A. Leader, M.D., for the Monofollicular Ovulation Induction Study Group

Transcription

1 Improved monofollicular ovulation in anovulatory or oligo-ovulatory women after a low-dose step-up protocol with weekly increments of 25 international units of follicle-stimulating hormone A. Leader, M.D., for the Monofollicular Ovulation Induction Study Group The Ottawa Fertility Centre, Division of Reproductive Medicine, Department of Obstetrics and Gynecology, Faculty of Medicine, University of Ottawa, Ottawa, Ontario, Canada Objective: To compare the effectiveness and efficiency of two low-dose step-up protocols for ovulation induction in women with anovulatory infertility (World Health Organization group II). Design: Open-label, prospective, randomized, group-comparative, multicenter study. Setting: Eighteen infertility centers in Europe and Canada. Patient(s): One hundred fifty-eight anovulatory or oligo-ovulatory infertile women. Intervention(s): Patients were randomly assigned to one of two protocols for one cycle of follitropin beta (rfsh) using a pen device. The starting dosage was 50 IU/day for 7 days. In the absence of follicles 12 mm, the daily dosage was increased by either 25 or 50 IU per week. Main Outcome Measure(s): The percentage of all subjects treated who ovulated after one treatment cycle (efficacy) and the total rfsh dose to reach ovulation (efficiency). Result(s): The 25-IU group had a higher incidence of monofollicular growth (41.3% of 80 vs. 21.8% of 78 women) and ovulation (81.3% vs. 60.3%), a lower cumulative rfsh dose (887 IU vs. 984 IU), and fewer cancellations due to hyperresponse ( 3 follicles 15 mm; 5.0% vs. 20.5%). Both protocols were well tolerated. Conclusion(s): Weekly increments of 25 IU in the daily dose were more effective and efficient than 50-IU increments. (Fertil Steril 2006;85: by American Society for Reproductive Medicine.) Key Words: Ovulation induction, low-dose step-up protocol, recombinant FSH, WHO group II anovulatory infertility Clomiphene citrate is the standard initial treatment for infertility in women with World Health Organization (WHO) group II anovulation. This treatment restores ovulation in approximately 70% of patients, and 35% achieve pregnancy within 6 months (1). Conception rates, however, are considerably lower with clomiphene citrate than with gonadotropin therapy, possibly because of the antiestrogenic effects of clomiphene citrate on cervical mucus and the endometrium or because of poorly monitored treatment cycles (2). Gonadotropin therapy is generally used in clomiphene citrate resistant patients and in those patients who do not conceive after repeated courses of clomiphene citrate (2, 3). The aim of ovulation induction with gonadotropins is to find the threshold dose of FSH required to develop a single preovulatory follicle and to avoid multifollicular recruitment (4, 5). As the ovarian threshold for FSH response varies Received July 13, 2005; revised and accepted November 26, Financial support received from NV Organon, Oss, the Netherlands. Best Clinical Paper at the 49th meeting of the Canadian Fertility and Andrology Society, which was held in Victoria, British Columbia, in 2003, and at the 59th annual meeting of the American Society of Reproductive Medicine, which was held in San Antonio, Texas, in Reprint requests: Arthur Leader, M.D., 955 Green Valley Crescent, Ottawa, Ontario, Canada K2C 3V4 (FAX: ; address: aleader@conceive.ca). among individuals (5), step-up protocols have gained wide acceptance (6 8). The major risk of some step-up protocols is the development of many follicles leading to cycle cancellation, severe ovarian hyperstimulation syndrome (OHSS), or multiple pregnancies (2, 7, 9, 10). To reduce the risk of these complications, various low-dose step-up and step-down regimens have been studied (10 14). In the low-dose step-up regimens, the amount of FSH used is reduced by using longer time intervals between dose increases and/or smaller dose increments (37.5 IU or 50 IU instead of 75 IU). These regimens have proved to be successful in reducing the risk of ovarian hyperstimulation while maintaining a satisfactory pregnancy rate (7, 10). However, these protocols were imprecise, as it was necessary for the patient to self-administer half or two-thirds of a 75-IU ampoule of urinary gonadotropins. With the availability of recombinant FSH (rfsh) preparations in vials containing 50 IU (follitropin beta) or 37.5 IU (follitropin alfa), it was easier for patients to use smaller doses. More recent studies (15, 16) have shown that a starting dose of 50 IU is adequate to induce ovulation. However, maintaining the 50-IU starting dose or even lower doses for 2 weeks may require that the treatment period be extended to more than 20 days to achieve ovulation (17) Fertility and Sterility Vol. 85, No. 6, June /06/$32.00 Copyright 2006 American Society for Reproductive Medicine, Published by Elsevier Inc. doi: /j.fertnstert

