Vol. 56, No.1, July 1991
FERTILITY AND STERILITY
Printed on acid-free paper in U.S.A.
Copyright © 1991 The American Fertility Society
The Lupron* screening test: tailoring the use of leuprolide acetate in ovarian stimulation for in vitro fertilizationt
Santiago L. Padilla, M.D. t Rosella D. Smith, M.D. Jairo E. Garcia, M.D. Women's Hospital Fertility Center, Department of Gynecology, Greater Baltimore Medical Center, Baltimore, Maryland
Study Objective: To evaluate the prognostic and therapeutic value of a Lupron (leuprolide acetate; Tap Pharmaceuticals, North Chicago, IL) screening test before ovarian stimulation for in vitro fertilization (IVF). Design: Prospective. Setting: Outpatient IVF program. Patients: Eighty patients exhibited four early estradiol (E 2) patterns. Patients with pattern A and B remained on a flare-up ovarian stimulation protocol. Patients with pattern C were randomized to three ovarian stimulation protocols. Patients with pattern D were treated with the flare-up protocol using a high pure follicle-stimulating hormone (FSH) dose. Results: Patterns A, B, C, andD occured in 44%,16%,25%, and 15% of the patients, respectively. The E2 pattern recurred in 77% of subsequent IVF cycles. Pattern A and B patients achieved a 41 % (23/56) and 22% (5/23) ongoing pregnancy rate (PR) per stimulated cycle. An early luteal phase Lupron protocol had the best ongoing PR per stimulated cycle (10/27, 37%) in patients with a pattern C response. Pattern D patients had a 20% (5/25) ongoing PR per stimulated cycle. Conclusion: The Lupron screening test allows prospective selection of stimulation protocols in ovulatory patients undergoing IVF. Early E2 patterns A and B should be treated with the flare-up protocol. Pattern C patients benefit from the luteal phase Lupron protocol and pattern D patients benefit from a high pure FSH flare-up protocol. Fertil Steril 56:79, 1991
We have reported that the early serum estradiol (E 2 ) response to Lupron (leuprolide acetate; Tap Pharmaceuticals, North Chicago, IL) during a flareup ovarian stimulation protocol for in vitro fertilization (IVF) has prognostic value. 1 A retrospective analysis revealed four distinct early E2 response patterns. Pattern A showed a prompt E2 elevation by menstrual cycle day 3, followed by a fall by cycle day 4. Pattern B showed a delayed E2 elevation by
Received December 3, 1990; revised and accepted March 12, 1991. * Tap Pharmaceuticals, North Chicago, Illinois. t Presented at the 46th Annual Meeting of The American Fertility Society, Washington, D.C., October 13 to 18, 1990. :j: Reprint requests: Santiago L. Padilla, M.D., Greater Baltimore Medical Center Physician's Pavilion, 6565 North Charles Street, Suite 207, Baltimore, Maryland 21204. Vol. 56, No.1, July 1991
cycle day 4, followed by a fall by cycle day 5 or 6. Pattern C showed persistent E2 elevation through cycle day 5. Pattern D lacked a significant early E2 response. The ongoing live-birth pregnancy rates (PRs) per stimulated cycles were extremely favorable for patterns A and B and significantly lower for patterns C and D. Pattern A is not only associated with a better PR but also with a significantly higher embryo implantation rate. 1 The vast majority of the multiple pregnancies occurred in pattern A patients. Before the current study, we did a retrospective analysis to evaluate the reproducibility of these E2 patterns. We studied 77 patients who underwent consecutive IVF attempts and found that the patterns A, B, C, and D recurred in 81 %, 62%, 76%, and 80% of subsequent IVF cycles, respectively. 2 The purpose of this prospective study is to evaluate the Padilla et al.
Lupron screening test for IVF
79
f
clinical value of a Lupron screening test before ovarian stimulation in IVF and to analyze different stimulation protocols for the patients who carried a poor prognosis in our original flare-up protocol, i.e., the patients who exhibited a pattern C or D response. MATERIALS AND METHODS
The study population consisted of infertility patients with tubal factor, endometriosis, or unexplained infertility referred for IVF from January 1 to December 1, 1989. Patients with male factor infertility were excluded to avoid variability secondary to sperm factors. All patients had regular menstrual cycles, a normal endometrial cavity documented by hysterography or hysteroscopy, a normal endometrial biopsy, and a normal semen analysis. The Lupron screening test was done by administering Lupron 0.75 mg subcutaneously (SC) for patients < 75 kg and 1 mg SC for patients 2': 75 kg. Injections were given every morning on cycle days 2, 3, and 4. A serum E2 level was obtained on cycle days 2, 3, 4, and 5. A serum follicle-stimulating hormone (FSH) level was obtained on cycle day 2. All daily serum samples were drawn before the Lupron injections. The serum was refrigerated at 4°C, and all samples from the same patient were run in the same assay. Patients with a serum E2 pattern A or B response underwent ovarian stimulation with a flare-up protocol. Briefly, Lupron was started with the identical dose as the Lupron screening test dose. Gonadotropin therapy was initiated on cycle day 5 with 75 IU of pure FSH (Metrodin; Serono Inc, Randolph, MA) and one ampule of human menopausal gonadotropin (hMG, Pergonal; Serono Inc, Randolph, MA) containing 75 IU of FSH and 75 IU of luteinizing hormone (LH) given intramuscularly (1M) at 8 A.M. and the same dose of pure FSH and hMG at 3 P.M. The dose was then adjusted in a step-down manner based on the patient's response as described earlier. 3 Patients with E2 pattern C were randomized prospectively by a randomization table to one of three protocols. Protocol 1 consisted of a combination protocol that started with 150 IU of pure FSH and two ampules of hMG on cycle days 3 and 4, followed by two ampules of hMG from cycle day 5 on. Protocol 2 was the flare-up protocol, but the Lupron dose was 1.5 mg SC for patients < 75 kg and 2 mg SC for patients 2': 75 kg. Protocol 3 was an early luteal phase Lupron protocol. Patients were monitored every 2 to 3 days starting on cycle day 15 with serum progesterone (P) levels. Lupron was started 80
Padilla et al.
