FERTILITY AND STERILITYt VOL. 69, NO. 3 (SUPPL. 2), MARCH 1998 Copyright ©1998 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A.
Pharmacokinetics of folliclestimulating hormone: clinical significance Zion Ben-Rafael, M.D.,* Tally Levy, M.D.,* and Joop Schoemaker, M.D.† Golda Meir Medical Center, Petah Tiqva, and Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel, and Free University Hospital, Amsterdam, the Netherlands
Objective: To review studies that examine the pharmacokinetics and pharmacodynamics of endogenous, as well as several exogenous FSH preparations. Design: Related studies were identified through a computerized bibliographic search. Patient(s): Initial pharmacodynamic studies were done in animal models and in women and men with either hypogonadotropic hypogonadism or suppressed hypothalamic-pituitary-gonadal axis. More recent studies evaluated FSH pharmacokinetics during ovulation induction treatment in women with normal ovulatory cycles or polycystic ovarian syndrome. Result(s): Various types of FSH exist according to their sialic acid content. High estrogen levels induce the secretion of less sialylated molecules with higher receptor affinity and an increased clearance rate. It appears that there is a threshold FSH level that should be reached to achieve an ovarian response. A very narrow range exists between the threshold and ceiling level for monofollicular growth. This threshold level is surpassed intentionally during IVF treatment cycles to induce multiple follicular recruitment. The threshold level can change under situations such as polycystic ovaries, perimenopause, oral contraceptives, and GnRH analogue treatment. Conclusion(s): To avoid the risk of ovarian hyperstimulation syndrome and multiple pregnancies, careful adjustments of serum FSH levels should be made by fine dosage modifications. By monitoring FSH levels and using less sialylated preparations, the efficacy of the treatment probably will improve. (Fertil Sterilt 1995;63:689 –700. ©1995 by American Society for Reproductive Medicine.) Key Words: FSH, pharmacokinetic, recombinant FSH, threshold level, PCOS
Received June 10, 1994. Reprint requests: Zion Ben Rafael, M.D., Department of Obstetrics and Gynecology, Golda Meir Medical Center, Petah Tiqva 49372, Israel (FAX: 972-3-9372445). *Department of Obstetrics and Gynecology, Golda Meir Medical Center, and Sackler School of Medicine, Tel Aviv University. †Division of Reproductive Endocrinology and Fertility, Department of Obstetrics and Gynecology, Free University Hospital. 0015-0282/98/$19.00 PII S0015-0282(97)00507-4
Follicle-stimulating hormone is an anterior pituitary glycoprotein that plays a crucial role in folliculogenesis and also is used frequently in the management of infertility. Clinicians are well aware of the relationship between the dose of FSH administered and the biological response in terms of E2 levels and the number of follicles growing in the ovaries. However, unlike other drugs, very little is known about the pharmacodynamics of endogenous and exogenous FSH. Obscurity remains regarding factors affecting the dose of FSH and the duration of treatment necessary to stimulate ovarian response leading to ovulation and pregnancy. The extreme narrowness of the therapeutic range between no effect and stimulation of several follicles, sometimes leading to hyperstimulation and higher rates of multiple pregnancies and abortions, is a major problem (1). Another enigma is the highly variable ovarian
response between different individuals, even to uniform, fixed treatment regimens. This may depend on the influence of endogenous FSH levels and the heterogeneity of FSH substances used to induce ovulation. The basic concept of safer gonadotropin administration that allows reduction in the risks involved in hyperstimulation requires precise understanding of the threshold hypothesis, first postulated by Brown (1) and later supported by others (2– 4). According to this theory, a certain threshold level of FSH should be exceeded before the preantral follicles that reach the FSH-dependent stage progress to maturation (5). This paper reviews the dynamics of FSH secretion, the pharmacokinetics of FSH preparations used clinically, and presents several studies that indicate the importance of evaluating serum FSH levels during treatment cycles.
DRUG PHARMACOKINETICS Safe and effective drug therapy requires delivery to target tissues in concentrations within the narrow range that yields efficacy without toxicity. Optimal precision in achieving concentrations of drugs within this therapeutic ‘‘window’’ can be achieved with regimens that are based on the kinetics of drug availability to target sites. The pharmacokinetics of a specific drug are dependent on the mode of administration, degree of absorption, and volume of distribution. Most drugs are eliminated as a first-order process, that is, the time required for the drug level in the plasma to reach one half the original value (the half-life or t1/2) will be the same regardless of the plasma level at the beginning of the treatment. Usually, a first-order elimination process reaches completion after three to four half-lives. With repeated administration, the drug accumulates in the body if it is given before complete elimination of the previous dose has occurred. Steady state is reached when the rate of administration equals the rate of elimination. If the time required to reach a steady state is too long, plasma levels can be achieved more rapidly by the administration of a loading dose at the beginning of the treatment, followed by smaller maintenance amounts. Elimination is carried out largely by the kidneys and liver in a clearance rate dependent on filtration, secretion, and reabsorption. The extent to which a drug is bound to plasma proteins also determines the fraction of drug extracted by the eliminating organs.
