Differences in ovarian stimulation in human menopausal gonadotropin treated woman may be related to follicle-stimulating hormone accumulation*

Transcription

1 FRTILITY AND STRILITY Copyright The American Fertility Society Vol. 46, No.4, October 1986 Printed in U.8A. Differences in ovarian stimulation in human menopausal gonadotropin treated woman may be related to follicle-stimulating hormone accumulation* Zion Ben-Rafael, M.D.t:J: Jerome F. Strauss III, M.D., Ph.D. Luigi Mastroianni, Jr., M.D. George L. Flickinger, V.M.D., Ph.D. Division of Reproductive Biology, Department of Obstetrics and Gynecology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania Two groups of normal ovulatory women who displayed either a marked (high responders; HR) or a more subtle (low responders; LR) ovarian response to a fixed dose of human menopausal gonadotropins (hmg) were evaluated for differences in blood levels of hormones. Serum follicle-stimulating hormone (FSH) levels doubled during the first 3 days of treatment (to approximately 20 miulml) in all patients; thereafter, the levels plateaued in LR but continued to rise steadily (to 35 miulml) in HR. In the latter group, rise in estradiol (z) and FSH was accompanied by an increase of luteinizing hormone (LH; two to five times) progesterone (P; four to eight times) testosterone (T; three to four times) and prolactin (PRL; 2 times) toward the end of the follicular phase. Positive correlation was found between FSH and 2 in HR and LR. Positive correlation was found, however, between LH, T, and P and between 2, P, and PRL only in HR. The extent of FSH accumulation in the circulation may be a principal factor in determining an individual's response to hmg therapy. Temporal changes of blood hormones indicated that the continuous rise in FSH levels in HR was associated with early luteinization of the follicles. Increased secretion of P in the follicular phase of these women (HR) probably synergized with the elevated 2 levels to elicit LH release. Similar changes in blood hormones were not found in LR. Fertil SteriI46:586, 1986 Received October 25, 1985; revised and accepted June 9, *Supported in part by United States Public Health Service grant HD , F05TW03580, and Mellon Foundation. tfogarty International (National Institutes of Health) fellow in Reproductive Biology and ndocrinology on leave from the Department of Obstetrics and Gynecology, the Chaim Sheba Medical Center, Tel-Hashomer and Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel. *Reprint requests: Zion Ben-Rafael, M.D., Department of Obstetrics and Gynecology, Division of Reproductive Biology, Hospital of the University of Pennsylvania, School of Medicine, 3400 Spruce Street, Philadelphia, Pennsylvania Despite numerous reports addressing effects of human menopausal gonadotropin (hmg) human chorionic gonadotropin (hcg) therapy,i-5 it is not known whether individual variability of ovarian responses to hmg and its repetition in successive cycles is due to ovarian or pituitary factors.6 Patient variability also has been attributed to differences in serum levels of follicle-stimulating hormone (FSH) and luteinizing hormone (LH) at the initiation of treatment. 7 This present study was done to evaluate differences in several blood hormones during follicular 586 Ben-Rafael et al. FSH accumulation in hmg-treated patients Fertility and Sterility

2 1.j II j <. i!, j' i l' ) phases of normal ovulatory patients who responded with either high or low estradiol (2) secretion to a fixed regimen of hmglhcg. Assessment of temporal relationships between hormones promotes the concept that the extent offsh accumulation in blood may be primarily responsible for individual variations in ovarian and pituitary responses to gonadotropin treatment. PATINTS MATRIALS AND MTHODS Serum from ten women who had treatment with hmg for in vitro fertilization and embryo transfer UVF-T) were selected retrospectively and arbitrarily were divided into two categories; five were low responders, with maximum 2 levels < 850 pg/ml (mean, 539 pg/ml), and five were high responders, with 2 levels> 1500 pg/ml (mean, 2312 pg/ml). The patients were selected consecutively until five were obtained from each category. The patients had normal ovulatory cycles and no history or physical findings of endocrinologic, renal, or hepatic disease and suffered from mechanical causes of infertility. Body weights (52.2 and 54 kg) and ages (30.8 and 31.5 years) were similar in low and high responders. Treatment protocol for follicular recruitment has been described.2 In short, 3 ampules of hmg (225 IU FSH and LH, Pergonal, Serono Laboratories, Inc., Randolph, MA) were