2 The present trial aimed to refine the rfsh regimen further. We evaluated and compared the efficacy and efficiency of two low-dose step-up protocols with rfsh (follitropin beta, which is marketed as Follistim (Organon USA, West Orange, NJ) in the United States and Puregon (NV Organon, Oss, The Netherlands) in other countries). Both regimens started with daily doses of 50 IU for 7 days. At weekly intervals, the daily dosage was increased (by 25 IU for one group and 50 IU for the other) for women without a follicle at least 12 mm in diameter. The Puregon Pen (Follistim Pen), which can adjust the dose of gonadotropins in 25-IU increments, was used as the self-injection device. MATERIALS AND METHODS Patients This study was funded by an unrestricted educational grant from Organon. Eighteen infertility centers in eight European countries and Canada (see Acknowledgements) participated in the recruitment of patients. The Independent Ethics Committee of each center approved the trial protocol. The study period lasted from June 2000 to January Eligible women had to be anovulatory or oligo-ovulatory (WHO group II infertility), with positive progestagen withdrawal bleeding or spontaneous menstrual bleeding. The women had to have been infertile for at least 1 year before randomization and had to have had no ovulation or conception during at least three preceding clomiphene citrate cycles. However, patients could not receive clomiphene citrate or gonadotropins within the 30 days before the start of study treatment. The women had to be years of age at screening and in good physical and mental health. Body mass index had to be no less than 18 kg/m 2 and no more than 33 kg/m 2. Patients had to have a normal uterine cavity, as documented by recent (within 3 years) hysteroscopy, hysterosalpingography, or sonohysterography assessment at screening. To exclude women with androgen-secreting tumors, the women had to have serum total T 5.0 nmol/l [the upper limit of normal for healthy nonhirsute eumenorrheic women is 2.94 nmol/l (18)] or free T levels within the local laboratory s reference range. The woman s male partner s semen had to have at least progressively motile sperm cells per milliliter, as documented no more than 2 years before randomization. Women were ineligible if they were pregnant or lactating; if they had previously been hospitalized because of OHSS; if they had untreated hyperprolactinemia; or if they had tumors of the ovary, breast, uterus, pituitary, or hypothalamus. Women with a gynecologic condition that is incompatible with pregnancy (e.g., severe uterine fibroid tumors or sexual organ malformation) would also be excluded. Other exclusion criteria included undiagnosed vaginal bleeding, primary ovarian failure, ovarian cysts or enlarged ovaries (unrelated to polycystic ovarian disease), and current or recent (past 12 months) history of drug or alcohol abuse. A pregnancy test was performed before the start of treatment. The study was approved by the Ethics Committee of each participating center. All patients gave written informed consent. This investigation was performed according to the Declaration of Helsinki, the International Conference on Harmonization guidelines, and Good Clinical Practice. Study Design This was an open-label, randomized, group-comparative, multicenter study to assess the effects of two different ovulation-induction regimens. Patients who met all selection criteria were randomly assigned to one of the two treatment groups by receiving a subject code number from a randomization list. Each center received this sequence of code numbers and allocated these numbers to the patients in the order of enrollment in the trial. Both regimens began with a daily dose of 50 IU of follitropin beta for 7 days. Weekly increments in the daily dose (25-IU increment in one regimen, 50-IU increment in the other) were permitted only for women who had no follicle of at least 12 mm. The patients administered the drug themselves, using the pen injection device supplied with cartridges containing 300 IU follitropin beta. Treatment started within 3 days after the initiation of a spontaneous or a progestagen-induced withdrawal menses. Before the first injection, an ultrasound scan was performed to assess follicular activity