Journal of in Vitro Fertilization and Embryo Transfer, Vol. 7, No. 6, 1990
The Use of Gonadotropin Releasing Hormone Agonist (GnRHa) in Good Responders Undergoing Repeat in Vitro Fertilization/Embryo Transfer (IVF/ET) 1 JAMES H. SEGARS, 2'3 GEORGE A. HILL, 2'4 STEPHANIE H. BRYAN, 2 CARL M. HERBERT, 11I,2 KEVIN G. OSTEEN, 2 B. JANE ROGERS, 2 and ANNE COLSTON WENTZ z'5
Submitted: January 29, 1990 Accepted: July 10, 1990
patients previously classified as "good responders" undergoing ovarian stimulation for IVF/ET.
The use of gonadotropin releasing hormone agonists (GnRHa) has been shown to improve the response in patients classified as "poor responders" undergoing ovarian stimulation for in vitro fertilization~embryo transfer (IVF/ ET). This study sought to determine whether GnRHa therapy would benefit patients undergoing IVF/ET who had been classified as "good responders" in prior attempts. Twenty-three patients who had completed a prior IVF/ET attempt but who failed to conceive underwent ovarian stimulation using a combination of GnRHa and human menopausal gonadotropin (hMG). Each patient's prior stimulation served as her control and consisted of clomiphene citrate (CC)/hMG in 18 patients and follicle stimulating hormone (FSH) and~or hMG in 5 patients. The numbers of oocytes retrieved, oocytes fertilized, embryos cleaved, and embryos transferred were all significantly greater in cycles treated with GnRHa/hMG compared to control cycles. The clinical pregnancy rate was 39% and the ongoing pregnancy rate was 26% during the cycle when GnRHa pretreatment was utilized. These data suggest that GnRHa therapy is of benefit even to those
KEY WORDS: gonadotropin releasing hormone (GnRH) ago-
nists; in vitro fertilization/embryotransfer; ovarian hyperstimulation. INTRODUCTION Despite recent technological improvements, success rates of in vitro fertilization/embryo transfer (IVF/ET) remain low. Cancellation of patients due to a poor ovarian response or premature luteinizing hormone (LH) surges accounts for 25 to 30% of patient attrition. Use of gonadotropin releasing hormone agonists (GnRHa) has been shown to improve the response in patients undergoing IVF/ET who have been classified as poor responders (1-4). Reports have also suggested that routine use of GnRHa for all patients undergoing IVF/ET may improve results by decreasing cancellations due to premature LH surges (5-8). Chetkowski et al. (9) prospectively compared the utilization of human menopausal gonadotropins (hMG) with and without the addition Of GnRHa in "unselected patients." These investigators were able to document that the use of the GnRHa with hMG resulted in an improvement in the fertilization and implantation rates and a decrease in the spontaneous abortion rate in patients undergoing IVF/ET. Furthermore, Macnamee et al. (10) recently demonstrated in a large prospective trial that short-term GnRHa treatment with hMG resulted in improvement when compared to a similar group of patients treated with a combination of clomiphene citrate (CC)/hMG. The ongoing pregnancy rate per treatment cycle was 27% in
i Presented at the 45th Annual Meeting of the American Fertility Society, November 1989, San Francisco, California. 2 Center for Fertility and Reproductive Research (C-FARR), Department of Obstetrics and Gynecology, Vanderbilt University Medical Center, Nashville, Tennessee 37232. 3 Present address: Building 6, Room 3A-15, Laboratory of Developmental Molecular Immunity, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland 20892. 4 To whom correspondence should be addressed at Center for Fertility and Reproductive Research, Department of Obstetrics and Gynecology, Room D-3223, Vanderbilt University Medical Center North, Nashville, Tennessee 37232. 5 Present address: Northwestern Medical Faculty Foundation, Inc., Prentice Pavilion, 333 East Superior Street, Suite 1165, Chicago, Illinois 60611.
327
0740-7769/90/1200-0327506.00/0 9 1990HenumPublishingCorporation
328
SEGARS ET AL.
the patients treated with the GnRHa/hMG protocol, compared to 14% for the CC/hMG protocol. Both of these studies (9,10) document an improvement in cycle outcome between groups of patients when GnRHa is utilized in the ovarian stimulation protocol. However, it remains unclear whether an individual patient classified as having a good response to ovarian stimulation would benefit by the addition of GnRHa to the stimulation regimen. Most studies addressing this question have employed a between groups analysis and have not specifically focused on "good responders." It may be that the improvement observed when all patients (i.e., groups of patients) are pretreated with GnRHa may be due solely to an improvement in the patients who are poor responders. This is an important point, for if both "poor" and "good responders" improve with adjunctive GnRHa treatment, then use of GnRHa would appear indicated for virtually all patients attempting IVF. For these reasons, we sought to determine if the adjunctive use of GnRHa (leuprolide acetate; LA) in patients previously classifted as having a good response during an IVF/ET cycle would result in an improved outcome during the GnRHa-treated cycle. The design differs from prior studies of GnRHa usage in normal responders in that responses of individual patients (not groups) were compared, thus minimizing interindividual variability.