administered intramuscularly from day 3 of the menstrual cycle for 5 to 7 days. Follicular maturation was assessed by daily measurements of serum 2 beginning on day 3 of the cycle and by daily ultrasonographic scanning, commencing on day 7. The hmg was discontinued when at least two follicles with mean diameters of> 1.5 cm in three planes were observed. The hcg was administered when the largest follicle reached 1.8 cm or if 2 levels tended to plateau (± 10%). Thus hcg (5000 IU, Profasi, Serono Laboratories, Inc., Randolph, MA) was administered 24 to 48 hours after the last dose of hmg. Blood was collected daily at 9:00 A.M., approximately 16 hours after the patients received hmg, and serum was stored at - 20 C until assayed. HORMONAL ASSAYS Hormones were measured with commercially available radioimmunoassays: FSH, LH (Amersham, Arlington Heights, IL), 2 (Radioassay System Laboratories, Carson City, CA), progesterone (P), testosterone (T), and prolactin (PRL) (Diagnostic Product Corp., Los Angeles, CA). Hormones in each sample were assayed in duplicate within the same assay. The sensitivities of the assays were FSH, 1.2 miu/ml; LH, 2 IU/ml; 2, 10 pg/ml; P, 0.05 ng/ml; T, 0.1 ng/ml; and PRL, 1 ng/ml. Interassay coefficients of variations were 6% for T and 2, 10% for P, 10% and 3.5% for FSH and LH, and 8% for PRL. Serum concentrations of FSH and LH represent contributions by endogenously secreted and exogenously administered hormones. STATISTICAL ANALYSS Results, normalized to the day ofhmg administration, are illustrated as daily means ± standard error of the mean (SM) and as individual values of each patient. Analysis of variance, followed by multiple-range tests for samples of unequal size,s, 9 was used to determine differences in circulating hormone levels between low and high responders. Spearman's correlation coefficients was used to determine relationships between different hormones. RSULTS Daily changes in serum FSH concentrations for low and high responders are shown in Figure 1. Basal concentrations were similar in the two groups (11.4 ± 0.6 and 10.9 ± 0.5). In low responders, the mean level of FSH increased two times to 21.5 ± 4.7 miu/ml during the first 3 days of gonadotropin therapy and then plateaued for the remainder of the follicular phase. In contrast, serum FSH levels rose steadily, to 35 ± 3.7 miui ml on day 6 of treatment in high responders. By day 4 and for the remainder of the regimen, the mean serum FSH levels in high responders were significantly higher than in low responders (P < 0.05 on days -4 and -3 and P < on days -2, -1, and 0). As with FSH, the mean pretreatment levels of 2 did not differ (62 ± 20.7 and 48 ± 9.2 pg/ml in low and high responders, respectively; P > 0.05) between low and high responders. In both groups, estrogen levels rose steadily throughout the follicular phase (Fig. 2); however, the mean peak levels of high responders (2312 ± 436 pg/ml) was Vol. 46, No.4, October 1986 Ben-Rafael et al. FSH accumulation in hmg-treated patients 587

3 40 30 " ::::> 20 CJ) u I o Days before hcg administration Figure 1 Serum FSH levels in low and high responders. The open circles represent the low responders (n = 5), and the closed circles represent the high responders (n = 5). Results (mean ± SM) are normalized to the day ofhcg administration. P < 0.05 on days - 4 and - 3; P < 0.01 on days - 2, -1, and O. approximately four times greater than that oflow responders (539 ± 80 pg/ml; P < 0.001). Positive correlations between FSH and 2 were found in both groups, with a more significant correlation coefficient in high responders (r = 0.929, P < 0.001) than in low responders (r = 0.830, P < 0.01). Individual variations in daily serum LH concentrations in high and low responders are shown in Figure 3. LH remained constant during gonadotropin administration, with one exception in the low responders. In contrast, increases (two to five times) in LH were observed toward the end of the follicular phase in all high responders. With one exception (52 miu/mi), peak LH levels in this group were less than normal preovulatory surges. LH levels were closely correlated with P (r = 0.814, P < 0.01) and T (r = 0.784, P < 0.02) levels in high responders but not in low responders. Temporal changes in serum P levels were similar to variations in LH concentrations (Fig. 4). In low responders, P levels remained constant throughout the follicular phase. In contrast, P concentrations in high responders began to rise 2 to 3 days before hcg administration and reached levels six times greater than basal levels on the day ofhcg treatment (0.26 ± 0.1 versus 1.71 ± 0.81 ng/ml, respectively; P < 0.01). Mean