DYNAMICS OF FSH SECRETION The pituitary FSH molecule is a heterodimeric glycoprotein composed of two dissimilar, noncovalently linked aand b-chains (6, 7). The a- and b-subunits are held together by electrostatic and hydrophobic forces that can be separated in vitro by treatment with acidified urea (6). The a-subunit contains 92 amino acids and is common to FSH, LH, thyroid-stimulating hormone, and placental-derived hCG (8, 9). Five disulphide bonds contribute to its tertiary structure. The b-subunit is composed of 115 amino acids sequenced in a hormone specific manner and six disulphide bonds (10, 11). The FSH molecule contains four asparagine-linked carbohydrate side chains attached at position 52 and 78 on the a-subunit (7, 8) and at position 7 and 24 on the b-subunit (10, 11). Up to 20 different FSH molecules with varying degrees of glycosylation have been purified (12), creating a spectrum of isoforms with differences in charge, bioactivity, and elimination half-lives (13, 14). These oligosaccharides have a role in the stabilization of the protein conformation. However, the more acidic, glycosylated molecules have a reduced affinity to the FSH receptor (2, 15). Loss of the sialic acid residue does not alter receptor recognition, suggesting that it is not necessary for receptor binding (16). On FERTILITY & STERILITYt
the other hand, the more glycosylated molecules have a longer half-life. Thus, it appears that desialylated molecules have a significantly decreased in vivo bioactivity (15, 16) probably because of increased plasma clearance. It was found that hepatic binding proteins isolated from mammalian liver specifically bind desialylated glycoproteins and direct them to lysosomal degradation (17). The various types of FSH are secreted according to the physiological requirements at a given time. The amount of sialic acid is influenced mainly by E2 levels (18, 19) and possibly by GnRH (14). During the reproductive years, when serum E2 concentration is high, the FSH molecule is less glycosylated with a shorter half-life but with greater receptor affinity (20, 21). Before puberty and after menopause, the low E2 levels induce the formation and secretion of more glycosylated forms (22, 23). Estrogen (E) treatment during menopause reduces the sialic acid content in the FSH molecule (2). To evaluate the differences in receptor affinity and activity of the different FSH forms, it may be more accurate to measure the FSH bioactivity instead of its immunoreactivity. Jia et al. (24) used E production by rat granulosa cells (GC) as an in vitro bioassay to measure serum concentrations of bioactive FSH compared with immunoreactive FSH. In women with normal cycles, the bioactive FSH levels exhibited a pattern resembling that of the immunoreactive FSH throughout the cycle. The FSH levels were higher during the early follicular phase and lower at the late follicular phase, with a preovulatory surge. The lowest levels were found during the luteal phase. The same group also evaluated FSH levels in women with hypergonadotropic and hypogonadotropic states and concluded that measurement of immunoreactive FSH correctly reflects the biological activity of FSH in the serum. Reddi et al. (25) used a modified rat GC bioassay and found that circulating bioactive FSH levels are maximal 12 to 13 days before the onset of the midcycle LH surge. Follicle-stimulating hormone induces cell differentiation and proliferation in the preovulatory follicle. When the FSH rise occurs at the beginning of the menstrual cycle, there is a cohort of preantral and early antral follicles 2 to 5 mm in diameter, capable of responding to FSH (26). Each of these follicles has its own sensitivity to FSH, expressed by its threshold level. When the FSH plasma concentration surpasses this threshold, the follicle enters its final rapid growth phase. The most sensitive follicle, that is, the first to start the growth phase, will become the leader. This dominant follicle secretes increasing amounts of inhibin and E that suppress pituitary FSH release through negative feedback, thus preventing the entry of new follicles to the rapid growth stage. As the FSH concentration (bioactive and immunoreactive) declines, less mature follicles that are still FSH-dependent cannot acquire enough aromatizing capability to create for the follicle a more estrogenic environment; hence they be41S
come predominantly androgenic and atretic. The dominant follicle, which has already acquired substantial aromatase activity through the increase in GC mass and FSH receptors, can thrive and achieve ovulation despite the FSH decline. These delicate steps demand a short-acting molecule with high-receptor affinity, achieved physiologically by secreting a more basic FSH molecule with a low sialic acid content. Although this phenomenon has been shown in the female rat, conflicting data have been published regarding FSH secretion in humans. Although Wide (21) claimed that there is apparently little difference in the form of FSH secreted at different phases of the cycle, others showed a shift of FSH isoforms from a basic pH range during midcycle to more acidic forms in the mid to late luteal phase (19). It is understandable that a further increase in FSH during the follicular phase occurring during gonadotropin treatment will elevate the FSH level above a ‘‘ceiling for monofollicular growth.’’ The FSH level will then surpass the threshold level of more follicles, and thus multifollicular growth will occur. Ultimately, this may lead to ovarian hyperstimulation. The difference between the threshold of the most sensitive follicle and the ceiling for the monofollicular growth is very small and may be as little as 10% of the administered dose (1).
DYNAMICS OF FSH METABOLISM Another mode of physiological regulation of FSH fluctuations is through the metabolic rate. Follicle-stimulating hormone is cleared from the circulation by the kidneys and liver. In premenopausal women there is a rapid clearance of the FSH molecule, thus achieving a shorter half-life, compared with post-menopausal women in whom the rise in FSH concentration is due to a decrease in clearance as well as to an increased secretion of a more glycosylated form. It is extremely difficult to measure the metabolic rate for FSH because there is a continuous secretion of endogenous hormone that prevents accurate measurements of exogenous FSH pharmacokinetics. The rate of metabolism is influenced also by the method of measurement and, in cases of exogenous administration, by the FSH preparation used. Most commercially available gonadotropin preparations are derived from the urine of postmenopausal women and thus have relatively more sialic acid and a longer half-life than pituitary FSH in pre-menopausal years. Yen et al. (27), by measuring endogenous FSH levels in postmenopausal women immediately after hypophysectomy, found that the disappearance of FSH from the blood has two components: an initial fast component corresponding to the volume of distribution with a t1/2 of 3.9 hours and a second slow component with a longer half-life of 70.4 hours, representing metabolism of the FSH molecule and excretion in the urine. Diczfalusy and Harlin (28) studied the clinical pharmacokinetic and pharmacodynamic properties of three 42S
commercially available gonadotropin preparations. After IV administration of 150 IU/L FSH in the form of Humegon (Organon International, Oss, the Netherlands), Pergonal (Serono Labs, Coinsins, Switzerland), and Metrodin (Serono Labs), the maximal concentrations (Cmax) were 27.5, 24.1, and 26.5 IU/L respectively, achieved after Tmax (time maximum) of 15.4, 16.9, and 16.9 minutes, respectively. The t1/2 of the fast component was 1.6, 2.3, and 2 hours, respectively, whereas the slow component was shorter than that found by Yen et al. (27): 11, 10, and 7.3 hours, respectively. These results demonstrate that the pharmacodynamic properties of the different urinary gonadotropin preparations are very similar when administered IV. The longer ‘‘slow compartment’’ found by Yen et al. (27) might be explained by a change in metabolism after hypophysectomy. The situation appears to be considerably different after IM administration of the same urinary gonadotropin preparations (28). When 150 IU/d FSH were given IM for 8 days, marked individual variations were found regarding peripheral FSH levels, and each patient tested seemed to exhibit characteristics individual FSH profiles irrespective of the preparation administered. The authors described two types of reactions: one with a gradual rise in FSH levels and the other with little, if any, rise in the peripheral FSH level during the treatment days. The first significant rise was observed after the discontinuation of the gonadotropin therapy. The t1/2 of FSH after IM administration exceeded 40 hours, which is approximately fourfold longer than after IV injection, and it took 4 to 5 days for the elevated FSH levels to return to the pretreatment values. These results probably reflect a depot effect at the site of injection. This depot effect and a longer half-life of a urinary gonadotropin preparation that is due to a higher content of sialic acid partially may explain the multiple follicular development, ovarian hyperstimulation, and high rates of multiple pregnancies that occur during induction protocols using urinary gonadotropins.