MATERIALS AND METHODS Twenty-three patients, ages 26 to 42 years [33.9 - 4.4 (mean +-- standard deviation)], admitted to the program for IVF/ET at Vanderbilt University Medical Center in Nashville, Tennessee, between October 1988 and February 1989 were evaluated. Eighteen patients (Group A) had previously undergone ovarian stimulation with CC (Serophene, Serono Laboratories, Inc., Randolph, M A ) / h M G (Pergohal, Serono Laboratories, Randolph, MA). Additional patients treated with follicle stimulating hormone (FSH; Metrodin, Serono Laboratories, Inc., Randolph, MA; n = 2) or hMG (n = 3) [Group B; n = 5] were included. All had successfully progressed through the previous cycle, had oocytes recovered, achieved fertilization, and underwent embryo transfer. No control cycle had been cancelled because of a poor estradiol (E2) response or a premature L H surge. By all criteria, these patients had
demonstrated a good response in the previous cycle; however, none had conceived. The 23 participating patients were begun on LA (Tap Pharmaceuticals, North Chicago, IL), 1 mg per day subcutaneously (SQ) beginning on cycle day 1. After 7 to 9 days of treatment, serum estradiol (Ez) was measured. If the serum E 2 level was less than 30 pg/ml, hMG was begun at 3 ampoules per day and the LA was decreased to 0.5 mg per day SQ. If the serum E2 was greater than 30 pg/ml, LA was continued at 1 mg/day SQ for 2 to 3 days and an E 2 and ultrasonogram (USG) were performed. If the E z remained greater than 30 pg/ml but no USG documentation of cysts was present, the LA dosage was changed to 0.5 mg SQ twice a day. Once suppression was documented by an E2 of less than 30 pg/ml, hMG was begun and the patients were followed with E2 and USG beginning on day 5. LA was continued at 0.5 mg per day SQ until the day of administration of human chorionic gonadotropin (hCG). Thirty-five hours following hCG, the patients underwent retrieval of oocytes with transvaginal USG guidance. Two patients in Group A were excluded for reasons not related to ovarian stimulation. The techniques of oocyte and embryo culture have been previously published (11). At the time of retrieval, oocytes were graded on a scale of 1 to 5 as previously described (11). Briefly, an oocyte graded as 1 was considered immature, 3 was considered mature preovulatory, and 5 was considered atretic. Grades 2 and 4 were considered slightly immature and slightly overmature, respectively, but were treated as mature oocytes. Rates of mature oocyte recovery, fertilization, embryo cleavage, and embryo transfer were compared in control (CC/hMG, FSH, hMG) and treated (GnRHa/hMG) cycles. Pregnancy rates are reported but not compared statistically, as all patients by design of the study had not conceived during a recent attempt at IVF/ET. As a further control for the possibility that the study patients had an initial poor CC/hMG (control) stimulation, the 16 patients in Group A were compared to a matched control group (Group C) of 16 patients undergoing CC/hMG stimulation in the current IVF/ET cycle. This additional control was an attempt to address the question of whether the previous (control) stimulation of the study patients was indicative of a good response on CC/hMG and to make sure that other improvements in the IVF/ET program over time were not responsible for the differences between the groups.
Journal o f in Vitro Fertilization and Embryo Transfer, VoI. 7, No. 6, 1990
329
GnRHa IN GOOD RESPONDERS UNDERGOING IVF/ET
Statistical Analysis Data are expressed as mean -+ standard deviation (SD). Analysis of data was performed for the study patients with a paired t test with the alpha level set at 0.05. Differences between control cycles in CC/ hMG-treated patients (Group A) and cycles in the second control group (Group C) were determined with Student's t test.
DISCUSSION
RESULTS In the study group, levels of estradiol at the time of hCG were not significantly different when compared in control and treated cycles. Control and treated cycles varied substantially, however, in the amounts of hMG administered and the number of days of hMG therapy (Table I). The rate of mature oocyte recovery (mean -+ SD) for all study patients in the GnRHa/hMG cycle was significantly greater than in control cycles (Table II). Perhaps as a result of the greater number of oocytes recovered at the time of surgery, fertilization, cleavage, and embryo transfer rates were also significantly greater in GnRHa/hMG vs control cycles in study patients. It did not appear that treatment with GnRHa resulted in any inhibition of embryo development in vitro. The difference between GnRHa/hMG and control cycles in the study group ultimately resulted in a greater number of embryos transferred in GnRHa/hMG-treated patients. Crude pregnancy rates are reported in Table II. Comparison of the pretreatment responses of the 16 CC/hMG patients (Group A) to an additional 16 patients undergoing recent CC/hMG stimulation failed to reveal any significant difference among the numbers of oocytes retrieved, oocytes fertilized, Table I. Characteristics of Ovarian Stimulation in Patients with and Without G n R H a Treatment a Study group (n = 21) Prior cycle Age E2 at the time ofhCG Ampoules o f h M G Days of stimulation
embryos cleaved, or embryos transferred (Table II). The 16 patients having a prior CC/hMG stimulation (Group A) exhibited significant differences between control and GnRHa-treated cycles when examined (Table II). Differences between former stimulated cycles and GnRHa/hMG cycles were not significant in Group B, perhaps due to the small group size (n = 5) (data not shown).
Group C, (n = 16) CC/hMG cycle GnRHa/hMG matched cycle control
33.3 -+ 4.6
--
33.4 -+ 3.6
1184 -+ 578 11.3 --+ 5.8
1351 - 472 25.6 --- 5.1"
1364 + 486 8.5 - 2.7
8.2 -+ 1.I
10.1 + 1.6"
8.5 -+ 1.2
a Data expressed as mean __- standard deviation. * P < 0.05 compared to prior stimulation.