and serum levels of FSH, LH, and E 2 were measured. The ultrasound and measurements of LH and E 2 were repeated on day 7 of treatment and every 2 3 days thereafter, up to and including the day of administration of hcg. On at least two occasions (5 10 days after the day of hcg administration), serum P was measured to confirm ovulation ( 25 nmol/l, as measured by the local laboratory). Ovulation could also be confirmed by ultrasound. Treatment was continued until one follicle of 18 mm was seen. Then hcg [10,000 IU of Pregnyl (NV Organon, Oss, The Netherlands)] was given as a single SC or IM injection to trigger ovulation. Subsequently, intrauterine insemination (IUI) could be undertaken. The hcg was withheld if there was an insufficient ovarian response after 35 days of treatment or in the case of ovarian hyperresponse ( 3 follicles of 15 mm). In the latter instance, conversion to IVF-ET was allowed. Treatment in this study was limited to one cycle only (with a maximum of 35 days), and the trial period ended with menstrual bleeding, a miscarriage, or clinical proof of an ongoing pregnancy. In the event of a confirmed ovulation, patients were asked to visit the clinic 3 6 weeks after the day of hcg administration or the last follitropin beta injection to detect a possible pregnancy by ultrasound. Pregnancy could also be assessed by measurement of serum hcg. If the woman was pregnant, ultrasound was performed at weeks to document an ongoing pregnancy. Fertility and Sterility 1767

3 FIGURE 1 Results for primary efficacy (A) and efficiency (B) outcome measures (all treated subjects). Endpoints The primary outcome measures were the percentage of all subjects treated who ovulated after one treatment cycle (efficacy) and the total follitropin beta dose to reach ovulation (efficiency). Women were considered to have ovulated if at least one serum progesterone (P) measurement after follitropin beta treatment was above the threshold value ( 25 nmol/l); if they became pregnant or had an ectopic pregnancy or miscarriage; or if other evidence for ovulation was gathered (e.g., by ultrasound). Patients who left the study and continued with IVF-ET were not considered to have ovulated. Secondary outcome measures included the percentage of women with a monofollicular cycle; the percentage of cycles canceled because of ovarian hypo- or hyperresponse; the total number of treatment days; the total number of follicles 2, 14, 16, and 18 mm; serum levels of E 2, FSH, and LH on the day of hcg administration or 1 3 days earlier; the size of the last follitropin beta dose; the ongoing pregnancy rate; and the multiple gestation rate. A cycle was classified as monofollicular if there was exactly one follicle 16 mm and no other follicles 12 mm. Throughout the study, safety was evaluated by monitoring the occurrence and nature of adverse events (AEs) and the incidence of OHSS. OHSS was graded as mild, moderate, or severe. Statistical Analysis For the efficacy and efficiency outcome measures, two-sided 95% confidence intervals (CI) were calculated for the overall treatment difference. The Cochran method stratified for center was used for nondichotomous parameters, and the Cochran- Mantel-Haenszel approach stratified for center was used for dichotomous parameters. With 90 evaluable patients in each treatment group, the ovulation rate could be estimated with a precision (SE) of approximately 5%. RESULTS Patients Of the 161 patients who met the inclusion criteria requirements, 158 women started treatment. Three women (all assigned to the 25-IU group) discontinued before treatment: one because of spontaneous pregnancy, one because of social problems, and one because of follicular cysts. Of the 158 women who began treatment, 80 patients were assigned to the 25-IU dose increment group and 78 patients to the 50-IU dose increment group. A total of 40 patients prematurely discontinued treatment. The main reason for discontinuation was the risk of ovarian hyperstimulation (five in the 25-IU group and 16 in the 50-IU group). Three women discontinued because of AEs: two of the women in the 25-IU group because of mild to moderate menstruation-like bleeding and one woman in the 50-IU group because of a moderate fever. In each group, one woman discontinued because of insufficient response and one because of spontaneous ovulation. Among the hcg-treated patients, seven women (two in the 25-IU group and five in the 50-IU group) had more than three follicles of 15 mm and continued with IVF. Three other women had too many follicles but did not continue with IVF; their discontinuations were classified as due to other reasons. One woman had an hcg injection but did not have IUI because of high body mass index. In total, 118 women completed the trial (68 in the 25-IU dose increment group and 50 in the 50-IU dose increment group). Table 1 shows a summary of the demographic, menstrual, and fertility characteristics of all women who started treatment. No clinically relevant differences were observed between the two groups. The baseline hormone levels were comparable in the two groups (Table 2). Primary Efficacy and Efficiency Parameters Figure 1 presents the ovulation rate and the mean total dose 1768 Leader Improved monofollicular ovulation Vol. 85, No. 6, June 2006