P levels increased in both groups on the day after hcg administration (5.5 ± 1.1 ng/ml in high responders and 1.31 ± 0.15 ng/ml in low responders) but remained significantly different (P < 0.01). Similar trends, although less striking, were noted when daily T levels were evaluated (Fig. 5). T did not vary throughout the follicular phase in low responders (from 0.26 ± 0.05 to 0.30 ± 0.03 ng/ml) but did increase significantly in high responders (from 0.16 ± 0.04 to 0.46 ± 0.08 ng/ml; P < 0.01). Nevertheless, T did not reach abnormal levels. T levels in high responders correlated positively with LH (see above), FSH (r = 0.923, P < 0.002) and with P (r = 0.742, P < 0.05). PRL did not change significantly in low responders (from 9.4 ± 1.6 to 14.6 ± 2.8 ng/ml; P > 0.05), but in high responders there was significant increase (from 10.4 ± 1.8 to 20.9 ± 3.8 ng/ ml; p < 0.02), such that levels reached the upper limit of normal values (Fig. 6). PRL levels after hcg administration did not differ significantly between low and high responders (16.5 ± 2.8 and 25 ± 4.2 ng/ml; P > 0.05). PRL levels in high responders correlated positively with 2 (r 0.710, P < 0.05) and P (r = 0.918, P < 0.02). DISCUSSION Healthy women of similar ages and weights and with normal menstrual cycles can have high c> Q. ( :0 e ID ::I Q) (/"J I o +1 Figure 2 Serum 2 levels in low and high responders. The open circles represent the low responders (n = 5), and the closed circles represent the high responders (n = 5). Results (mean ± SM) are normalized to the day ofhcg administration. P < 0.05 on day - 5; P < 0.01 on days - 4 and - 3; P < on days - 2, -1, and O. 588 Ben-Rafael et al. FSH accumulation in hmg-treated patients Fertility and Sterility

4 a Mean 20 - Individual levels ly variable ovarian responses to a uniform treatment with hmg. For the purposes of evaluating these differences, we arbitrarily categorized patients as either low responders (peak 2 < 850 pg/ml) or high responders (peak 2 > 1500 pg/ml). It has been suggested that individual variability in response to gonadotropins and its reproducibility during successive treatment cycles may be attributed to differences in the number of ovarian follicles, number of FSH or LH receptors in ovarian cells, or basal levels of FSH and LH before treatment. 7 It is important that in this study the basal levels of FSH and LH were similar in both groups. In accordance with previous studies,1o-12 we observed a rise in serum FSH during the first few days of hmg treatment in all individuals. There I 0 Days before hcg administration. 2 g' :::> g... o 20 :x: 0- 'N : Q) -::J -' - Mean - Individual levels hcg I o... a I lot Mean - Individual levels o I 0 Days before hcg administration Figure 3 Mean and individual variations in daily serum LH levels in low and high responders. Results are normalized to the day of hcg administration. Mean serum LH levels in high responders on day - 1 were significantly higher than in low responders or pretreatment (P < 0.05) levels Figure 4 Mean and individual variations in daily serum progesterone levels in low and high responders. Results are normalized to the day of hcg administration. Mean serum progesterone levels in high responders starting on day - 2 were significantly higher than in low responders or pretreatment (P < 0.05) levels. Vol. 46, No.4, October 1986 Ben-Rafael et a1. FSH accumulation in hmg-treated patients 589

5 0.8 c Q) c e If) 0.2 hormones, we suggest that differences in the metabolism of the exogenously administered FSH are involved primarily in determining the different gonadal reactions, because significant differences in 2 between the two groups were found 1 day earlier than the divergence of FSH profiles. Individual differences in serum profiles of FSH and 2 that emerged within a few days after initiation of therapy are probably responsible for variations in blood levels of other gonadal and pituitary hormones during the latter part of the follicular phase. Increased LH concentrations were found during the last few days of hmg treatment almost exclusively in high responders I 1.0 -Mean Individual levels c 0.6 o Q) c o If) o +- If) / I hcg o I 0 +1 Figure 5 Mean and individual variations in daily serum testosterone levels in low and high responders. Results are normalized to the day of hcg administration. Mean serum testosterone levels in high responders on day 0 were significantly higher than in low responders or pretreatment (P < 0.01) levels. after, two strikingly different profiles of serum FSH and consequently of other hormones were seen throughout the remainder of the follicular phase. In low responders, the FSH levels (approximately 20 miu/ml) remained constant despite continued therapy; in high responders, the levels continue to rise for several more days (to 35 miui mi). Because the FSH in the blood of the hmg-treated women probably was derived from injected... C 20 -o o "0 Cl. 30 -Mean - Individuollevels hcg o +\ I Figure 6 Mean and individual variations in daily serum prolactin levels in low and high responders. Results are normalized to the day of hcg administration. Mean serum prolactin levels in high responders on day 0 were significantly higher than pretreatment (P < 0.02) levels but similar to the level in low responders. 590 Ben-Rafael et al. FSH accumulation in hmg-treated patients Fertility and Sterility