Investigators continue to search for IVF/ET stimulation protocols which will maximize oocyte recovery and ultimately improve pregnancy rates. The CC/hMG regimen has proven superior in terms of oocytes recovered, fertilization and cleavage rates, and numbers of embryos transferred. Unfortunately, this has not translated into improved pregnancy rates (12). Several studies have compared the use of GnRHa to the standard management of patients undergoing ovarian stimulation in poor responders. These patients generally had exhibited a poor E 2 response to hMG stimulation or had manifested a premature L H surge in prior cycles. The data in these reports suggest that this group of patients clearly benefits from GnRHa pretreatment (2,4,8). Neveu e t al. (8) have investigated the adjunctive use of GnRHa treatment in a group of normal patients attempting IVF/ET. Their study compared 10 patients being treated with FSH alone to I0 patients stimulated with G n R H a and FSH. The group treated with GnRHa had a statistically greater number of oocytes retrieved and number of embryos/ oocyte (fertilization rate), in contrast to the studies by Neveu e t al. (8) and Macnamee e t al. (10), we find that the greater number of oocytes retrieved, greater fertilization rate, and greater cleavage rate resulted in an increased number of embryos available for transfer. This may be a significant finding, since the number of embryos transferred correlates with the rate of pregnancy. Our data confirm the data of Chetkowski e t al. (9) but extend their findings and demonstrate that a preselected group of "good responders" will benefit from adjunctive short-term use of GnRHa. Frydman e t al. (6) also reported the efficacy of adjunctive GnRHa in IVF/ET patients with previous normal responses. The study by Frydman e t al. (6) employed a retrospective design with a comparison to a prior CC/hMG stimulation. In the normal
Journal o f in Vitro Fertilization and Embryo Transfer, Vol. 7, No. 6, 1990
SEGARS ET AL.
330
Table II. Results of Oocyte Recovery in Patients With and W i t h o u t G n R H a T h e r a p y a
Prior stim.
GnRHa/hMG
CC/hMG
GnRHa/hMG
Group C (matched controls for CC/hMG) (n = 16), CC/hMG
5.6 --- 2,9 3.7 +- 1.9
8.2 - 3.9* 4.9 -+ 2.5*
5.6 -+ 2.8 3.8 -+ 2.1
8.5 -+ 4.1" 5.3 +- 2.6*
5.2 + 2.8 3.2 -+ 2.4
2.9 -+ 1.4 2.6 -+ 1.1 0b
4.3 +- 2.2* 3.4 -+ 1.1' 9/21
2.7 --- 1.4 2.5 - 1.2 0b
4.7 +- 2.2* 3,4 +- 1.1" 6/16
3.0 +- 2.2 2.7 -+ 1.5 3/16
Group A (prior C C / H M G ) (n = 16)
All study patients (n = 21)
Mature oocytes recovered Oocytes fertilized E m b r y o s undergoing cleavage E m b r y o s transferred Pregnancies
a Data e x p r e s s e d as m e a n -+ standard deviation. b By study design, patients had not conceived in prior IVF/ET attempts. * P < 0.05 compared to prior stimulation.
ovulatory patients undergoing IVF/ET, the ongoing pregnancy rate increased from 10.4% in prior cycles to 22.4% in the agonist-treated group. No data were presented to determine whether this difference was due to an increase in the number of oocytes recovered or if the changes in pregnancy rate could have been due to other factors. The observation that the use of GnRHa for all patients undergoing stimulation results in an improved outcome does not satisfactorily address the question of whether GnRHa should be given to good (normal) responding patients. That is, the improvement in the total group's response (all patients) may be due solely to an improvement among the poor responders. To address the question of whether adjunctive use of GnRHa is of benefit to "good responders," we evaluated a group of patients who had a satisfactory previous response to ovarian stimulation in an IVF attempt but who did not conceive. This design permitted a paired comparison of the prior stimulation to the current GnRHa/hMG cycle. The design minimized the interindividual variability in the response to ovarian stimulation. Our results suggest that GnRHa/hMG therapy is of benefit to patients who have previously progressed to surgery and embryo transfer and could be classified as good responders on a CC/hMG regimen. A significantly greater number of mature oocytes was retrieved, which resulted in a greater number of embryos available for transfer. The ability of this study to detect a difference in the number of embryos transferred (vs earlier studies) is likely due to the comparison of individual patients and not groups of patients. Because this study compared the response to a clomiphene/hMG stimulation with a GnRHa/hMG regimen, it is possible that the observed differences may be due to other fac-
tors (i.e., clomiphene or hMG dosage) and not GnRHa use alone. Other studies comparing GnRHa/ hMG to hMG stimulations have noted similar directional trends (13,14), therefore it would not seem that the presence of clomiphene in the control stimulations would be entirely responsible for the improvements observed. The clinical pregnancy rate of all patients treated with GnRHa was 39% (9 of 23). There have been two spontaneous abortions, one ectopic pregnancy, and six ongoing pregnancies. Unfortunately, we cannot make a statistical statement regarding pregnancy rates, since by design the patients chosen for study had not conceived in their prior cycle. There were six pregnancies in the GnRHa Group A (6 of 16; 37%) and three pregnancies in the matched controis (Group C; 18%). Interestingly, the percentage of pregnancies in the matched control group is consistent with that of our center for a CC/hMG stimulation (18%). Further studies will focus on this issue in an attempt to determine whether the observed improvements in outcome translate to an increased pregnancy rate in good responders. As an additional control, we evaluated a group of 16 patients (Group C) undergoing CC/hMG stimulation in the same cycle as the study group receiving GnRHa/hMG therapy (Group A). Comparison of these 16 patients to 16 study patients' CC/hMG stimulation revealed that there was no significant difference in the rates of oocyte recovery, fertilization, cleavage, and embryo transfer in the matched control group (Group C) as compared to the paired control group (Group A) of study patients. Thus, it appears that we had chosen a study group that was representative of patients undergoing stimulation with CC/hMG and not a group of poor responders. These data reveal that in those patients known to
Journal o f in Vitro Fertilization and Embryo Transfer, Vol. 7, No. 6, 1990
GnRHa IN GOOD RESPONDERS UNDERGOING IVF/ET
be good responders during a CC/hMG stimulation, adjunctive treatment with GnRHa significantly improves the rates of oocyte recovery, oocyte fertilization, embryo cleavage, and number of embryos transferred. The study suggests that the improvement observed in ovarian stimulation when GnRHa therapy is utilized to treat all patients is due not only to improvement in patients who demonstrate a poor response to ovarian stimulation regimens, but also to improvement in patients who have demonstrated prior good responses. Studies are ongoing to determine whether the observed increase in number of embryos transferred results in a significant improvement in pregnancy outcome.