4 TABLE 1 Demographic, menstrual, and fertility characteristics, all-subjects-treated population. 25-IU group (n 80) 50-IU group (n 78) Demographic characteristics, mean SD Age (y) Height (cm) Weight (kg) Body mass index (kg/m 2 ) Menstrual cycle characteristics, n (%) Amenorrhea 22 (27.5) 27 (34.6) Cycle length 41 days 33 (41.3) 33 (42.3) Cycle length 41 days 25 (31.3) 18 (23.0) Female fertility characteristics, mean SD Early abortions n Parity n Duration of infertility (y) Male fertility characteristics, mean SD Sperm count (cells 10 5 ml) Percentage of motile sperm (%) of follitropin beta in both treatment groups. More of the women in the 25-IU group than in the 50-IU group ovulated (65 [81.3%] vs. 47 [60.3%]). The estimated treatment difference (adjusted for center) of 18.6 percentage points was statistically significant (P.009; 95% CI of difference, percentage points). With an ovulation rate of 81.3% in the 25-IU group and 60.3% in the 50-IU group, 4.8 subjects would need to be treated with the 25-IU increment regimen to obtain one more ovulation. Among the hcgtreated patients, the ovulation rate was 89.9% for the 25-IU group and 78.9% for the 50-IU group. The mean ( SD) total dose of follitropin beta administered was lower in the 25-IU group than in the 50-IU group ( IU vs IU). The estimated treatment difference (adjusted for center) was also statistically significant ( IU; P.013; 95% CI of difference, to 34.3). Secondary Efficacy Parameters Table 3 shows the results and statistical analyses on most secondary efficacy outcome measures. The percentage of patients with a monofollicular cycle was nearly twice as high in the 25-IU group as in the 50-IU group (41.3% vs. 21.8%, respectively, of all subjects treated). The rate of cancellation due to ovarian hyperresponse was considerably lower in the 25-IU group, which also had a lower mean threshold dose (the last dose of follitropin beta before hcg administration) and a lower mean serum E 2 level. The mean serum LH level was higher in the 25-IU group. However, that difference was largely due to three outliers (showing sharply increasing LH values) in one center. Excluding this center from the analysis resulted in a minor difference between the two treatment groups (7.16 and 6.28 U/L, respectively), which was no longer statistically significant. About 30% of patients in both treatment groups remained at the starting dose of 50 IU. More than half of the women in both groups increased the TABLE 2 Baseline hormone values, all-subjectstreated population. 25 IU (n 80) 50 IU (n 78) PRL (mu/l) a PRL ( g/l) a TSH (mu/l) FSH (U/L) LH (U/L) T total (nmol/l) b Free T (nmol/l) b E 2 (pmol/l) Pr (nmol/l) a PRL was measured in mu/l in nine centers, in g/l in seven centers, and in either mu/l or g/l in one center. b Two centers measured free T, whereas total T was measured in the other centers. Fertility and Sterility 1769

5 TABLE 3 Results on secondary parameters, all-subjects-treated population. Parameter 25 IU group (n 80) 50 IU group 95% CI of (n 78) Difference a difference P Monofollicular cycle rate, % of AST Mean no. of treatment days to 1.0 NS Mean threshold dose a of rfsh (IU) to Cancellation rate due to Low ovarian response, % of AST to 7.0 NS Ovarian hyperresponse, % of AST to Mean serum E 2 (pmol/l) b to Mean serum FSH (U/L) b to 0.86 NS Mean serum LH (U/L) b No. of miscarriages (% of AST, % of pregnancies) 2 (2.5, 11) 3 (3.8, 23) Ongoing pregnancy rate, n (% of AST) 16 (20.0) 10 (12.8) to 17.9 NS Multiple gestation rate, n (% of AST, % of ongoing pregnancies) 2 (2.5, 12.5) 0 Note: NS not significant; AST all-subjects-treated. P.05. a Estimated treatment difference adjusted for center using the Cochran (Mantel-Haenszel) approach. b Measured on the day of hcg injection or 1 3 days earlier; restricted to hcg-treated subjects. dose once during treatment; similar percentages ( 10%) in both groups had two or three dose increments. No