6 The short half-life of LH and the time (16 hours) between hmg injection and blood collection make it likely that these increases oflh primarily represent pituitary secretion of the gonadotropin. It has been shown in women 7, 13 and in monkeys 6, 14, 15 that LH surges are often absent in hmgstimulated cycles, despite 2 increases above normal midcycle levels. Suppression of LH secretion in gonadotropin-stimulated monkeys have been attributed to inhibition of the pituitary by an ovarian factor. 14, 15 Nevertheless, increases in serum LH can occur during stimulated cycles, but the response often is blunted, compared with the normal midcycle surge.6,7 Further, LH surges have been observed in patients whose serum 2 level reaches 2000 pg/ml. 16, 17 However, differences in 2 levels alone between high and low responders do not seem to account for the variations in LH secretion. The estrogen rise in low responders (who failed to secrete LH) was similar to that seen during normal cycles and often reached peak levels above that of normal women. Differences in the profiles of serum P are a more likely explanation for the presence or absence of LH surges in hmg-stimulated women. In high responders, surges in LH were accompanied and often preceded by a substantial rise in blood P, but P remained low and constant in low responders. A brief rise of P before ovulation is well documented in normal menstrual cycles, 18, 19 and recent studies have indicated that this P can synergize with estrogens in promoting pituitary secretion of LH.19 We suggest that the accelerated maturation of follicles in high responders promoted by the substantial accumulation of serum FSH leads to an earlier luteinization offollicular cells; this in turn facilitates the secretion of P when high responders continued to receive hmg. In contrast, ovarian follicles in low responders apparently had not reached the stage of development at which increases in P secretion could occur in response to hmg treatment. Only after injection of hcg were the levels of P elevated in this group. Positive correlation between LH, P, and T indicates that the significant rise to T in high responders also can be attributed to the rises in LH. However, FSH effect on T secretion cannot be determined based on these data. The preovulatory rise in serum PRL in high responders confirms previous observations that increases of this hormone are found in association with rising 2 levels in stimulated cycles.6, 11,20,21 The close correlation between P and PRL in high responders supports the findings that P may synergize with 2 to stimulate PRL secretion. 20 Further investigations are required to establish whether such increase in PRL can adversely affect fertility. The most significant finding of this study is that individual variations in response to hmg stimulation seems to be. related to the extent of FSH accumulation in serum. The few published studies on pharmacokinetics of hmg provide no clues as to the reasons for differences in FSH accumulation. Differences in absorption and clearance or metabolic alteration of the injection hormones resulting in slower clearance remain possible explanations. Changes in immunoreactivity of FSH between preparations cannot account for the differences between the groups, because the same batch was used for treating all of the patients. Further, high and low responders are known to repeat their hormonal patterns in successive cycles,7,22 independent of the lot of hmg. We suggest that further investigation of the absorption and clearance of hmg might provide a better understanding of the variable patterns of response. 