Suppression of the ovary using a gonadotropin releasing-hormone agonist prior to stimulation for oocyte retrieval*

IIII

I

:

: II

FERTILITY AND STERILITY

Vol. 48, No.5, November 1987 Printed in U.S.A.

Copyright c 1987 The American Fertility Society

Suppression of the ovary using a gonadotropin releasinghormone agonist prior to stimulation for oocyte retrieval*

Dominique de Ziegler, M.D. Marcelle I. Cedars, M.D. Denise Randle, R.N.

John K. H. Lu, Ph.D. Howard L. Judd, M.D. David R. Meldrum, M.D.t

Department of Obstetrics and Gynecology, University of California Los Angeles (UCLA) School of Medicine, Los Angeles, California

Five women with prior suboptimal ovarian stimulation for in vitro fertilization and embryo transfer (IVF-ET) were pretreated with a long-acting gonadotropin-releasing hormone (GnRH) agonist beginning in the midluteal phase of the preceding menstrual cycle. The four women with normal luteal function had castrate estrogen levels following regression of the corpus luteum, whereas one woman with an abnormal luteal phase and perimenopausal levels of gonadotropins had an agonistic response. In the three women with adequate stimulation, 20 oocytes were obtained and one woman became pregnant. Initiation of GnRH agonist therapy during the luteal phase of a normal menstrual cycle may be an efficient way of obtaining ovarian suppression without an agonistic response. Fertil Steril 48:807, 1987

The extent of multiple follicular growth in response to stimulation is one of the most important factors influencing successful pregnancy following in vitro fertilization and embryo transfer (IVFET). Some women respond to stimulation with a small number of asynchronous follicles (poor responders). We have observed an unusually good response in women with menstrual intervals longer than 35 days, suggesting that stimulation earlier in the recruitment process may lead to a better ovarian response.! In order to allow stimulation earlier in the evolution of ovarian processes resulting in selection of a dominant follicle, we have preceded stimulation in poor responders by ovarian suppression using a long-acting gonadotropin-releasing hormone agonist (GnRH-a). MATERIALS AND METHODS

Five women who had experienced one or more unsuccessful cycles of ovarian stimulation in our Received April 6, 1987; revised and accepted June 15, 1987. * This work was supported by grants from the U.S. Public Health Service (AG-01512, AG-4810, and CA-23093). t Reprint requests: David R. Meldrum, M.D., AMI-South Bay Vol. 48, No.5, November 1987

IVF-ET program volunteered for study. These patients had previously experienced a suboptimal response to ovarian stimulation with human menopausal gonadotropins (hMG), resulting in cancellation in one or more completed unsuccessful cycles. All subjects had regular ovulatory cycles. A long-acting agonist of GnRH [(Imbzl) DHis 6 ,Pro9 -NEt]-GnRH (GnRH-a; 100 JLg) was self-administered subcutaneously daily for 14 days, starting on the sixth day of basal body temperature (BBT) elevation. Recording of BBT was continued during treatment and occurrence of menstrual bleeding was recorded by the subjects. Individuals whose fallopian tubes were not absent or occluded were advised to use barrier contraception during the cycle they received GnRH -a. On the second day after discontinuation of GnRH -a treatment, a pelvic ultrasound scan was performed to exclude the presence of cysts, and ovarian stimulation was carried out in an identical manner to prior unsatisfactory cycles. Three ampules of hMG (75 units of follicle-stimulating horHospital IVF Center, 514 North Prospect Avenue, Redondo Beach, California 90277. de Ziegler et a1.

Ovarian response following suppression

807

mone [FSH] and 75 units of luteinizing hormone [LH] per ampule, Serono Laboratories, Inc., Randolph, MA) were injected daily. The dose was kept constant until the fifth day of therapy, when daily ultrasound scans were begun to assess ovarian response. Pelvic ultrasonography was carried out using a Philips 2000 sector scanner (Philips Ultrasound Systems, Santa Ana, CA) transabdominally or transvaginally, as appropriate. The dose ofhMG then was increased to 5 ampules/day (2 ampules in the morning and 3 in the evening). Human chorionic gonadotropin (hCG) was administered (10,000 IU intramuscularly [1M]) when two follicles were at least 1.5 cm in their longest diameters and estradiol (E 2) levels appeared appropriate for the number of follicles. If development of a single dominant follicle was visualized, hCG was withheld and the IVF cycle was cancelled. The hCG injection was timed so that it preceded laparoscopy by 34! to 35 hours. Insemination of oocytes, ET, and treatment with progesterone (P; 25 mg 1M daily) after transfer were carried out according to our standard procedures. 2 Blood samples were obtained at 9 A.M. immediately before and every 48 hours during GnRH-a administration, then daily until hCG was given or the IVF cycle was cancelled. The serum levels of FSH, E 2, estrone (E 1 ), and P were measured using previously described radioimmunoassay (RIA) techniques. 3-6 During hMG administration, serum E2 was determined by rapid RIA using a p25 E2 kit (Pantex, Santa Monica, CA).