statistically significant differences were found for the cancellation rate due to low response, the total number of treatment days, and the mean serum FSH level. Nor was the difference in ongoing pregnancy rate (20.0% in the 25-IU group vs. 12.8% in the 50-IU group) statistically significant. No ectopic pregnancies occurred. There were two multiple pregnancies (one twin and one triplet), both in women in the 25-IU group. Both of the women with multiple pregnancies had one 14-mm follicle, one 15-mm follicle, and one follicle 18 mm at the last ultrasound examination. Figure 2 shows the mean number of follicles in each of the four size classes, measured on the day of hcg administration FIGURE 2 Number and size of follicles on day of hcg administration or 1 3 days earlier. (A) All subjects who received hcg treatment (n 124) and (B) all patients who received hcg but did not convert to IVF (n 118) Leader Improved monofollicular ovulation Vol. 85, No. 6, June 2006

6 FIGURE 3 Cancellation rate due to ovarian hyperresponse by treatment period (all-subjects-treated). days. Two women in the 25-IU group and one woman in the 50-IU group had mild OHSS. Three women left the study because of the development of an AE: in two cases because of a menstruation-like bleeding irregularity (25-IU group) classified as mild and moderate and both possibly related to study medication; in one case because of fever (50-IU group) classified as moderate and not related to study medication. (or 1 3 days earlier). All women who received hcg were taken into account, including (Fig. 2A) and excluding (Fig. 2B) those who converted to an IVF cycle. More follicles were observed in the 50-IU group, especially in the size classes of mm (Fig. 2A). In this 50-IU group, therefore, more women converted to an IVF cycle after the hcg injection (five vs. two women). If these seven women were excluded, the total number of follicles was only slightly higher in the 50-IU group and the distribution of follicles over the different size classes was comparable for both treatment groups (Fig. 2B). Figure 3 shows the cancellation rate due to ovarian hyperresponse by treatment period. A considerably higher number of women in the 50-IU group developed an ovarian hyperresponse: 20.5 % versus 5.0% in the 25-IU group (P.004). Safety The incidence of AEs was comparable in the two treatment groups. In total, 12.7% of the women reported an AE that was considered possibly, probably, or definitely related to study medication (10.0% in the 25-IU group and 15.4% in the 50-IU group). The most frequently reported AEs were headache (11.3% in the 25-IU group and 9.0% in the 50-IU group) and abdominal pain (gynecological) (12.5% and 9.0%, respectively). Two patients (one in the 25-IU group and one in the 50-IU group) were hospitalized because of OHSS, which is classified as a serious adverse event (SAE). Both patients recovered from their SAE, although in one case recovery took 40 DISCUSSION This study shows that both dose regimens with rfsh (follitropin beta) are effective and safe for inducing ovulation in anovulatory or oligo-ovulatory women (WHO group II). The use of smaller dose increments (25 IU instead of 50 IU) resulted in a higher ovulation rate (81.3% vs. 60.3% of all subjects treated), with a lower total rfsh requirement and a lower threshold dose. The regimen with the 25-IU increments was therefore more effective and efficient than the regimen with the 50-IU increments. These results support the concept that a follitropin beta regimen with weekly dose increments of only 25 IU clearly reduces the risk of hyperstimulation in patients who are particularly sensitive to gonadotropins. The rate of cancellation due to ovarian hyperresponse was clearly reduced in the 25-IU group of our study (5.0% vs. 20.5% in the 50-IU group; P.004). However, the lower cancellation rate did not account for the entire higher incidence of ovulation in the 25-IU group. The incidence of monofollicular cycles among patients who ovulated did not differ between the 25-IU and 50-IU groups. Over the past decade, protocols with varying doses of FSH have been studied for ovulation induction. The common objective of these studies was to achieve maturation and ovulation of a single dominant follicle, without inducing ovarian hyperstimulation. Today, the chronic low-dose step-up regimen is the most frequently used. Compared with other conventional step-up regimens, the chronic low-dose step-up regimen of FSH tends to produce a higher rate of monofollicular development and slightly improved pregnancy rates (7, 10). In addition, the occurrence of