RECOMBINANT HUMAN FSH For many years, infertility treatment has been based on gonadotropins extracted and purified from the urine of postmenopausal women. The expression of human FSH in Chinese hamster ovary cells transfected with the a and b subunit genes (29) resulted in the synthesis of recombinant human FSH (FSH). The polypeptide backbone of the FSH is identical to that of natural FSH. However, there are various different isoform profiles according to carbohydrate residue composition that closely resemble those seen in natural human FSH (15). The charge heterogenicity and bioactivity of FSH were confirmed by chromatofocusing, receptor displacement, and in vitro and in vivo studies (15, 16, 30). It also appears that modulation of the isoform content of a given FSH preparation either during fermentation or purification can change its bioactivity (15). One of the main Vol. 69, No. 3 (Suppl. 2), March 1998
advantages of FSH is the lack of other hormones such as LH. Indeed, it is the only preparation with a biochemical purity of over 99% (30). All natural FSH preparations were found to be contaminated with small or large amounts of other proteins from pituitary or urinary origin. Mannaerts et al. (30) compared the in vivo and in vitro activity of recombinant FSH with those of pituitary FSH, urinary FSH, and hMG. Specific in vivo bioactivity and immunoreactivity of recombinant FSH were higher than those of urinary FSH and hMG, although bioactivity to immunoreactivity ratios of these preparations were not significantly different. Compared with a highly purified pituitary FSH, the recombinant FSH had a comparable high specific in vivo bioactivity, but because the pituitary FSH had a lower immunoreactivity, it resulted in a significantly increased in vivo bioactivity:immunoreactivity ratio. Receptor displacement and in vitro bioassay studies based on induction of aromatase activity in immature rat Sertoli cells and T production in mouse Leydig cells demonstrated comparable activity of recombinant, urinary, and pituitary FSH (30). Although in vitro and in vivo animal studies showed that recombinant FSH and FSH derived from natural sources are indistinguishable, clinical pharmacokinetics studies in humans are scarce. Le Cotonnec et al. (31) assessed the pharmacokinetics of recombinant FSH and urinary FSH (Gonal-F and Metrodin; Serono Laboratories, Aubonne, Switzerland) after a single IV injection in 12 pituitary down regulated healthy females. They found a two-compartment model in both preparations. The initial (distribution) half-life of FSH was approximately 2 hours, and the true terminal (elimination) half-life of FSH appeared to be slightly under 1 day (17 6 5 hours [mean 6 SD] for urinary FSH and 17 6 4 hours for recombinant FSH). Approximately 20% of the administered urinary derived FSH and 10% of the recombinant FSH dose was extracted into the urine. Thus, moderate renal impairment does not require dosage adjustments of FSH. The renal clearance was found to be slightly greater for urinary FSH (0.1 6 0.05 L/h) than for recombinant FSH (0.07 6 0.04 L/h). These values that are less than the glomerular filtration rate may indicate that human FSH is reabsorbed after filtration, or that high molecular weight glycosylated human FSH (approximately 31 kd) is too large to be excreted freely, or that a renal metabolism of the molecule may exist. With both preparations, the initial volume of distribution was found to be 4 L, which corresponds well to serum volume. The volume of distribution at steady state was approximately 11 L, which approaches the volume of extracellular water. Mannaerts et al. (32) assessed the pharmacokinetics and pharmacodynamic properties of recombinant FSH after a single IM injection of 300 IU recombinant FSH (Org 32489, CP 90073; Organon International) in eight women and seven men suffering from gonadotropin deficiency. Serum FSH levels were raised at 30 minutes and returned to baseline values 264 hours after the injection. The mean t1/2 was not significantly different between the sexes, FERTILITY & STERILITYt
being 44 6 14 hours in females and 32 6 12 hours in males. These values are in agreement with those reported for urinary FSH and hMG (28, 33). The mean Cmax values were significantly lower in females than in male volunteers (4.3 6 1.7 versus 7.4 6 2.8 IU/L), and mean Tmax was also significantly longer in females than in males (27 6 5 versus 14 6 8 hours), suggesting that the absorption of recombinant FSH is slower in women than in men. Indeed, a negative correlation was found between body weight and serum FSH levels. Because gluteal fat is greater in women (34), part of the drug was deposited probably in their SC adipose layers rather than IM, leading to a less rapid absorption. Similar results were obtained in a study comparing single dose pharmacokinetics of 150 IU Gonal-F administered by IV, IM, and SC routes (35). A one-compartment pharmacokinetic model with first-order absorption can describe recombinant FSH pharmacokinetics. After IM and SC administration, there is a longer elimination half-life (37 6 25 hours) compared with IV administration (18 6 6 hours). The mean Cmax was similar (3 6 1 IU/L) after the IM and the SC administration, although recombinant FSH is absorbed apparently faster after SC administration as indicated by a shorter Tmax for the SC route (16 6 10 hours), when compared with the IM route (25 6 10 hours), and a shorter absorption half-life observed with the SC administration (5 6 4 hours) than with the IM route (8 6 4 hours). Recombinant human FSH has a bioassay:immunoassay ratio of 1.6 and urinary FSH a ratio of 1.15 (31). After an IV bolus of 300 IU Gonal-F, the ratio drops by 30% immediately, corresponding to the mixture of exogenous recombinant FSH with endogenous FSH. The ratio declines by 15% after administration of 150 IU of recombinant FSH. Subsequently, the serum FSH bioassay:immunoassay ratio remains unchanged for approximately 2 hours and then progressively increases to maximum on day 2 after 150 IU (31, 35) and day 4 after 300 IU (31). Thereafter the ratio decreases returning to normal by day 7 (31). Another study compared the pharmacodynamic properties of recombinant and urinary FSH in eight gonadotropin deficient subjects who received an IM injection of 300 IU FSH (36). No significant differences were apparent between the bioactive FSH levels (measured by immature rat in vitro GC cell bioassay), although the immunoreactive FSH level at 72 hours was significantly higher after urinary FSH than after recombinant FSH. Nevertheless, there was no significant changes in the bioactive:immunoreactive ratios of FSH within time and between sexes after recombinant or urinary FSH. An interesting finding was that in all subjects the bioactive:immunoreactive ratio was significantly higher after recombinant FSH than after urinary derived FSH treatment, indicating that recombinant FSH probably contains relatively more basic isoforms (35, 36). Shoham et al. (37) recently presented data regarding dose-related increases of serum FSH after multiple admin43S
istration of recombinant FSH in two hypogonadotropic women. The women received IM injections of 75 IU/d for 7 days, followed by 150 IU/d for another 7 days and 225 IU/d for a further 7 days. For each recombinant FSH dose applied, steady state levels of serum FSH concentrations were reached in 3 to 5 days. Plateau serum FSH concentrations were reached at day 17 after a total administration of 2,250 IU recombinant FSH. Serum FSH concentrations decreased to baseline levels after 10 to 12 days. After SC administration of 150 IU/d recombinant FSH for 7 days (35), a steady state was reached within 4 days. This indicates that on the basis of pharmacokinetic data, physicians should wait at least 4 days to assess the efficacy of a given dose and not modify it too frequently. Nevertheless, there is a large interpatient variability in the ovarian response to recombinant FSH that mainly arises from individual diversities in ovarian sensitivity to FSH, rather than from differences in pharmacokinetics (3, 38). In accordance with the two-cell theory, recombinant FSH can induce follicular growth, although without concomitant increase in E2 secretion (37, 39). Similar data have been published also for urinary FSH (40). Interestingly, inhibin, which rose earlier than E2, was found to be an excellent early indicator of follicular development (38). It is important to note that recombinant FSH was well tolerated, no drug-related adverse reactions were noted, and no serum antirecombinant FSH antibodies were detected (32, 37, 39). It seems that in the future recombinant gonadotropins will be the treatment of choice for ovulation induction because these preparations are devoid of inactive contaminants, thus allowing SC administration to have a higher bioactivity and improved batch-to-batch consistency.