331
4.
5.
6.
7.
8.
ACKNOWLEDGMENTS
The authors wish to acknowledge the excellent secretarial assistance of Ruth Robertson. This study was supported in part by a materials grant from TAP Pharmaceuticals, North Chicago, Illinois. The authors acknowledge the superb technical assistance of Christina Bastias, M.D., Peter Fetterolf, Ph.D., Melanie Freeman, M.S., and Thomas Thompson, Research Assistant.
REFERENCES 1. Awadalla SG, Friedman CI, Chin NW, Dodds W, Park JM, Kim MH: Follicular stimulation for in vitro fertilization using pituitary suppression and human menopausal gonadotropins. Fertil Steril 1987;48:811-815 2. Palermo R, Amodeo G, Navot D, Rosenwaks Z, Cittadini E: Concomitant gonadotropin-releasing hormone agonist and menotropin treatment for the synchronized induction of multiple follicles. Fertil Steril 1988;49:290-295 3. de Ziegler D, Cedars MI, Randle D, Lu JKH, Judd HL, Meldrum DR: Suppression of the ovary using a gonadotro-
9.
10.
11.
12.
13.
14.
pin releasing-hormone agonist prior to stimulation for oocyte retrieval. Fertil Steril 1987;48:807-810 Rutherford A J, Subak-Sharpe RJ, Dawson KJ, Margara RA, Franks S, Winston RML: Improvement of in vitro fertilisation after treatment with buserelin, an agonist of luteinizing hormone releasing hormone. Br Med J 1988;296:1765-1768 Frydman R, Belaisch-Allart J, Parneix I, Forman R, Hazout A, Testart J: Comparison between flare up and down regulation effects of luteinizing hormone-releasing hormone agonists in an in vitro fertilization program. Fertil Steril 1988; 50:471-475 Frydman R, Parneix 1, Belasich-Allart J, Forman R, Hazout A, Fernandez H, Testart J: LHRH agonists in IVF: Different methods of utilization and comparison with previous ovarian stimulation treatments. Hum Reprod 1988 ;3:559-561 Rainhorn JD, Forman RG, Belaisch-Allart J, Hazout A, Fries N, Testart J, Frydman R: One year's experience with programmed retrieval at IVF. Hum Reprod 1987;2:491-494 Neveu S, Hedon B, Bringer J, Chinchole JM, Arnal F, Humeau C, Cristol P, Viala JL: Ovarian stimulation by a combination of gonadotropin-releasing hormone agonist and gonadotropins for in vitro fertilization. Fertil Steril 1987; 47:639-643 Chetkowski RJ, Kruse LR, Nass TE: Improved pregnancy outcome with the addition of leuprolide acetate to gonadotropins for in vitro fertilization. Fertil Steril 1989;52:250--255 Macnamee MC, Howles CM, Edwards RG, Taylor PJ, Elder KT: Short-term luteinizing hormone-releasing hormone agonist treatment: Prospective trial of a novel ovarian stimulation regimen for in vitro fertilization. Fertil Steril 1989; 52:264-269 Hill GA, Freeman M, Bastias MC, Rogers BJ, Herbert CM, Osteen KG, Wentz AC: The influence of embryo quality on pregnancy rate in a program for in vitro fertilization/embryo transfer. Fertil Steril (in press) Diamond MP, Hill GA, Webster BW, Herbert CM, Rogers BJ, Osteen KG, Maxson WS, Vaughn WK, Wentz AC: Comparison of human menopausal gonadotropin, clomiphene citrate, and combined human menopausal gonadotropin-clomiphene citrate stimulation protocols for in vitro fertilization. Fertil Steril 1986;46:1108-1112 Stone BA, Serafini PC, Quinn P, Kerin JF, Marts RP: Gonadotropin and estradiol during ovarian stimulation in women treated with leuprolide acetate. Obstet Gynecol 1989 ;73:990 Meldrum D: GnRHa agonists as adjuncts for in vitro fertilization. Ohstet Gynecol Surv 1989;44:314
Journal o f in Vitro Fertilization and Embryo Transfer, Vol. 7, No. 6, 1990
The Use of Gonadotropin Releasing Hormone Agonist (GnRHa) in Good Responders Undergoing Repeat in Vitro Fertilization/Embryo Transfer (IVF/ET) 1 JAMES H. SEGARS, 2'3 GEORGE A. HILL, 2'4 STEPHANIE H. BRYAN, 2 CARL M. HERBERT, 11I,2 KEVIN G. OSTEEN, 2 B. JANE ROGERS, 2 and ANNE COLSTON WENTZ z'5
Submitted: January 29, 1990 Accepted: July 10, 1990
patients previously classified as "good responders" undergoing ovarian stimulation for IVF/ET.