severe OHSS was avoided (7) and the multiple pregnancy rate was clearly reduced (7, 10). However, these low-dose step-up regimens using a starting dose of 75 IU FSH for 14 days can result in overstimulation when the more potent premixed recombinant FSH preparations are used. About 10% 20% of cycles are canceled because of ovarian hyperresponse (11, 19, 20, 21), and the development of a single dominant follicle was achieved in only about 50% (37% 79%) of noncanceled cycles (19 21). In the large multicenter studies using the chronic low-dose step-up regimen with urinary FSH (ufsh) or rfsh for ovulation induction, the ovulation rates are around 60% of noncanceled cycles (10, 22) and around 70% of all started cycles (11). Cumulative ongoing pregnancy rates (observed Fertility and Sterility 1771

7 after a maximum of three treatment cycles) reported in these studies are about 45% of noncanceled cycles (22) and about 25% of all started cycles (11). The efficacy results of clinical studies on ovulation induction are highly dependent on the population used for analyses, the number of treatment cycles, and the way the data are presented. The present study used intent-to-treat analysis and expressed the outcome measures (e.g., ovulation rate, monofollicular cycle rate, and pregnancy rate) as percentages of all started cycles (the study included only one treatment cycle). In contrast, most of the above-mentioned studies on the standard low-dose step-up regimens did not use intent-totreat analyses (7, 10, 20 22) and presented cumulative results after a maximum of three cycles (7, 11, 20 22). Therefore, results of the present study are difficult to compare with these studies of the standard low-dose step-up regimen. Results of two previous studies (15, 16) had suggested that a starting dose of 50 IU rfsh (followed by weekly dose increments of 50 IU, if needed) is sufficient to initiate follicular development in clomiphene citrate resistant patients. The ovulation rates reported in these studies (6/11 [55%] and 90/125 [72%] of started cycles, respectively) were in line with those reported in the 50-IU group in our study. A small, randomized cross-over study (17) compared the follicular response of two low starting doses of rfsh (50 IU of follitropin beta vs IU of follitropin alfa). These daily dosages were maintained for 2 weeks; if there was no evidence of follicular development, the daily dose was increased by 50-IU (follitropin beta) or 37.5 IU (follitropin alfa) at weekly intervals. There was an interval of at least a month between treatment cycles. Fifteen clomiphene citrate resistant patients received both treatment regimens. All treatment cycles were ovulatory. Sixty percent of the follitropin beta cycles and 67% of the follitropin alfa cycles were monofollicular; the difference was not statistically significant. The difference in magnitude of the results between the latter and the three other studies may be explained by the small sample size of 15 women, in which none of the cycles were canceled. Two recent studies of women with polycystic ovary syndrome (PCOS) provide further evidence that a 50-IU starting dose of follitropin beta and weekly 25-IU increments in the daily dose provides improved efficacy with a lower risk of OHSS. One of these studies (14) compared a step-up regimen versus a step-down regimen, both with follitropin beta, in 83 women with PCOS. The step-up regimen used a starting dosage of 50 IU daily for up to 14 days, followed by weekly increments of 25 IU in the daily dosage, up to a maximum daily dosage of 100 IU in the first cycle. The following two cycles could start at a dosage of 75 IU/day and could increase to a daily dosage of 125 IU/day. In the step-down regimen, a starting dosage of 100 IU was used until a follicle of 9 mm was detected, then the dosage was decreased to 75 IU/day for 3 days and then to 50 IU/day until the day before hcg administration. If no follicular development was evident after day 5, the daily dosage was increased to 150 IU. When a follicle of 9 mm was present, the dosage was decreased to 125 IU/day for 3 days, then 100 IU/day for 3 days, and then 75 IU/day until the day before hcg administration. The step-up protocol had a higher incidence of monofollicular growth (68.2% of cycles vs. 32% of cycles; P.0001) and a higher ovulation rate (70.3% of cycles vs. 61.7% of cycles; P.02). The other study (23) compared two low-dose step-up regimens of rfsh in 100 clomiphene citrate resistant women with PCOS: one group received follitropin alpha (starting dose of 75 IU and dose increments of 37.5 IU), and the other