THE THRESHOLD THEORY The stimulation program most often used during ovulation induction cycles is the so-called ‘‘step-up’’ induction protocol, that is, induction is started with a certain dose of gonadotropins that is increased if after approximately 5 to 7 days no sign of ovarian response has been observed. Once the daily effective dose has been achieved, the same dose is continued until the development of a preovulatory follicle. The effect of FSH at the ovarian level is dependent on plasma concentration. This, in turn, is influenced not just by the dose administered but also by endogenous FSH secretion, metabolic clearance rate, and volume of distribution, which are individual and differ from woman to woman. Since Brown’s study in 1978 (1), it has been known that the ovary probably responds to a certain FSH threshold level. This is seen in spontaneous ovarian cycles as well as during gonadotropin therapy. By stepwise increments of 10% to 30% in the dose of the gonadotropin, Brown was able to find the threshold level for the development of a single follicle. 44S
Several other studies also demonstrated the existence of such a threshold (41). It appears, however, that the precise dose is individual, depends on endogenous FSH metabolism, and most probably on individual and different ovarian sensitivity. Zeleznik and Kubik (42) investigated the relationship between plasma gonadotropin concentration and the initiation and maintenance of preovulatory follicular growth in Macaque monkeys. The monkeys, after receiving GnRH analogue (GnRH-a), were given pulsatile infusion of human FSH. The authors showed that by elevating plasma FSH concentrations to a threshold level of 15 to 20 mIU/mL (conversion factor to SI unit, 1.00), E2 concentration began to rise above baseline values. Subsequently, the duration at which FSH levels were kept above threshold influenced the number of follicles that were stimulated. Interestingly, this increased ovarian activity occurred in the presence of a constant plasma concentration of FSH. A second important conclusion was that once the FSH infusion was reduced, despite a progressive decline in plasma FSH levels, E2 concentrations continued to rise in a manner similar to that seen during the spontaneous follicular phase, that is, the sensitivity of the leading follicle for FSH increases and its threshold level decreases. These results, therefore, clearly support Brown’s (1) threshold hypothesis. The first to address the nature of the individual response to gonadotropin treatment were Ben-Rafael et al. (3). They noticed that women of similar ages and weights with normal menstrual cycles had highly variable ovarian responses to a uniform treatment with hMG during an IVF-ET program. To evaluate the differences, they arbitrarily categorized the patients as low responders (peak E2 level , 850 pg/mL [conversion factor to SI unit, 3.671]) or high responders (peak E2 level . 1,500 pg/mL) while treated with a fixed regimen of 3 hMG ampules (225 IU/d). Two strikingly different profiles of serum FSH were observed. In the low-responder group, FSH increased to 20 mIU/mL within 3 days and then remained constant despite continued injections of hMG. In contrast, in the high-responder group, the levels continued to rise to 35 mIU/mL with subsequent higher E2 levels. Thus, the individual variation in response to exogenous FSH seemed to be related to the extent of FSH accumulation in the serum. The same group (43) investigated whether different patterns of endocrine response could be modified in high and low responders by changing the dose of hMG in the same women. It was demonstrated in high responders that treatment with three hMG ampules per day was associated with a steady rise and doubling of FSH blood levels. When these women received two ampules, significantly lower blood levels of FSH were attained. In contrast, when the low responders were treated with either two or three hMG ampules per day, the serum FSH levels did not differ significantly: the levels of FSH increased slightly during the 1st day and then plateaued for the remainder of the follicular phase. Interestingly, despite changes in the amount of gonaVol. 69, No. 3 (Suppl. 2), March 1998
dotropin administered, the hormone patterns were repetitive in the high and low responders, and the number of large follicles remained the same. The low responders had higher pretreatment FSH levels, suggesting impaired follicular development and less responsive ovaries. When basal FSH is elevated, the response to stimulation is reduced. If the threshold theory is applied, then the threshold level in these women may have changed, necessitating a larger dose of hMG to achieve an ovarian response. Forman et al. (44) also presented data supporting this theory. They examined two groups of patients: one group received a fixed schedule of ovulation induction with clomiphene citrate (CC) and hMG in a standard dose of two ampules on alternate days from the 2nd to the 10th day of the cycle; the second group received the same treatment protocol with the addition of FSH for the first 2 days of the follicular phase. By calculating the mean FSH:LH ratios, the authors showed that patients with a low pre-stimulation FSH:LH ratio could benefit from the addition of FSH to the stimulation protocol, whereas this was ineffective in patients with a normal or high FSH:LH ratio. This, in their view, supports the threshold hypothesis and the suggestion that once the threshold level is attained, further increase in the gonadotropin ratio is unlikely to be associated with an improved response. On the contrary, high-responder patients may benefit from lowering the FSH dose. This lower dose may provide a better follicular environment with less luteinization and perhaps a better uterine receptivity, without compromising the number of large follicles (43). Furthermore, the addition of urinary FSH to hMG treatment (45) did not prove to be superior to hMG alone. The initial higher FSH levels did not change the individual ovarian response, supporting the previous findings that follicular recruitment does not appear to depend entirely on the dose of gonadotropins. The metabolism of FSH and the final plasma concentration may be the significant determinants. It is thus suggested that a high FSH dose should be used only in situations were a low dose has failed. Differences in serum FSH levels are also probably the basis for ovarian hyporesponsiveness that is encountered in patients suppressed with GnRH-a (46) or oral contraceptives (OC) (47). It has been shown that the follicular phase is prolonged occasionally after discontinuing OC, and a brief treatment-free interval is needed to re-establish a normal ovarian-pituitary relationship in the subsequent spontaneous cycle (48). Treatment of IVF patients with OC for a short duration caused poor response in certain women associated with profound lowering of gonadotropin secretion during and immediately after administration of the combined pill (47). On the contrary, in normal responders FSH levels on day 3 of the cycle returned to levels that were similar to those of women who were not suppressed by OC (47). It appears that women with suppressed pituitary function may need to return to a certain threshold level of FSH for early follicular FERTILITY & STERILITYt
recruitment if a desirable ovarian response is to be achieved during hMG stimulation. This can be accomplished usually by administering a larger initial daily FSH dose. Notwithstanding, FSH pharmacodynamics are altered greatly under GnRH-a. Scheele et al. (49) studied FSH pharmacodynamics in two groups of patients: with and without GnRH-a treatment. It appeared that in the GnRH-a group, the increased FSH plasma concentration per unit of FSH administered was twice as high as that of the group without the analogue. This can be explained partly by the absence of endogenous negative feedback of follicular secretions on analogue-treated pituitary FSH (Van der Meer M, Schoemaker J, unpublished observation). The changes were so prominent that an effect of GnRH-a on the metabolic clearance rate of FSH could not be excluded. According to all the data presented, the FSH concentration during the early follicular phase is of crucial importance. The rise of FSH levels above a specific, individual threshold value will initiate the entry of follicles into the rapid growth phase. The duration and extent to which the FSH concentration is above the threshold level will determine the number of follicles that will reach the final maturation stage. If ovulation is to be restricted to a single follicle, it is essential that the FSH levels fall rapidly below the threshold. The long half-life of the urinary gonadotropin preparations complicate these fine adjustments. Even if the amount of FSH is reduced, several follicles have already been recruited and will continue to grow despite the decline in FSH concentration. On the other hand, the long half-life of these preparations and a high initial FSH dose are advantageous if the final goal is multiple follicular maturation. This is essential for IVF because pregnancy rates (PRs) are related to the number of good-quality oocytes recovered and embryos transferred to the uterus.