2 levels of high responders in this study group are similar to the reported levels in anovulatory women suffering from complications associated with gonadotropin treatment.23,24 The small number of women and the occurrence of one pregnancy within our study do not, however, provide enough information for evaluating relationships between FSH accumulation, hormonal profiles, and the success of IVF-T. RFRNCS 1. Garcia J, Jones GS, Acosta AA, Wright G Jr: Human menopausal gonadotropin/human chorionic gonadotropin follicular maturation for oocyte aspiration: Phase 1, Fertil Steril 39:167, Ben-Rafael Z, Kopf GS, Blasco L, Flickinger GL, Tureck RW, Strauss JF, Mastroianni L Jr: Follicular maturation parameters associated with the failure of oocyte retrieval fertilization and cleavage in vitro. Fertil Steril 45:51, Laufer N, DeCherney AH, Haseltine FP, Polan ML, Mezer HC, Dlugi AM, Sweeney D, Nero F, Naftolin F: The use of high-dose human menopausal gonadotropin in an in vitro fertilization program. Fertil Steril 40:734, Mashiach S, Dor J, Rudak, Ben-Rafael Z, Rabinovitch 0, Serr DM, Nebel L, Goldman B: Comparison of methods for induction of ovulation in patients undergoing in vitro fertilization and embryo transfer. Am J Reprod Immunol 6:57,1984 Vol. 46, No.4, October 1986 Ben-Rafael et al. FSH accumulation in hmg-treated patients 591

7 5. Jone HW Jr, Acosta AA, Andrews MC, Garcia J, Jones GS, Mantzavinos T, McDowell J, Sandow B, Veeck L, Whibley T, Wilkes C, Wright G: The importance offollicular phase to success and failure in vitro fertilization. Fertil Steril40:317, Collins RL, Williams RF, Hodgen GD: ndocrine consequences of prolonged ovarian hyperstimulation: hyperprolactinemia, follicular atresia, and premature luteinization. Fertil Steril 42:436, Jones GS: Update on in vitro fertilization. ndocrinol Rev 5:62, Snedecor GW, Cochran WG: Statistical methods, sixth edition. Ames, IA, Iowa State University Press, 1967, p Kramer CY: xtension of multiple range tests to group means with unequal numbers of replications. Biometrics 12:307, Yuen BH, Sy L, Cannon W: Regulation of ovarian follicular and luteal function during treatment with exogenous gonadotropin in anovulatory infertility. Am J Obstet Gynecol 140:629, Healy DL, Burger HG: Serum follicle-stimulating hormone, luteinizing hormone, and prolactin during the induction of ovulation with exogenous gonadotropin. J Clin ndocrinol Metab 56:474, Snyder DL, Jaffe RB, Midgley AR Jr, Bolte : Regulation of human gonadotropins: XI. Hormonal profiles in normal and infertile women treated with gonadotropins. Fertil Steril 24:706, Fowler R, dwards RG, Walters, Chan STH, Steptoe PC: Steroidogenesis in preovulatory follicles of patients given human menopausal and chorionic gonadotropins as judged by the radioimmunoassay of steroids in follicular fluids. J ndocrinol 77:161, Sopelak VM, Hodgen GD: Blockade of the estrogeninduced luteinizing hormone surge in monkeys: a nonsteroidal, antigenic factor in porcine follicular fluid. Fertil Steril 41:108, Schenken RS, Hodgen GD: Follicle-stimulating hormone induced ovarian hyperstimulation in monkeys: blockade of the luteinizing hormone surge. J Clin ndocrinol Metab 57:50, Ferraretti AP, Garcia J, Acosta AA, Jones GS: Serum luteinizing hormone during ovulation induction with human menopausal gonadotropin for in vitro fertilization in normally menstruating women. Fertil Steril 40:742, Rossavik IK, Gibbons W, Findley W, Young RL, Dodson MG, Poindexter AN: Mathematical methods to enhance the value of ultrasound data in the in vitro fertilization and embryo transfer program. (Abstr 120) Presented at the Forty-first AnilUal Meeting of the American Fertility Society, Chicago, IL, September 28 to October 2, Trounson AO, Calabrese R: Change in plasma progesterone concentration around the time of the luteinizing hormone surge in women superovulated for in vitro fertilization. J Clin ndocrinol Metab 59:1075, Hoff JD, Quigley M, Yen SSC: Hormonal dynamics at midcycle: are-evaluation. J Clin ndocrinol Metab 57: 792, Williams RF, Gianfortoni JG, Hodgen GD: Hyperprolactinemia induced by an estrogen-progesterone synergy: quantitative and temporal effects of estrogen priming in monkeys. J Clin ndocrinol Metab 60:126, Kemmann, Gemzell CA, Beinert WC, Beling CB, Jones JR: Plasma prolactin changes during the administration of human menopausal gonadotropins in nonovulatory women. Am J Obstet GynecoI129:145, Diamond MP, Wentz AC, Vaughn WK, Webster BW, Herbert CM, Osteen KG, Maxson WS: Outcome of successive cycles of ovulation induction in the same individual. Fertil Steril 43:369, Ben-Rafael Z, Dor J, Mashiach S, Blankenstein J, Lunenfeld B, Serr DM: Abortion rate in pregnancies following ovulation induced by human menopausal gonadotropin/human chorionic gonadotropin. Fertil Steril 39:157, Lunenfeld B, Insler V: Gonadotropins. In Diagnosis and Treatment of Functional Infertility. Berlin, Grosse Verlag, 1978, p Ben-Rafael et al. FSH accumulation in hmg-treated patients Fertility and Sterility