received follitropin beta (starting dose of 50 IU and dose increments of 25 IU). The difference in the percentage of women who ovulated (89% for follitropin alpha and 98% for follitropin beta) was not statistically significant. However, the follitropin beta group had more follicles 16 mm (2.1 vs. 1.6, P.05) and fewer canceled cycles (4% vs. 15.7%, P.05) than the follitropin alpha group had. These two new studies further confirm our findings that the starting dose of 50 IU follitropin beta, followed by weekly dose increments as low as 25 IU, is an optimal treatment alternative, further minimizing the risk for ovarian hyperstimulation. In the present study, one woman (1.3%) in each group was hospitalized because of OHSS. This indicates that both dose regimens are adequate to minimize the risk of OHSS in a group of patients particularly sensitive to it. The low occurrence of multiple gestations (only two cases) supports the importance of careful monitoring of follicle development during ovarian stimulation in PCOS patients. A surprising finding was the relatively low incidence of miscarriage (11.1% of pregnancies) in our 25-IU group. Most large series of pregnant patients with PCOS report an early miscarriage rate of 20% 30% (7). However, a similarly low miscarriage rate (15%) was also observed in a small study of the sequential step-up and step-down regimen (20). This study suggests that both the cumulative follitropin beta dose needed and the threshold follitropin beta dose are reduced if smaller dose increments are used (25 IU instead of 50 IU). The mean duration of treatment was similar (13 14 days) in both dose groups. This treatment period is comparable to those reported with other regimens using either ufsh or rfsh (24). The pen injection device used in this study is a convenient way for patients to give themselves precise doses of follitropin beta in increments of 25 IU. In conclusion, a low-dose step-up protocol for follitropin beta (administered with the pen injection device), with a starting dose of 50 IU rfsh plus smaller (25 IU as opposed to 50 IU) weekly increments (in the absence of a follicle at least 12 mm in diameter), produces a more controlled ovarian stimulation suited for ovulation induction. The protocol with the smaller weekly increments was more effective and efficient, as shown by the greater percentage of women with monofollicular growth, the higher percentage of women who ovulated, 1772 Leader Improved monofollicular ovulation Vol. 85, No. 6, June 2006

8 the lower rate of cancellation due to a hyperresponse, and the lower total dose of rfsh (follitropin beta) used. Acknowledgments: The participating IVF study centers and investigators of the rfsh Dose Increments for Ovulation Induction Study Group: Canada: Fertility Centre, Ottawa (A. Leader) Denmark: Herlev Amtssygehus, Herlev (S. Lindenberg); Fertility Clinic, Brædstrup Hospital, Brædstrup (F. Hald) France: Jean Verdier Hospital, Bondy (J.-N. Hugues); private practice, Paris (S. Alvarez); private practice, Paris (M. Savale, now deceased) Germany: Department of Gynecology, Endocrinology, and Reproductive Medicine, University of Heidelberg (T. Strowitzki); Otto von Guericke University Magdeburg, Magdeburg (J. Kleinstein); private practice, Munich (H. K. Rjosk) Norway: St. Olavs Hospital, Trondheim (S.B. Kjøtroslash;d); Fylkessjukehuset i Haugesund, Haugesund (R. Kolvik) Spain: La Paz University Hospital, Madrid (M. Cuadrado Mangas) Sweden: Women s Clinic, Sahlgrenska University Hospital, Göteborg (B. Edén); Carl von Linné Clinic, Uppsala (T. Bergh) The Netherlands: Catharina Hospital, Eindhoven (B. C. Schoot); University Medical Center Utrecht, Utrecht (P. van Zonneveld) United Kingdom: Leeds General Infirmary, Leeds (A. H. Balen); Hope Hospital, Manchester (H. M. Buckler) REFERENCES 1. Macgregor AH, Johnson JE, Bunde CA. Further clinical experience with clomiphene citrate. Fertil Steril 1968;19: Balen A. Ovulation induction optimizing results and minimizing risks. Hum Fertil (Camb) 2003;6(2 Suppl):S42 S Farhi J, Homburg R, Lerner A, Ben Rafael Z. The choice of treatment for anovulation associated with polycystic ovary syndrome following failure to conceive with clomiphene. Hum Reprod 1993;8: Scheele F, Schoemaker J. The role of follicle-stimulating hormone in the selection of follicles in human ovaries: a survey of the literature and a proposed model. Gynecol Endocrinol 1996;10: van der Meer M, Hompes PG, Scheele F, Schoute E, Veersema S, Schoemaker J. Follicle stimulating hormone (FSH) dynamics of low dose step-up ovulation induction with FSH in patients with polycystic ovary