The use of gonadotropin releasing hormone agonists (GnRHa) has been shown to improve the response in patients classified as "poor responders" undergoing ovarian stimulation for in vitro fertilization~embryo transfer (IVF/ ET). This study sought to determine whether GnRHa therapy would benefit patients undergoing IVF/ET who had been classified as "good responders" in prior attempts. Twenty-three patients who had completed a prior IVF/ET attempt but who failed to conceive underwent ovarian stimulation using a combination of GnRHa and human menopausal gonadotropin (hMG). Each patient's prior stimulation served as her control and consisted of clomiphene citrate (CC)/hMG in 18 patients and follicle stimulating hormone (FSH) and~or hMG in 5 patients. The numbers of oocytes retrieved, oocytes fertilized, embryos cleaved, and embryos transferred were all significantly greater in cycles treated with GnRHa/hMG compared to control cycles. The clinical pregnancy rate was 39% and the ongoing pregnancy rate was 26% during the cycle when GnRHa pretreatment was utilized. These data suggest that GnRHa therapy is of benefit even to those
KEY WORDS: gonadotropin releasing hormone (GnRH) ago-
nists; in vitro fertilization/embryotransfer; ovarian hyperstimulation. INTRODUCTION Despite recent technological improvements, success rates of in vitro fertilization/embryo transfer (IVF/ET) remain low. Cancellation of patients due to a poor ovarian response or premature luteinizing hormone (LH) surges accounts for 25 to 30% of patient attrition. Use of gonadotropin releasing hormone agonists (GnRHa) has been shown to improve the response in patients undergoing IVF/ET who have been classified as poor responders (1-4). Reports have also suggested that routine use of GnRHa for all patients undergoing IVF/ET may improve results by decreasing cancellations due to premature LH surges (5-8). Chetkowski et al. (9) prospectively compared the utilization of human menopausal gonadotropins (hMG) with and without the addition Of GnRHa in "unselected patients." These investigators were able to document that the use of the GnRHa with hMG resulted in an improvement in the fertilization and implantation rates and a decrease in the spontaneous abortion rate in patients undergoing IVF/ET. Furthermore, Macnamee et al. (10) recently demonstrated in a large prospective trial that short-term GnRHa treatment with hMG resulted in improvement when compared to a similar group of patients treated with a combination of clomiphene citrate (CC)/hMG. The ongoing pregnancy rate per treatment cycle was 27% in
i Presented at the 45th Annual Meeting of the American Fertility Society, November 1989, San Francisco, California. 2 Center for Fertility and Reproductive Research (C-FARR), Department of Obstetrics and Gynecology, Vanderbilt University Medical Center, Nashville, Tennessee 37232. 3 Present address: Building 6, Room 3A-15, Laboratory of Developmental Molecular Immunity, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland 20892. 4 To whom correspondence should be addressed at Center for Fertility and Reproductive Research, Department of Obstetrics and Gynecology, Room D-3223, Vanderbilt University Medical Center North, Nashville, Tennessee 37232. 5 Present address: Northwestern Medical Faculty Foundation, Inc., Prentice Pavilion, 333 East Superior Street, Suite 1165, Chicago, Illinois 60611.
327
0740-7769/90/1200-0327506.00/0 9 1990HenumPublishingCorporation
328
SEGARS ET AL.
the patients treated with the GnRHa/hMG protocol, compared to 14% for the CC/hMG protocol. Both of these studies (9,10) document an improvement in cycle outcome between groups of patients when GnRHa is utilized in the ovarian stimulation protocol. However, it remains unclear whether an individual patient classified as having a good response to ovarian stimulation would benefit by the addition of GnRHa to the stimulation regimen. Most studies addressing this question have employed a between groups analysis and have not specifically focused on "good responders." It may be that the improvement observed when all patients (i.e., groups of patients) are pretreated with GnRHa may be due solely to an improvement in the patients who are poor responders. This is an important point, for if both "poor" and "good responders" improve with adjunctive GnRHa treatment, then use of GnRHa would appear indicated for virtually all patients attempting IVF. For these reasons, we sought to determine if the adjunctive use of GnRHa (leuprolide acetate; LA) in patients previously classifted as having a good response during an IVF/ET cycle would result in an improved outcome during the GnRHa-treated cycle. The design differs from prior studies of GnRHa usage in normal responders in that responses of individual patients (not groups) were compared, thus minimizing interindividual variability.