RESULTS As shown on Figure 1, four of our five subjects receiving GnRH -a had complete ovarian suppres-

I

~

100

10

80

8

y 60

6

40

4

'"

~

'" 0:

0-

0

I

6

E "m 0:

e

~

w 20

2

0

0

i

a. I

234

5

6

7

8

9

~

" ~ ~ ~

GnRH analOIi (dOY!l.L

Figure 1 Mean ± SEM E2 (pg/ml) and P (ng/ml) levels immediately prior to GnRH-a treatment on the sixth day of BBT elevation and every 48 hours thereafter in four subjects whose ovarian function was suppressed at the end of GnRH-a treatment. 808

de Ziegler et at Ovarian response following suppression

200

10

160

8

~

120

6

~co

~

80

4

§

';;

40

0:

8:

e

w

I

cr'

';;

...... "0----0- .. --0...

2

"''''----0-- __0

:g,

e

a. 0

0 I

2 3 4

5 6

7 8

9 10"

12 13 14

GnRH analog (days I

Figure 2 Serum E2 and P concentrations before GnRH-a treatment and every 48 hours thereafter in a patient with an agonistic response to GnRH-a.

sion at the end of the 14-day treatment. Prior to treatment with GnRH -a, mean ± standard error (SE) serum levels of E2 and P were 99 ± 22 pg/ml. and 8.8 ± 1.1 ng/ml, respectively, which were within the expected range for the midluteal phase of the menstrual cycle. At the end of treatment, serum El and E 2 levels were in the castrate range at 17 ± 8 and 5 ± 1 pg/ml, respectively. Initiation of GnRH -a treatment during the midluteal phase did not appear to have an impact on corpus luteum function, as suggested by the lack of alteration seen in the expected pattern of hormone secretion. The decline in serum E2 levels paralleled those of P on the 13th day of the BBT rise, indicating a normal duration of the luteal phase despite GnRH -a treatment, and menstrual bleeding occurred at the expected time. A moderate elevation in serum LH took place after initiation of GnRH -a treatment, rising from the pretreatment value of 4.5 ± 1.4 mlU/ml to 25 ± 5 mlU/ml 48 hours later. Subsequently, LH levels stayed in that range throughout GnRH-a treatment. Elevation in serum FSH levels after initiation of GnRH-a was less pronounced and briefer, followed by a decline to concentrations at or below pretreatment levels. After completion of GnRH -a treatment, ultrasound examination did not reveal any ovarian enlargement in these four patients. In sharp contrast, the remaining subject manifested a different pattern of hormone secretion after GnRH -a administration. As shown in Figure 2, prior to GnRH-a treatment, the serum P concentration was abnormally low in this last subject (2.1 ng/ml) compared with values expected and found during the midluteal phase in the four other subjects. That the low serum P concentration in this subject reflected an abnormal luteal phase rather than inaccurate timing is indicated by occurFertility and Sterility