THE APPLICATION OF THE FSH THRESHOLD LEVEL THEORY IN PATIENTS WITH POLYCYSTIC OVARIAN SYNDROME (PCOS) Successful induction of ovulation in patients with PCOS remains a dilemma. In these patients, the cyclic pattern of FSH and LH is typically non-existent, and the chronic subthreshold levels of FSH may allow some follicular growth but are insufficient to facilitate maturation and ovulation. Clomiphene citrate is used to induce ovulation in these women. However, a significant number of patients with PCOS fail to respond to CC therapy. In such patients, treatment with a gonadotropin is frequently successful. Despite close monitoring, multiple pregnancies occur in one of three treatment (50, 51) cycles and ovarian hyperstimulation may occur in up to 60% of cycles (52). When human gonadotropins are used for induction of ovulation, it is accepted that some of the spontaneous ovu45S
latory cycle features cannot be reproduced. The use of pharmacological gonadotropin doses results in supraphysiological levels of FSH, provoking initial development of a large cohort of follicles. During the course of therapy, recruitment of additional follicles does not cease unless FSH levels drop below the threshold. In patients with PCOS, this phenomenon is intensified because of the inherited tendency toward multifollicular development caused by the continuous subthreshold FSH levels and also because of intrinsic ovarian growth factors that may have an important role in the pathogenesis of PCOS (53, 54). In these patients, even a slight increase in the serum FSH concentration is sometimes sufficient to overcome chronic anovulation.
Shoham et al. (60) also compared a conventional protocol (starting dose of 75 IU urinary FSH for 7 days with dose increments of 75 IU every 7 days) with a low-dose protocol with a starting dose of 75 IU/d for 7 days, and if no ovarian response was achieved, the initial dose was increased by 37.5 IU/d every 7 days. If multiple follicles developed, the dose was decreased to 37.5 IU/d in the next cycle. The low-dose protocol resulted in a significant reduction in the number of leading follicles (1.5 6 0.67 compared with 4.4 6 2.1), decreased serum E2 concentrations (1,539 pmol/L compared with 4,050 pmol/L), and higher rates of ovulation. As a result, five patients conceived after low-dose treatment, compared with none in the conventional protocol group.
LOW-DOSE GONADOTROPIN TREATMENT
The authors concluded that during low-dose treatment, the individual FSH threshold level was reached with the smallest effective gonadotropin dose. This threshold dose was then maintained to achieve the lowest number of mature preovulatory follicles, thus avoiding hyperstimulation and multiple gestations.
Many treatment regimens have been applied to avoid overstimulation complications. Polson et al. (55) suggested that monofollicular development could be induced by the low-dose gonadotropin protocol. This was confirmed by others (56, 57). The particular features of this regimen are the prolonged administration of 75 mIU/d (1 ampule) gonadotropin for 7 to 14 days as a starting dose and small (37.5 mIU) dose increments if there is no follicular response. Kamrava et al. (56) showed that the administration of 40 IU FSH one time per day for 4 weeks to two patients with PCOS resulted in successful pregnancies. Seibel et al. (57) also demonstrated that chronic low-dose FSH is capable of reversing the hormonal imbalance and can induce ovulation in PCOS, even without the administration of hCG. Furthermore, only a single preovulatory follicle developed in most cases. Polson et al. (58), by using low-dose pulsatile SC administration of FSH (75 IU/d for 14 days, later increased by 37.5 IU/d at weekly intervals), were able to induce ovulation in 70% treatment cycles in patients with PCOS. Of these ovulatory cycles, 70% were monofollicular, 50% of the treated women became pregnant, and most (4 of 5) were singleton gestations. Because it was not clear whether the success of treatment was related to the pulsatile mode of administration or low dose of FSH used, the same group (58) compared low-dose pulsatile treatment with the efficacy of a single IM/d injection of the same dose for the same duration. There were no significant differences between pulsatile and IM administration, suggesting that the favorable results were related to the low-dose FSH. Sagle and co-workers (59), in a comparative randomized study, compared the effects of urinary FSH and hMG using a low-dose regimen in 30 women with PCOS. The results of the study confirmed the efficacy of low-dose FSH in ovulation induction. Low doses of urinary FSH and hMG were equally successful in inducing ovulation, suggesting that the success of treatment depended on the low gonadotropin dose used rather than the presence or absence of LH in preparation. 46S
Schoemaker et al. (4) approached the threshold level in patients with PCOS with a different treatment concept. They induced monofollicular growth with FSH levels that marginally exceeded the threshold. Fourteen patients received IV administration of urinary FSH with a maximum of three treatment cycles. In the first cycle, the serum FSH level was kept at 1 IU/L above the prestimulation value for the first 10 days. The dose was then increased to achieve a higher serum level by increments of 1 IU/L at weekly intervals until follicular growth was observed by ultrasound (US). Thus the individual FSH threshold level was determined. In subsequent cycles, stimulation was started at a level of 1 IU/L below the threshold level of the preceding cycle. This approach resulted in 76% ovulatory cycles, of which 84% were monofollicular. In subsequent treatment cycles, the FSH level was manipulated just above the threshold, and preliminary data from the same investigators showed that only minor changes in the FSH daily dose (19 IU/d) may have determined whether monofollicular or multifollicular growth would occur. From these encouraging results, it appears that the best treatment approach for patients with PCOS is by way of the FSH threshold concept. However, authors who advocate a slow rise in the FSH dose to reach the threshold level experience a prolonged follicular phase. This phenomenon may represent a follicular phase of several cohorts of follicles, most of which are insensitive to the low FSH levels. Nevertheless, it is possible to overcome this effect by defining the individual FSH threshold levels in patients with PCOS during a low-dose treatment cycle and establishing the gonadotropin dosage needed to initiate ovarian response (Ben-Rafael Z, Levy T, unpublished data). In subsequent cycles, it was possible to achieve the threshold level in a shorter period by slightly elevating the initial gonadotropin dose (by 37.5 IU) and then lowering it when ovarian reVol. 69, No. 3 (Suppl. 2), March 1998