MATERIALS AND METHODS Twenty-three patients, ages 26 to 42 years [33.9 - 4.4 (mean +-- standard deviation)], admitted to the program for IVF/ET at Vanderbilt University Medical Center in Nashville, Tennessee, between October 1988 and February 1989 were evaluated. Eighteen patients (Group A) had previously undergone ovarian stimulation with CC (Serophene, Serono Laboratories, Inc., Randolph, M A ) / h M G (Pergohal, Serono Laboratories, Randolph, MA). Additional patients treated with follicle stimulating hormone (FSH; Metrodin, Serono Laboratories, Inc., Randolph, MA; n = 2) or hMG (n = 3) [Group B; n = 5] were included. All had successfully progressed through the previous cycle, had oocytes recovered, achieved fertilization, and underwent embryo transfer. No control cycle had been cancelled because of a poor estradiol (E2) response or a premature L H surge. By all criteria, these patients had
demonstrated a good response in the previous cycle; however, none had conceived. The 23 participating patients were begun on LA (Tap Pharmaceuticals, North Chicago, IL), 1 mg per day subcutaneously (SQ) beginning on cycle day 1. After 7 to 9 days of treatment, serum estradiol (Ez) was measured. If the serum E 2 level was less than 30 pg/ml, hMG was begun at 3 ampoules per day and the LA was decreased to 0.5 mg per day SQ. If the serum E2 was greater than 30 pg/ml, LA was continued at 1 mg/day SQ for 2 to 3 days and an E 2 and ultrasonogram (USG) were performed. If the E z remained greater than 30 pg/ml but no USG documentation of cysts was present, the LA dosage was changed to 0.5 mg SQ twice a day. Once suppression was documented by an E2 of less than 30 pg/ml, hMG was begun and the patients were followed with E2 and USG beginning on day 5. LA was continued at 0.5 mg per day SQ until the day of administration of human chorionic gonadotropin (hCG). Thirty-five hours following hCG, the patients underwent retrieval of oocytes with transvaginal USG guidance. Two patients in Group A were excluded for reasons not related to ovarian stimulation. The techniques of oocyte and embryo culture have been previously published (11). At the time of retrieval, oocytes were graded on a scale of 1 to 5 as previously described (11). Briefly, an oocyte graded as 1 was considered immature, 3 was considered mature preovulatory, and 5 was considered atretic. Grades 2 and 4 were considered slightly immature and slightly overmature, respectively, but were treated as mature oocytes. Rates of mature oocyte recovery, fertilization, embryo cleavage, and embryo transfer were compared in control (CC/hMG, FSH, hMG) and treated (GnRHa/hMG) cycles. Pregnancy rates are reported but not compared statistically, as all patients by design of the study had not conceived during a recent attempt at IVF/ET. As a further control for the possibility that the study patients had an initial poor CC/hMG (control) stimulation, the 16 patients in Group A were compared to a matched control group (Group C) of 16 patients undergoing CC/hMG stimulation in the current IVF/ET cycle. This additional control was an attempt to address the question of whether the previous (control) stimulation of the study patients was indicative of a good response on CC/hMG and to make sure that other improvements in the IVF/ET program over time were not responsible for the differences between the groups.
Journal o f in Vitro Fertilization and Embryo Transfer, VoI. 7, No. 6, 1990
329
GnRHa IN GOOD RESPONDERS UNDERGOING IVF/ET
Statistical Analysis Data are expressed as mean -+ standard deviation (SD). Analysis of data was performed for the study patients with a paired t test with the alpha level set at 0.05. Differences between control cycles in CC/ hMG-treated patients (Group A) and cycles in the second control group (Group C) were determined with Student's t test.
DISCUSSION
RESULTS In the study group, levels of estradiol at the time of hCG were not significantly different when compared in control and treated cycles. Control and treated cycles varied substantially, however, in the amounts of hMG administered and the number of days of hMG therapy (Table I). The rate of mature oocyte recovery (mean -+ SD) for all study patients in the GnRHa/hMG cycle was significantly greater than in control cycles (Table II). Perhaps as a result of the greater number of oocytes recovered at the time of surgery, fertilization, cleavage, and embryo transfer rates were also significantly greater in GnRHa/hMG vs control cycles in study patients. It did not appear that treatment with GnRHa resulted in any inhibition of embryo development in vitro. The difference between GnRHa/hMG and control cycles in the study group ultimately resulted in a greater number of embryos transferred in GnRHa/hMG-treated patients. Crude pregnancy rates are reported in Table II. Comparison of the pretreatment responses of the 16 CC/hMG patients (Group A) to an additional 16 patients undergoing recent CC/hMG stimulation failed to reveal any significant difference among the numbers of oocytes retrieved, oocytes fertilized, Table I. Characteristics of Ovarian Stimulation in Patients with and Without G n R H a Treatment a Study group (n = 21) Prior cycle Age E2 at the time ofhCG Ampoules o f h M G Days of stimulation
embryos cleaved, or embryos transferred (Table II). The 16 patients having a prior CC/hMG stimulation (Group A) exhibited significant differences between control and GnRHa-treated cycles when examined (Table II). Differences between former stimulated cycles and GnRHa/hMG cycles were not significant in Group B, perhaps due to the small group size (n = 5) (data not shown).
Group C, (n = 16) CC/hMG cycle GnRHa/hMG matched cycle control
33.3 -+ 4.6
--
33.4 -+ 3.6
1184 -+ 578 11.3 --+ 5.8
1351 - 472 25.6 --- 5.1"
1364 + 486 8.5 - 2.7
8.2 -+ 1.I
10.1 + 1.6"
8.5 -+ 1.2
a Data expressed as mean __- standard deviation. * P < 0.05 compared to prior stimulation.