rence of menstrual bleeding at the expected time concomitant with a decrease in serum P below 1 ng/ml. In this patient, initiation of treatment with GnRH-a resulted in a marked agonistic response of the ovary (Fig. 2) with rapidly rising serum El and E2 levels reaching 221 and 462 pg/ml, respectively, at the end of GnRH-a treatment, and development of three large follicles exceeding 3 cm on ultrasound examination, thus precluding administration of hMG. The decreasing concentration of P induced a clinically normal menstrual period, despite concomitantly increasing serum estrogen levels. Initiation of GnRH -a induced a more pronounced elevation in LH level from 19.2 mIU/ml before GnRH-a to 75.8 mIU/ml 48 hours after the onset of treatment. In contrast, no elevation in serum FSH was discernible. An unexpected finding was that the basal FSH and, to a lesser extent, LH levels were unexpectedly elevated at 33.8 and 19.2 mIU/ml, respectively. We have subsequently confirmed perimenopausal levels of gonadotropins in this subject, despite persistence of regular menstrual cycles. Ofthe four subjects who were given hMG, stimulation of multiple follicular development was achieved in three subjects, while in the fourth, the IVF cycle was cancelled because of the development of a single dominant follicle. In the three subjects undergoing laparoscopy, a total of 25 follicles had developed (Table 1), which exceeded the total of nine follicles during the best previous IVF cycle for each patient. Similarly, a larger number of oocytes (20) were retrieved in the three IVF cycles following GnRH-a treatment. Patient number 5 became pregnant and delivered a healthy baby girl. Of the total of 20 oocytes retrieved in these three patients, six were immature, with germinal vesicles present. This unusual number of immature oocytes suggests that criteria for hCG administration may have to be modified when pretreatment with Table 1 Outcome of IVF in Four Patients· Who Recieved hMG for Follicular Stimulation After Temporary Ovarian Suppression with GnRH-a Patient

Prior cycle

After GnRH-a

E2 (on day of hCG administration)

3 4 5

385 566 780

1002 866 812

Follicles (oocytes) (>1 cm)

3 4 5

2 (-) 3 (2) 4 (3)

7 (8) 8 (2) 10 (10)

• One patient again responded with a single dominant follicle. Vol. 48, No.5, November 1987

GnRH-a is given because of the differing status of follicular activity at the initiation of stimulation. DISCUSSION

The initial stimulation of gonadotropin release observed in the present study during the luteal phase was markedly less than we previously observed using the same agonist and dose during the early follicular phase of the menstrual cycle. 7 This finding is consistent with a previous report of blunting of the initial response to GnRH agonist administration in the luteal phase of the menstrual cycleS and confirms a similar study using continu0us GnRH agonist infusion to initiate treatment of uterine myomata. 9 These observations, and our previous finding that 4 out of 16 women receiving the same treatment beginning on day 2 of the menstrual cycle had an agonist response,lO suggest that the midluteal phase of a normal menstrual cycle may be an ideal time for initiation of GnRH -a therapy. Initiation of GnRH-a therapy during the luteal phase may require confirmation of normal luteal function using rapid measurements of serum P or urinary pregnandiol glucuronide, which are now available. During agonist administration, the level of FSH rose transiently and returned to concentrations at or below the luteal level prior to drug injection. Thus the normal rise of FSH during late luteal and early follicular phases, which initiates follicle growth, did not occur. The rise of LH during agonist therapy has been shown to be due to detection of nonbioactive LH fragments. ll During administration of GnRH analogues, bioactive FSH 12 and LHll are markedly suppressed, whereas immunoreactive gonadotropin concentrations do not show these changes. The marked suppression of E 2 levels thus provides a more accurate reflection of circulating bioactive gonadotropins. The occurrence of pregnancy in subject number 5 after transfer of three cleaving embryos and the delivery of a healthy baby girl suggest that GnRH-a had no deleterious effects on developing oocytes when administered for 14 days preceding hMG stimulation. However, the unusual number of immature oocytes retrieved led us to postulate that either GnRH-a or the consequent decrease in ovarian estrogen production may have modified the maturation of developing oocytes relative to follicle size, thus affecting the appropriate timing for hCG administration, or that hMG stimulation early in the process of recruitment allowed some de Ziegler et al.