sponse was noticed. This method shortened the follicular phase and at the same time avoided the recruitment and maturation of many follicles. On application of this method, it is most important to establish the daily FSH level to timeously make suitable adjustments. It is also suggested that the effective daily dose for patients with PCOS is highly variable because some patients respond with ovarian hyperstimulation to 1 hMG ampule (75 IU) per day, whereas others require 3– 4 ampules per day to initiate an ovarian response. Van Weissenbruch (61) quantitatively approached the FSH threshold in induction of ovulation protocols in three groups of patients. Group 1 were patients with severe hypogonadotrophic hypogonadism; group 2 were patients with regular menstrual cycles whose pituitary gonadotropin secretion had been suppressed by continuous GnRH infusion, and group 3 patients were diagnosed as having PCOS. During induction of ovulation with IV administration of gonadotropin in a regular step-up protocol, Van Weissenbruch (61) monitored FSH levels and follicular growth by daily US. Once the leading follicle reached a 12-mm diameter and on the assumption that there was a linear growth of 2 mm/d, they were able to extrapolate the FSH levels at the time of follicular selection for dominance (4 to 5 mm in diameter). It appeared that in group 1 (hypogonadotropic hypogonadal patients), it was necessary for the FSH level to be .7.8 IU/L to induce follicular growth. No follicular growth occurred below this level. Similar results were found in group 2 (normally cycling women under pituitary suppression). Whenever follicular growth occurred, the growth initiating FSH level was .7.8 IU/L threshold. The situation differed entirely in the patients with PCOS. Between the levels of 6.3 and 9.8 IU/L follicular growth either took place or did not. The threshold appeared to vary from patient-to-patient, and even to a certain, albeit lesser extent, from cycle-to-cycle in the same patient. This phenomenon may be explained by the varying sensitivity or threshold level in patients with PCOS. Another explanation may be the extreme variations in body weight seen in these women. Chong et al. (62) demonstrated that patients who have normal 610% ideal body weight (IBW) are more likely to respond to lower doses of hMG than patients who have .10% IBW, and in particular those who are .25% of their IBW. Why heavier women may need more hormone to induce ovulation is unknown. It may be related to the greater amount of body surface, inadequate E2 metabolism, and decreased sex hormone-binding globulin. All these are found in patients with PCOS with elevated body mass index (BMI). Also, the IM absorption of the drug is slower and incomplete in obese patients because of increased SC fat or fat infiltration of the muscle. This latter explanation concerning the depot effect may be in accordance with Van Weissenbruch’s data (61) that in patients FERTILITY & STERILITYt
treated by IV gonadotropin, there was no correlation whatsoever between BMI and the FSH threshold level. To simulate the physiological follicular phase, it may be necessary not only to reduce the amount of FSH administered but also to change the molecular form. This approach is proposed by Baird (2) who believes that to mimic the changes in FSH pharmacodynamics that occur during the normal follicular phase, it may be necessary to change the sialic acid content of the exogenous FSH administered. He advised the use of different gonadotropin preparations as the basis of a physiological regimen. Follicle-stimulating hormone with a relatively long half-life could be given in the early follicular phase of the cycle to activate the small follicles. This preparation would then be replaced by one with a shorter half-life to facilitate the adjustment of the doses required to simulate the fall in FSH concentration during the mid and late follicular phase. By this ‘‘physiological’’ induction protocol, a single follicle will hopefully ovulate, and the hazards of hyperstimulation and multiple pregnancies will be avoided.
CONCLUSIONS Gonadotropins have an important role in induction of ovulation. Since the introduction of hMG for clinical use, many approaches have been directed at maximizing their efficacy in increasing PRs while controlling the hyperstimulation effects. To achieve this end, the treatment regimen should simulate a normal menstrual cycle. This aim demands a profound understanding of the pharmacodynamics of FSH and its influence on the ovary. It is now obvious that the most important controlling factor is the serum FSH concentration. At the beginning of the follicular phase, surpassing a ‘‘threshold’’ FSH level will induce the development of a cohort of follicles. A very narrow range exists between such a threshold and ‘‘ceiling’’ level for monofollicular growth that, if surpassed, bears the risk of hyperstimulation. Thus, the best treatment regimen is to reach this threshold level by the administration of low-dose gonadotropins and preferably in the future gonadotropins with a low sialic acid content. Once an ovarian response has been achieved, the lowest effective dose should be continued until ovulation. In contrast, in IVF cycles the basic concept is to induce multiple follicular development. This is achieved by the administration of a high initial FSH dose to intentionally surpass the threshold level. The threshold FSH level is probably constant in normally cycling women. However, it may change under different situations such as PCOS, OC, pre-treatment with a GnRH-a, and perimenopause. It is concluded that the daily serum FSH concentration may be the most important single factor during gonadotropin therapy, and adjustments of serum FSH levels can be made by dosage modification. Measurement of FSH levels has increased our 47S