Investigators continue to search for IVF/ET stimulation protocols which will maximize oocyte recovery and ultimately improve pregnancy rates. The CC/hMG regimen has proven superior in terms of oocytes recovered, fertilization and cleavage rates, and numbers of embryos transferred. Unfortunately, this has not translated into improved pregnancy rates (12). Several studies have compared the use of GnRHa to the standard management of patients undergoing ovarian stimulation in poor responders. These patients generally had exhibited a poor E 2 response to hMG stimulation or had manifested a premature L H surge in prior cycles. The data in these reports suggest that this group of patients clearly benefits from GnRHa pretreatment (2,4,8). Neveu e t al. (8) have investigated the adjunctive use of GnRHa treatment in a group of normal patients attempting IVF/ET. Their study compared 10 patients being treated with FSH alone to I0 patients stimulated with G n R H a and FSH. The group treated with GnRHa had a statistically greater number of oocytes retrieved and number of embryos/ oocyte (fertilization rate), in contrast to the studies by Neveu e t al. (8) and Macnamee e t al. (10), we find that the greater number of oocytes retrieved, greater fertilization rate, and greater cleavage rate resulted in an increased number of embryos available for transfer. This may be a significant finding, since the number of embryos transferred correlates with the rate of pregnancy. Our data confirm the data of Chetkowski e t al. (9) but extend their findings and demonstrate that a preselected group of "good responders" will benefit from adjunctive short-term use of GnRHa. Frydman e t al. (6) also reported the efficacy of adjunctive GnRHa in IVF/ET patients with previous normal responses. The study by Frydman e t al. (6) employed a retrospective design with a comparison to a prior CC/hMG stimulation. In the normal
Journal o f in Vitro Fertilization and Embryo Transfer, Vol. 7, No. 6, 1990
SEGARS ET AL.
330
Table II. Results of Oocyte Recovery in Patients With and W i t h o u t G n R H a T h e r a p y a
Prior stim.
GnRHa/hMG
CC/hMG
GnRHa/hMG
Group C (matched controls for CC/hMG) (n = 16), CC/hMG
5.6 --- 2,9 3.7 +- 1.9
8.2 - 3.9* 4.9 -+ 2.5*
5.6 -+ 2.8 3.8 -+ 2.1
8.5 -+ 4.1" 5.3 +- 2.6*
5.2 + 2.8 3.2 -+ 2.4
2.9 -+ 1.4 2.6 -+ 1.1 0b
4.3 +- 2.2* 3.4 -+ 1.1' 9/21
2.7 --- 1.4 2.5 - 1.2 0b
4.7 +- 2.2* 3,4 +- 1.1" 6/16
3.0 +- 2.2 2.7 -+ 1.5 3/16
Group A (prior C C / H M G ) (n = 16)
All study patients (n = 21)
Mature oocytes recovered Oocytes fertilized E m b r y o s undergoing cleavage E m b r y o s transferred Pregnancies
a Data e x p r e s s e d as m e a n -+ standard deviation. b By study design, patients had not conceived in prior IVF/ET attempts. * P < 0.05 compared to prior stimulation.
ovulatory patients undergoing IVF/ET, the ongoing pregnancy rate increased from 10.4% in prior cycles to 22.4% in the agonist-treated group. No data were presented to determine whether this difference was due to an increase in the number of oocytes recovered or if the changes in pregnancy rate could have been due to other factors. The observation that the use of GnRHa for all patients undergoing stimulation results in an improved outcome does not satisfactorily address the question of whether GnRHa should be given to good (normal) responding patients. That is, the improvement in the total group's response (all patients) may be due solely to an improvement among the poor responders. To address the question of whether adjunctive use of GnRHa is of benefit to "good responders," we evaluated a group of patients who had a satisfactory previous response to ovarian stimulation in an IVF attempt but who did not conceive. This design permitted a paired comparison of the prior stimulation to the current GnRHa/hMG cycle. The design minimized the interindividual variability in the response to ovarian stimulation. Our results suggest that GnRHa/hMG therapy is of benefit to patients who have previously progressed to surgery and embryo transfer and could be classified as good responders on a CC/hMG regimen. A significantly greater number of mature oocytes was retrieved, which resulted in a greater number of embryos available for transfer. The ability of this study to detect a difference in the number of embryos transferred (vs earlier studies) is likely due to the comparison of individual patients and not groups of patients. Because this study compared the response to a clomiphene/hMG stimulation with a GnRHa/hMG regimen, it is possible that the observed differences may be due to other fac-
tors (i.e., clomiphene or hMG dosage) and not GnRHa use alone. Other studies comparing GnRHa/ hMG to hMG stimulations have noted similar directional trends (13,14), therefore it would not seem that the presence of clomiphene in the control stimulations would be entirely responsible for the improvements observed. The clinical pregnancy rate of all patients treated with GnRHa was 39% (9 of 23). There have been two spontaneous abortions, one ectopic pregnancy, and six ongoing pregnancies. Unfortunately, we cannot make a statistical statement regarding pregnancy rates, since by design the patients chosen for study had not conceived in their prior cycle. There were six pregnancies in the GnRHa Group A (6 of 16; 37%) and three pregnancies in the matched controis (Group C; 18%). Interestingly, the percentage of pregnancies in the matched control group is consistent with that of our center for a CC/hMG stimulation (18%). Further studies will focus on this issue in an attempt to determine whether the observed improvements in outcome translate to an increased pregnancy rate in good responders. As an additional control, we evaluated a group of 16 patients (Group C) undergoing CC/hMG stimulation in the same cycle as the study group receiving GnRHa/hMG therapy (Group A). Comparison of these 16 patients to 16 study patients' CC/hMG stimulation revealed that there was no significant difference in the rates of oocyte recovery, fertilization, cleavage, and embryo transfer in the matched control group (Group C) as compared to the paired control group (Group A) of study patients. Thus, it appears that we had chosen a study group that was representative of patients undergoing stimulation with CC/hMG and not a group of poor responders. These data reveal that in those patients known to
Journal o f in Vitro Fertilization and Embryo Transfer, Vol. 7, No. 6, 1990
GnRHa IN GOOD RESPONDERS UNDERGOING IVF/ET
be good responders during a CC/hMG stimulation, adjunctive treatment with GnRHa significantly improves the rates of oocyte recovery, oocyte fertilization, embryo cleavage, and number of embryos transferred. The study suggests that the improvement observed in ovarian stimulation when GnRHa therapy is utilized to treat all patients is due not only to improvement in patients who demonstrate a poor response to ovarian stimulation regimens, but also to improvement in patients who have demonstrated prior good responses. Studies are ongoing to determine whether the observed increase in number of embryos transferred results in a significant improvement in pregnancy outcome.
