity is decreased significantly, we investigated the use of the step-down principle developed by Fauser et al. (15), which allowed us to stimulate follicles to full maturation using a decremental dose regimen while reducing the number of functionally active mature follicles. In addition, it has been shown that the different ovarian responses in high versus low responders to COH are related to the extent of FSH accumulation in the late follicular phase (20). Moreover, LH was not administered. The objective of this strategy was to produce a small FSH window. Initially, we had to reach the FSH threshold, then the doses of gonadotropins were reduced to obtain FSH levels that dropped below the FSH threshold. By mimicking the dynamics of FSH in natural s (21), we attempted to elicit the recruitment and maturation of a relatively small number of follicles while the rest of the cohort was driven to atresia; our main goal, however, was to decrease E 2 levels in this group of patients. As a consequence, patients in whom at least one IVF attempt failed in which 3 4 good-quality embryos were transferred and E 2 levels were 3,000 pg/ml on the day of hcg administration were voluntarily assigned either to repeat the same protocol or to undergo the step-down regimen. Estradiol levels on the day of hcg administration were significantly lower with the step-down regimen compared with the standard protocol (1,919 477 pg/ml versus 5,271 613 pg/ml, respectively, P.001). The implantation rate and PR were 29.3% and 64.2%, respectively, in patients undergoing the step-down regimen versus 8.5% and 24.2%, respectively, in those receiving the standard protocol (P.02 and P.0002, respectively). The standard protocol was our control to ensure that the results obtained in the next with the step-down approach were not simply because of the repetition of the procedure. These clinical results suggest that in high responders to COH, endometrial receptivity can be improved by lowering abnormally high E 2 levels. The major difficulty with the step-down regimen is to determine the individual FSH threshold in clinical practice because the threshold for administered FSH may differ as much as 10-fold for different patients (22). For this reason, five s (17.2%) were cancelled because the FSH threshold was not reached and follicle development was arrested, as demonstrated by an abrupt 80% decrease in the E 2 level. Although our experience is limited, all our cancellations were induced by decreasing the FSH dose to 1 or 1 ampule 2 after 6 days of treatment when extremely high follicle development was observed. Learning from this, we resolved the problem in the next. No cancellations were recorded when the FSH dosage was maintained at 1 1 ampules/d after day 6 of treatment, and it is known that maxi- 2 mum serum levels after IM injections correlate well with body weight (23). For an individualized approach, we need more knowledge about the pharmacokinetics of the purified and recombinant gonadotropins. The possible mechanism(s) responsible for our clinical observations remain unknown. However, there may be an altered regulation of steroid receptors in the human endometrium because in COH s a significant reduction in the nuclear receptors for progesterone and E 2 has been demonstrated in both glands and stroma (24). Further studies are needed to clarify this point. Moreover, this study does not consider secretory products other than ovarian steroids that may be altered. Nonetheless, our clinical data suggest that impaired endometrial receptivity in superovulated patients with high levels of E 2 can be improved by decreasing E 2 levels in the next, thereby increasing the implantation rate and PR. References 1. Assisted reproductive technology in the United States and Canada: 1994 results generated from the American Society for Reproductive Medicine/Assisted Reproductive Technology Registry. Fertil Steril 1996;66:697 705. 2. Conaghan J, Hardy K, Handyside AH, Winston RML, Leese HJ. Selection criteria for human embryo transfer: a comparison of pyruvate uptake and morphology. J Assist Reprod Genet 1993;10:21 30. 