Ovarian response following suppression

809

r follicles to develop which otherwise would have rapidly become atretic. New criteria for completion of follicular maturation and injection of hCG may need to be established when ovarian suppression with GnRH-a precedes stimulation with hMG. The outcome of IVF cycles that followed ovarian suppression with GnRH -a was moderately encouraging. In monkeys given the same dose of gonadotropin before and during administration of a GnRH antagonist,13 a tendency toward a lower incidence of poor response was observed. It was hoped that, by stopping GnRH-a, endogenous gonadotropin secretion would recover to a level similar to that in a spontaneous cycle, thus allowing examination of the single variable of the differing stage of follicle maturation, whereas, in the previous investigation, total gonadotropin stimulation of the ovaries may have been less during GnRH analogue administration because of the continuing suppression of endogenous gonadotropin secretion. Although our prior study of GnRH-a treatment of normally cycling women for 28 days indicated rapid recovery of cyclic ovarian function,14 it is possible that endogenous gonadotropin secretion was not fully recovered at the time of initiation of hMG stimulation. Recently, Sher et al. 15 obtained adequate ovarian responses in 20 women who failed to respond adequately to 3 ampules of hMG and were treated with progressively increasing doses of hMG, seemingly obviating the need to devise other therapeutic approaches for poor responders, although the pregnancy rate in those women was not given. The role of GnRH agonist suppression is particularly helpful in the treatment of women who manifest premature LH surges during preparation for IVF -ET by maintaining suppression during ovarian stimulation. Further refinements of the hMG dose and the timing of hCG administration following GnRH-a suppression hold the possibility of improving follicle response together with obviating the problems inherent in the occurrence of premature LH elevations. Acknowledgments. The GnRH agonist was kindly supplied by Ortho Pharmaceutical Corporation. We thank Elisa Obnial, John Lee, and Philip LaPolt for expert technical help and Teresa McGrath for efficient assistance in preparing the manuscript. REFERENCES 1. Cedars MI, Steingold KA, de Ziegler D, Chetkowski R, Hamilton F, Meldrum DR: Outcome of ovarian stimulation,

810

de Ziegler et al.

Ovarian response following suppression

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

in vitro fertilization and embryo transfer in oligomenorrheic women. Unpublished data Meldrum DR, Chetkowski R, Steingold KA, de Ziegler D, Cedars MI, Hamilton M: Evolution of a highly successful in vitro fertilization-embryo transfer program. Fertil Steril 48:86,1987 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 30:325, 1970 Yen SSC, Llerena LA, Little B, Pearson OH: Disappearance rates of endogenous luteinizing hormone and chorionic gonadotropin in man. J Clin Endocrinol Metab 28:163, 1968 De Vane GW, Czekala NM, Judd HL, Yen SSC: Circulating gonadotropins, estrogen and androgens in polycystic ovarian disease. Am J Obstet Gynecol 121:496, 1975 Anderson DC, Hopper BR, Lasley BL, Yen sse: A simple method for the assay of eight steroids in small volumes of plasma. Steroids 28:170, 1976 De Fazio J, Meldrum DR, Lu JKH, Vale WW, Rivier JE, Judd HL, Chang RJ: Acute ovarian responses to a long-act-· ing agonist of gonadotropin-releasing hormone in ovulatory women and women with polycystic ovarian disease. Fertil Steril 44:453, 1985 Araki S, Chikazawa K, OMotoyama M, Ijima K, Abe N, Tamada T: Reduction in pituitary desensitization and prolongation of gonadotropin release by estrogen during continuous administration of gonadotropin-releasing hormone in women: its antagonism by progesterone. J Clin Endocrinol Metab 60:590, 1985 Healy DL, Lawson SR, Abbott M, Baird DT, Fraser HM: Toward removing uterine fibroids without surgery: subcutaneous infusion of a luteinizing hormone-releasing hormone agonist commencing in the luteal phase. J Clin Endocrinol Metab 63:619, 1986 Steingold KA, Cedars M, Lu JKH, Randle D, Judd HL, Meldrum DR: Treatment of endometriosis with a long-acting gonadotropin-releasing hormone agonist. Obstet Gynecol 69:403, 1987 Meldrum DR, Tsao Z, Monroe SE, Braunstein GD, Sladek J, Lu JKH, Vale W, Rivier J, Judd HL, Chang RJ: Stimulation of LH fragments with reduced bioactivity following GnRH agonist administration in women. J Clin Endocrinol Metab 58:755, 1984 Dahl KD, Pavlou SN, Kovacs WJ, Hsueh AJW: The changing ratio of serum bioactive to immunoreactive follicle-stimulating hormone in normal men following treatment with a potent gonadotropin releasing hormone antagonist. J Clin Endocrinol Metab 63:792, 1986 Kenigsberg D, Littman BA, Williams RF, Hodgen GD: Medical hypophysectomy. II. Variability of ovarian response to exogenous gonadotropins. Fertil Steril 42:116, 1984 Meldrum DR, Chang RJ, Lu J, Vale W, Rivier J, Judd HL: "Medical oophorectomy" using a long-acting GnRH agonist: a possible new approach to the treatment of endometriosis. J Clin Endocrinol Metab 54:1081, 1982 Sher G, Knutzen V, Stratton C, Chotiner H, Mayville J: In vitro fertilization and embryo transfer: two-year experience. Obstet Gynecol 67:309, 1986

Fertility and Sterility