331
4.
5.
6.
7.
8.
ACKNOWLEDGMENTS
The authors wish to acknowledge the excellent secretarial assistance of Ruth Robertson. This study was supported in part by a materials grant from TAP Pharmaceuticals, North Chicago, Illinois. The authors acknowledge the superb technical assistance of Christina Bastias, M.D., Peter Fetterolf, Ph.D., Melanie Freeman, M.S., and Thomas Thompson, Research Assistant.
REFERENCES 1. Awadalla SG, Friedman CI, Chin NW, Dodds W, Park JM, Kim MH: Follicular stimulation for in vitro fertilization using pituitary suppression and human menopausal gonadotropins. Fertil Steril 1987;48:811-815 2. Palermo R, Amodeo G, Navot D, Rosenwaks Z, Cittadini E: Concomitant gonadotropin-releasing hormone agonist and menotropin treatment for the synchronized induction of multiple follicles. Fertil Steril 1988;49:290-295 3. de Ziegler D, Cedars MI, Randle D, Lu JKH, Judd HL, Meldrum DR: Suppression of the ovary using a gonadotro-
9.
10.
11.
12.
13.
14.
pin releasing-hormone agonist prior to stimulation for oocyte retrieval. Fertil Steril 1987;48:807-810 Rutherford A J, Subak-Sharpe RJ, Dawson KJ, Margara RA, Franks S, Winston RML: Improvement of in vitro fertilisation after treatment with buserelin, an agonist of luteinizing hormone releasing hormone. Br Med J 1988;296:1765-1768 Frydman R, Belaisch-Allart J, Parneix I, Forman R, Hazout A, Testart J: Comparison between flare up and down regulation effects of luteinizing hormone-releasing hormone agonists in an in vitro fertilization program. Fertil Steril 1988; 50:471-475 Frydman R, Parneix 1, Belasich-Allart J, Forman R, Hazout A, Fernandez H, Testart J: LHRH agonists in IVF: Different methods of utilization and comparison with previous ovarian stimulation treatments. Hum Reprod 1988 ;3:559-561 Rainhorn JD, Forman RG, Belaisch-Allart J, Hazout A, Fries N, Testart J, Frydman R: One year's experience with programmed retrieval at IVF. Hum Reprod 1987;2:491-494 Neveu S, Hedon B, Bringer J, Chinchole JM, Arnal F, Humeau C, Cristol P, Viala JL: Ovarian stimulation by a combination of gonadotropin-releasing hormone agonist and gonadotropins for in vitro fertilization. Fertil Steril 1987; 47:639-643 Chetkowski RJ, Kruse LR, Nass TE: Improved pregnancy outcome with the addition of leuprolide acetate to gonadotropins for in vitro fertilization. Fertil Steril 1989;52:250--255 Macnamee MC, Howles CM, Edwards RG, Taylor PJ, Elder KT: Short-term luteinizing hormone-releasing hormone agonist treatment: Prospective trial of a novel ovarian stimulation regimen for in vitro fertilization. Fertil Steril 1989; 52:264-269 Hill GA, Freeman M, Bastias MC, Rogers BJ, Herbert CM, Osteen KG, Wentz AC: The influence of embryo quality on pregnancy rate in a program for in vitro fertilization/embryo transfer. Fertil Steril (in press) Diamond MP, Hill GA, Webster BW, Herbert CM, Rogers BJ, Osteen KG, Maxson WS, Vaughn WK, Wentz AC: Comparison of human menopausal gonadotropin, clomiphene citrate, and combined human menopausal gonadotropin-clomiphene citrate stimulation protocols for in vitro fertilization. Fertil Steril 1986;46:1108-1112 Stone BA, Serafini PC, Quinn P, Kerin JF, Marts RP: Gonadotropin and estradiol during ovarian stimulation in women treated with leuprolide acetate. Obstet Gynecol 1989 ;73:990 Meldrum D: GnRHa agonists as adjuncts for in vitro fertilization. Ohstet Gynecol Surv 1989;44:314
Journal o f in Vitro Fertilization and Embryo Transfer, Vol. 7, No. 6, 1990