syndrome. Hum Reprod 1994;9: Franks S, Polson DW, Sagle M. Low-dose gonadotropin regimens for induction of ovulation. In: Filicori M, Flamigni C, eds. Proceedings of the Symposium on Ovulation Induction: Basic science and clinical advances. January 20 22, 1994, Palm Beach, Florida. Amsterdam: Excerpta Medica, Homburg R, Levy T, Ben Rafael Z. A comparative prospective study of conventional regimen with chronic low-dose administration of folliclestimulating hormone for anovulation associated with polycystic ovary syndrome. Fertil Steril 1995;63: White DM, Polson DW, Kiddy D, Sagle P, Watson H, Gilling-Smith C, et al. Induction of ovulation with low-dose gonadotropins in polycystic ovary syndrome: an analysis of 109 pregnancies in 225 women. J Clin Endocrinol Metab 1996;81: Wang CF, Gemzell C. The use of human gonadotropins for the induction of ovulation in women with polycystic ovarian disease. Fertil Steril 1980;33: Hedon B, Hugues JN, Emperaire JC, Chabaud JJ, Barbereau D, Boujenah A, et al. A comparative prospective study of a chronic low dose versus a conventional ovulation stimulation regimen using recombinant human follicle stimulating hormone in anovulatory infertile women. Hum Reprod 1998;13: Coelingh Bennink HJ, Fauser BC, Out HJ. Recombinant folliclestimulating hormone (FSH; Puregon) is more efficient than urinary FSH (Metrodin) in women with clomiphene citrate resistant, normogonadotropic, chronic anovulation: a prospective, multicenter, assessor-blind, randomized, clinical trial. European Puregon Collaborative Anovulation Study Group. Fertil Steril 1998;69: Schoot DC, Hop WC, de Jong FH, van Dessel TJ, Fauser BC. Initial E 2 response predicts outcome of exogenous gonadotropins using a stepdown dose regimen for induction of ovulation in polycystic ovary syndrome. Fertil Steril 1995;64: van Santbrink EJ, Fauser BC. Urinary follicle-stimulating hormone for normogonadotropic clomiphene-resistant anovulatory infertility: prospective, randomized comparison between low dose step-up and stepdown dose regimens. J Clin Endocrinol Metab 1997;82: Christin-Maitre S, Hugues JN. A comparative randomized multicentric study comparing the step-up versus step-down protocol in polycystic ovary syndrome. Hum Reprod 2003;18: Hayden CJ, Rutherford AJ, Balen AH. Induction of ovulation with the use of a starting dose of 50 units of recombinant human folliclestimulating hormone (Puregon). Fertil Steril 1999;71: Hoomans E, Voortman G. Efficacy and efficiency of ovulation induction in anovulatory women using a low-dose step-up scheme with recombinant FSH (Puregon). Hum Reprod 1999;14:327 [abstract book 1]. 17. Balasch J, Fabregues F, Creus M, Casamitjana R, Puerto B, Vanrell JA. Recombinant human follicle-stimulating hormone for ovulation induction in polycystic ovary syndrome: a prospective, randomized trial of two starting doses in a chronic low-dose step-up protocol. J Assist Reprod Genet 2000;17: Knochenhauer ES, Key TJ, Kahsar-Miller M, Waggoner W, Boots LR, Azziz R. Prevalence of the polycystic ovary syndrome in unselected black and white women of the southeastern United States: a prospective study. J Clin Endocrinol Metab 1998;83: Balasch J, Fabregues F, Creus M, Puerto B, Penarrubia J, Vanrell JA. Follicular development and hormone concentrations following recombinant FSH administration for anovulation associated with polycystic ovarian syndrome: prospective, randomized comparison between low-dose step-up and modified step-down regimens. Hum Reprod 2001;16: Hugues JN, Cedrin-Durnerin I, Avril C, Bulwa S, Herve F, Uzan M. Sequential step-up and step-down dose regimen: an alternative method for ovulation induction with follicle-stimulating hormone in polycystic ovarian syndrome. Hum Reprod 1996;11: Yarali H, Bukulmez O, Gurgan T. Urinary follicle-stimulating hormone (FSH) versus recombinant FSH in clomiphene citrate-resistant, normogonadotropic, chronic anovulation: a prospective randomized study. Fertil Steril 1999;72: Loumaye E, Martineau I, Piazzi A, O Dea L, Ince S, Howles C, et al. Clinical assessment of human gonadotrophins produced by recombinant DNA technology. Hum Reprod 1996;11(Suppl 1): Gurgan T, Demirol A. Comparison of two different low-dose step-up regimens of recombinant FSH in polycystic ovary disease patients. Poster presented at the 19th annual ESHRE meeting; Madrid, Spain, June 29 to July 2, Bayram N, van Wely M, van der Veen F. Recombinant FSH versus urinary gonadotrophins or recombinant FSH for ovulation induction in subfertility associated with polycystic ovary syndrome. Cochrane Database Syst Rev 2001;2:CD Fertility and Sterility 1773