3. Paulson RJ, Sauer MV, Lobo RA. Embryo implantation after human in vitro fertilization: importance of endometrial receptivity. Fertil Steril 1990;53:870 4. 4. Garcia JE, Acosta AA, Hsiu JG, Jones HW. Advanced endometrial maturation after ovulation induction with human menopausal gonadotrophin/human chorionic gonadotrophin for in vitro fertilization. Fertil Steril 1984;41:31 7. 5. Bladford AJ, Najmabadi S, Paulson RJ. Ultrastructural characteristics of the luteal phase endometrium in donors undergoing controlled ovarian hyperstimulation. Fertil Steril 1997;67:625 30. 6. Simón C, Mercader A, Frances A, Gimeno MJ, Polan ML, Remohí J, et al. Hormonal regulation of serum and endometrial IL-1, IL-1, and IL-1ra: IL-1 endometrial microenvironment of the human embryo at the apposition phase under physiological and supraphysiological steroid level conditions. J Reprod Immunol 1996;31:165 84. 7. Pellicer A, Ruiz A, Castellví RM, Calatayud C, Ruiz M, Tarín JJ, et al. Is the retrieval of high numbers of oocytes desirable in patients treated with gonadotrophin-releasing hormone analogues (GnRHa) and gonadotrophins? Hum Reprod 1989;4:536 40. 8. Tarín JJ, Pellicer A. Consequences of high ovarian response to gonadotrophins: a cytogenetic analysis of unfertilized human oocytes. Fertil Steril 1990;54:665 70. 9. Simón C, Cano F, Valbuena D, Remohí J, Pellicer A. Clinical evidence for a detrimental effect on uterine receptivity of high serum estradiol levels in high and normal responder patients. Hum Reprod 1995;10:2432 7. 10. Pellicer A, Valbuena D, Cano F, Remohí J, Simón, C. Lower implantation rates in high responders: evidence for an altered endocrine milieu during the preimplantation period. Fertil Steril 1996;65:1190 5. 238 Simón et al. Step-down regimen in high responders Vol. 70, No. 2, August 1998

11. Testart J, Belaisch-Allart J, Forman R, Gazengel A, Hazout A, Strubb N, et al. Influence of different stimulation treatments on oocyte characteristics and in vitro fertilizing ability. Hum Reprod 1989;4:192 7. 12. Toner J, Brzyski R, Oehninger S, Veeck L, Simonetti S, Muasher S. Combined impact of the number of preovulatory oocytes and cryopreservation on IVF outcome. Hum Reprod 1991;6:284 9. 13. Chenette PE, Sauer MV, Paulson RJ. Very high serum estradiol levels are not detrimental to clinical outcome of in vitro fertilization. Fertil Steril 1990;54:858 63. 14. Tarín JJ, Sampaio MC, Calatayud C, Castellví RM, Bonilla-Musoles F, Pellicer A. Relativity of the concept high responder to gonadotrophins. Hum Reprod 1992;7:19 22. 15. Fauser BCJM, Donderwinkel P, School DC. The step-down principle in gonadotrophin treatment and the role of GnRH analogues. Baillieres Clin Obstet Gynaecol 1993:7:309 30. 16. Fossum GT, Davidson A, Paulson RJ. Ovarian hyperstimulation inhibits embryo implantation in the mouse. J In Vitro Fertil Embryo Transfer 1989;6:7 10. 17. Smith MG. On the interruption of pregnancy in the rat by injection of ovarian follicular extract. Johns Hopkins Hospital Bulletin 1926;39: 203 14. 18. McLean-Morris J, Van Wagenen G. Interception: the use of the preovulatory estrogens to prevent implantation. Am J Obstet Gynecol 1973;115:101 6. 19. Gidley-Baird AA, O Neil C, Sinosich MJ, Porter RN, Pike IL, Saunders DM. Failure of implantation in human in vitro fertilization and embryo transfer patients: the effects of altered progestrone/estrogen ratios in humans and mice. Fertil Steril 1986;45:69 75. 20. Ben Rafael Z, Strauss JF, Mastroianni L, Flickinger GL. Differences in ovarian stimulation in HMG stimulated women may be related to FSH accumulation. Fertil Steril 1986;46:586 92. 21. Hillier SG. Regulation of follicle estrogen biosynthesis: a survey of current concepts. J Endocrinol 1981;89:3 18. 22. Brown JB. Pituitary control of ovarian function: concepts derived from gonadotrophin therapy. Aus N Z J Med 1978;18:47 54. 23. Mannaerts B, Shoham Z, Schoot B, Bouchard P, Harlin J, Fauser B, et al. Single-dose pharmacokinetics and pharmacodynamics of recombinant human FSH (ORG 32489*) in gonadotropin-deficient volunteers. Fertil Steril 1993;59:108 14. 24. Hadi FH, Chantler E, Anderson E, Nicholson R, McClelland RA, Seif MW. Ovulation induction and endometrial steroid receptors. Hum Reprod 1994;9:2405 10. FERTILITY & STERILITY 239