understanding of the processes involved in follicle stimulation. In the future, monitoring FSH levels along with E2 and US may help to control ovarian stimulation, particularly in low- and high-responder patients. References 1. Brown JB. Pituitary control of ovarian function-concepts derived from gonadotropin therapy. Aust NZJ Obstet Gynaecol 1978;18:47–54. 2. Baird DT. A model for follicular selection and ovulation: lessons from superovulation. J Steroid Biochem 1987;27:15–23. 3. Ben-Rafael Z, Strauss JF III, Mastroianni L Jr, Flickinger GL. Differences in ovarian stimulation in human menopausal gonadotropin treated women may be related to follicle-stimulating hormone accumulation. Fertil Steril 1986;46:586 –92. 4. Schoemaker J, Van Weissenbruch MM, Van der Meer M. New approaches with the FSH threshold principle in polycystic ovarian syndrome. Ann NY Acad Sci 1993;687:296 –300. 5. Fritz MA, Speroff L. The endocrinology of the menstrual cycle: the interaction of folliculogenesis and neuroendocrine mechanisms. Fertil Steril 1982;38:509 –29. 6. Reichert LE Jr, Ward DN. On the isolation and characterization of the alpha and beta subunits of human pituitary follicle-stimulating hormone. Endocrinology 1974;94:655– 64. 7. Rathnam P, Saxena BB. Primary aminoacid sequence of follicle-stimulating hormone from human pituitary glands. J Biol Chem 1975;17: 6735– 42. 8. Fiddes JC, Goodman HM. Isolation, cloning and sequence analysis of the cDNA for the a-subunit of human chorionic gonadotropin. Nature 1979;281:351– 6. 9. Pierce JG, Parsons TF. Glycoprotein hormones: structure and function. Annu Rev Biochem 1981;50:465–95. 10. Shome B, Parlow AF. Human follicle stimulating hormone: first proposal for the amino acid sequence of the hormone-specific b-subunit (hFSHb). J Clin Endocrinol Metab 1974;39:203–5. 11. Watkins PC, Eddy R, Beck AK, Velucci V, Leverone B, Tanzi RE, et al. Sequence and regional assignment of the human follicle stimulating hormone b-subunit gene to the short arm of human chromosome 11. DNA 1987;6:205–12. 12. Stanton PG, Robertson DM, Burgon PG, Schmauk-White B, Hearn MTW. Isolation and physicochemical characterization of human follicle-stimulating human isoforms. Endocrinology 1992;130:2820 –32. 13. Chappel SC, Ulloa-Aguirre A, Coutifaris C. Biosynthesis and secretion of follicle stimulating hormone. Endocr Rev 1983;4:179 –211. 14. Ulloa-Aguirre A, Espinoza R, Damian-Matsumura P, Chappel SC. Immunological and biological potencies of different molecular species of gonadotropin. Hum Reprod 1988;3:491–501. 15. Cerpa-Poljak A, Bishop LA, Hort YJ, Chin CKH, DeKroon R, Mahler SM, et al. Isoelectric charge of recombinant human follicle-stimulating hormone isoforms determines receptor affinity and in vitro bioactivity. Endocrinology 1993;132:351– 6. 16. Galway AB, Hsueh AJW, Keene JL, Yamoto M, Fauser BCJM, Boime I. In vitro and in vivo bioactivity of recombinant human folliclestimulating hormone and partially deglycosylated variants secreted by transfected eukaryotic cell lines. Endocrinology 1990;127:93–100. 17. Kawasaki T, Ashwell G. Chemical and physical properties of a hepatic membrane protein that specifically binds asialoglycoproteins. J Biol Chem 1976;251:1296 –1302. 18. Chappel SC, Bethea CL, Spies HG. Existence of multiple forms of follicle-stimulating hormone within the anterior pituitaries of cyanomolgue monkeys. Endocrinology 1984;115:452– 61. 19. Padmanabhan V, Lang LL, Sonstein J, Kelch RP, Beitins IZ. Modulation of serum follicle-stimulating hormone bioactivity and isoform distribution by estrogenic steroids in normal women and in gonadal dysgenesis. J Clin Endocrinol Metab 1988;67:465–73. 20. Galle P, Ulloa-Aguirre A, Chappel SC. Effects of estradiol phenobarbitone and LHRH upon the isoelectric focusing profile of pituitary follicle-stimulating hormone in ovariectomized hamster. J Endocrinol 1983;99:31– 40. 21. Wide L. Male and female forms of human follicle-stimulating hormone is serum. J Clin Endocrinol Metab 1982;55:682– 8. 22. Ulloa-Aguirre A, Chappel SC. Multiple species of follicle-stimulating hormone exist within the anterior pituitary gland of male golden hamsters. J Endocrinol 1982;95:257– 63. 23. Chappel SC, Ulloa-Aguirre A, Ramaley JA. Sexual maturation in female rats: time course of the appearance of multiple species of anterior pituitary FSH. Biol Reprod 1982;28:196 –207. 24. Jia XC, Kessel B, Yen SSC, Tucker EM, Hsueh AJW. Serum bioactive follicle-stimulating hormone during the human menstrual cycle and in hyper and hypogonadotrophic states: application of a sensitive granulosa cell aromatase bioassay. J Clin Endocrinol Metab 1986;62:1243–9.
25. Reddi K, Wickings EJ, McNeilly AS, Baird DT, Hillier SG. Circulating bioactive follicle stimulating hormone and immunoreactive inhibin levels during the normal human menstrual cycle. Clin Endocrinol (Oxf) 1990;33:547–57. 26. Gougeon A. Dynamics of follicular growth in the human: a model from preliminary results. Hum Reprod 1986;1:81–7. 27. Yen SSC, Llerena LA, Pearson OH, Littell AS. Disappearance rates of endogenous follicle-stimulating hormone in serum following surgical hypophysectomy in man. J Clin Endocrinol Metab 1970;30:325–9. 28. Diczfalusy E, Harlin J. Clinical-pharmacological studies on human menopausal gonadotropin. Hum Reprod 1988;3:21–7. 29. Keene JL, Matzuk MM, Otani T, Fauser BC, Galway AB, Hsueh AJ, et al. Expression of biologically active human follitropin in Chinese hamster ovary cells. J Biol Chem 1989;246:4769 –75. 30. Mannaerts B, De Leeuw R, Geelen J, Van Ravestein A, Van Wezenbeek P, Schuurs A, et al. Comparative in vitro and in vivo studies on the biological characteristics of recombinant human follicle-stimulating hormone. Endocrinology 1991;129:2623–30. 31. le Cotonnec J-Y, Porchet HC, Beltrami V, Khan A, Toon S, Rowland M. Clinical pharmacology of recombinant human follicle-stimulating hormone (FSH). I. Comparative pharmacokinetics with urinary human FSH. Fertil Steril 1994;61:669 –78. 32. Mannaerts B, Shoham Z, Schoot D, Bouchard P, Harlin J, Fauser B, et al. Single-dose pharmacokinetics and pharmacodynamics of recombinant human follicle-stimulating hormone (Org 32489) in gonadotropindeficient volunteers. Fertil Steril 1993;59:108 –14. 33. Jockenhivel F, Fingscheidt U, Khan SA, Behre HM, Nieschlag E. Bio and immuno-activity of FSH in serum after intramuscular injection of highly purified urinary human FSH in normal men. Clin Endocrinol (Oxf) 1990;33:573– 84. 34. Zuidema S, Pieters FA, Duchateau GS. Release and absorption rate aspects of intramuscular injected pharmaceuticals. Int J Pharm 1988; 47:1–12. 35. le Cotonnec J-Y, Porchet HC, Beltrami V, Khan A, Toon S, Rowland M. Clinical pharmacology of recombinant human follicle-stimulating hormone. II. Single doses and steady state pharmacokinetics. Fertil Steril 1994;61:679 – 86. 36. Matikainen T, De Leeuw R, Mannaerts B, Huhtaniemi I. Circulating bioactive and immunoreactive recombinant human follicle-stimulating hormone (Org 32489) after administration to gonadotropin deficient subjects. Fertil Steril 1994;61:62–9. 37. Shoham Z, Mannaerts B, Insler V, Coelingh-Bennink H. Induction of follicular growth using recombinant human follicle-stimulating hormone in two volunteer women with hypogonadotropic hypogonadism. Fertil Steril 1993;59:738 – 42. 38. Porchet HC, le Cotonnec J-Y, Loumaye E. Clinical pharmacology of recombinant human follicle-stimulating hormone. III. Pharmacokineticpharmacodynamic modeling after repeated subcutaneous administration. Fertil Steril 1994;61:687–95. 39. Schoot D, Coelingh-Bennink H, Mannaerts B, Lamberts S, Bouchard P, Fauser B. Human recombinant follicle-stimulating hormone induces growth of preovulatory follicles without concomitant increase in androgen and estrogen biosynthesis in a woman with isolated gonadotropin deficiency. J Clin Endocrinol Metab 1992;74:1471–3. 40. Shoham Z, Balen A, Patel A, Jacobs HS. Results of ovulation induction using human menopausal gonadotropin or purified follicle-stimulating hormone in hypogonadotropic hypogonadism patients. Fertil Steril 1991;56:1048 –53. 41. Van Weissenbruch MM, Schoemaker HC, Drexhage HA, Schoemaker J. Pharmaco-dynamics of human menopausal gonadotropin (hMG) and follicle-stimulating hormone (FSH). The importance of the FSH concentration in initiating follicular growth in polycystic ovary-like disease. Hum Reprod 1993;8:813–21. 42. Zeleznik AJ, Kubik CJ. Ovarian responses in Macaques to pulsatile infusion of follicle-stimulating hormone (FSH) and luteinizing hormone: increased sensitivity of the maturing follicle to FSH. Endocrinology 1986;119:2025–32. 43. Benadiva CA, Ben-Rafael Z, Strauss JF III, Mastroianni L Jr, Flickinger GL. Ovarian response of individuals to different doses of human menopausal gonadotropin. Fertil Steril 1988;49:997–1001. 44. Forman RG, Demouzon J, Feinstein MC, Testart J, Frydman R. Studies on the influence of gonadotropin levels in the early follicular phase on the ovarian response to stimulation. Hum Reprod 1991;6:1137– 43. 45. Benadiva CA, Ben-Rafael Z, Blasco L, Tureck R, Mastroianni L Jr. Flickinger GL. An increased initial follicle-stimulating hormone/luteinizing hormone ratio does not affect ovarian responses and the outcomes of in vitro fertilization. Fertil Steril 1988;50:777– 81. 46. Ben-Rafael Z, Lipitz S, Bider D, Mashiach S. Ovarian hyporesponsiveness in combined gonadotropin-releasing hormone agonist and menotropin therapy is associated with low serum follicle-stimulating hormone levels. Fertil Steril 1991;55:272–5. 47. Benadiva CA, Ben-Rafael Z, Blasco L, Tureck R, Mastroianna L Jr.
Vol. 69, No. 3 (Suppl. 2), March 1998
50. 51. 52. 53. 54.
Flickinger GL. Ovarian response to human menopausal gonadotropin following suppression with oral contraceptives. Fertil Steril 1988;50: 516 – 8. Riklein TA, Mishell DR. Gonadotropin, prolactin and steroid hormone levels after discontinuation of oral contraceptives. Am J Obstet Gynecol 1977;127:585–9. Scheele F, Hompes PGA, van der Meer M, Schoute E, Schoemaker J. The relationship between follicle-stimulating hormone dose and level and its relevance for ovulation induction with adjuvant gonadotropinreleasing hormone-agonist treatment. Fertil Steril 1993;60:620 –5. Kemmann E, Tavalcoli F, Shelden RM, Jones JR. Induction of ovulation with menotropins in women with polycystic ovarian disease. Am J Obstet Gynecol 1981;141:58 – 62. Wang CF, Gemzell C. The use of human gonadotropins for the induction of ovulation in women with polycystic ovarian disease. Fertil Steril 1980;33:479 – 86. Raj SG, Berger MG, Grimes EM, Taymor ML. The use of gonadotropins for the induction of ovulation in women with polycystic ovarian disease. Fertil Steril 1977;28:1280 – 4. Tonetta SA, DiZerega GS. Intragonadal regulation of follicular maturation. Endocr Rev 1989;10:205–29. Homburg R, Pariente C, Lunenfeld B, Jacobs HS. The role of insulin-like growth factor I (IGF-I) and IGF binding protein I (IGFBP-I) in the pathogenesis of polycystic ovary syndrome. Hum Reprod 1992;7:1379 – 83. Polson DW, Mason HD, Saldahna MBY, Franks S. Ovulation of a single dominant follicle during treatment with low dose pulsatile folli-
FERTILITY & STERILITYt
60. 61. 62.
cle-stimulating hormone in women with polycystic ovary syndrome. Clin Endocrinol (Oxf) 1987;26:205–12. Kamrava MM, Seibel MM, Berger MJ, Thompson I, Taymor ML. Reversal of persistent anovulation in polycystic ovarian disease by administration of chronic low dose follicle-stimulating hormone. Fertil Steril 1982;37:520 –3. Seibel MM, Kamrava MM, McArdle C, Taymor ML. Treatment of polycystic ovary disease with chronic low dose follicle-stimulating hormone: biochemical changes and ultrasound correlation. Int J Fertil 1984;29:39 – 42. Polson DW, Mason HD, Kiddy DS, Winston RML, Margara R, Franks S. Low dose follicle-stimulating hormone in the treatment of polycystic ovary syndrome: a comparison of pulsatile subcutaneous with daily intramuscular therapy. Br J Obstet Gynaecol 1989;96:746 – 8. Sagle MA, Hamilton-Fairley D, Kiddy DS, Franks S. A comparative randomized study of low dose human menopausal gonadotropin and follicle-stimulating hormone in women with polycystic ovarian syndrome. Fertil Steril 1991;55:56 – 60. Shoham Z, Patel A, Jacobs HS. Polycystic ovarian syndrome: safety and effectiveness of stepwise and low dose administration of purified follicle-stimulating hormone. Fertil Steril 1991;55:1051– 6. Van Weissenbruch MM. Gonadotropins for induction of ovulation, immunological, pharmacological and clinical studies [dissertation]. Amsterdam: Free University, 1990. Chong AP, Rafael RW, Forte CC. Influence of weight in the induction of ovulation with human menopausal gonadotropin and human chorionic gonadotropin. Fertil Steril 1986;46:599 – 603.