Lupron screening test for IVF
when ovulation was detected, as evident by a serum P > 3 ng/mL. Patients continued the Lupron until menses. If a serum E2 was 250 pg/mL, ultrasound (US) follicular monitoring was started. Patients on the FSH/hMG (protocol 1) had daily LH monitoring starting with US monitoring. Ovarian stimulation was continued until two follicles were at least 16 mm in size for all Lupron protocols and 15 mm in size for the pure FSH/hMG group without Lupron. That evening 10,000 U of hCG were injected 1M between 9 P.M. and 1 A.M. A transvaginal US-guided oocyte retrieval was scheduled 35 hours after the hCG injection. A maximum of five oocytes were inseminated with 150 X 103 washed motile sperm/mL of insemination media 3 to 26 hours after aspiration, depending on the degree of maturation. If cryopreservation was elected, up to 16 oocytes were inseminated, and a maximum of four embryos were transferred fresh. The embryo laboratory details have been described earlier. 4 Briefly, insemination media consisted of Ham's F-10 medium (Gibco, Grand Island, NY) with 7.5% heated-inactivated fetal cord serum. On day 1 after insemination, the cumulus was removed mechanically to look for signs of fertilization. If fertilization did not occur, the oocytes were reinseminated. Fertilized oocytes with three or more pronuclei were discarded. After fertilization, the embryos were transferred to growth media that consisted of Ham's F-lO with 15% fetal cord serum. Oocytes and embryos were incubated in 5% CO 2 in air at 37°C. Embryo transfer (ET) was done 72 hours after oocyte retrieval. All patients who underwent ET received supplemental P throughout the luteal phase. If pregnancy was established, the patient was given 250 mg of 17hydroxyprogesterone caproate 1M every week until the 12th week of gestation. Transvaginal US was done 3 to 4 weeks after ET for determination of Fertility and Sterility
fetal heart activity and detection of multiple pregnancies. Clinical pregnancies were established when two serum {3-hCG titers, drawn at least 10 days after ET, were rising, with US and/or histologic evidence of pregnancy. An ongoing pregnancy was defined as an intrauterine pregnancy of at least 14 weeks of gestation with presence of fetal heart activity. Serum E2 and FSH were measured in duplicate. Serum E2 was assayed with a commercial doubleantibody radioimmunoassay (RIA) kit available from Diagnostic Products Inc, (Los Angeles, California). Estradiol (1251 labeled) was used, and six E2 standards were run with concentrations between 20 and 3,600 pg/mL. Interassay coefficient of variation (CV) was between 4.6% and 13.7% depending on the level of hormone measured. The intra-assay CV depended on the level of hormone and ranged between 2.8% and 11.4%. Both of the high CVs were associated with the low values of E 2. The E2 antiserum was highly specific. The highest cross-reactivity to other natural steroids is with estrone (2.4%) and estriol (0.5%). Serum FSH was measured with a double-antibody RIA from Amersham Corporation (Arlington Heights, IL). Follicle-stimulating hormone 25 I labeled) and six standards between 0 and 145 mIU/mL were used for the assay. Intra-assay CV was between 3.1 % and 5.2%, and interassay CV was between 6% and 10.7%. The cross-reactivity of the FSH antiserum was 0.9% with thyroid-stimulating hormone, 0.08% with LH, and 0.06% with hCG. All samples were repeated when the intra-assay CV between duplicates was> 10%. Statistical evaluation was performed with Fisher's exact test.
e
RESULTS Eighty patients had the Lupron screening test. These patients underwent a total of 153 IVF cycles
Table 1 Frequency of the Early Serum E2 Pattern During the Lupron Screening Test and Reproducibility in the Subsequent Flare-Up Lupron Ovarian Stimulation for IVF Subsequent IVF cycles pattern Lupron screening pattern A B C D a b
(n = 35) (n = 13) (n = 20) (n = 12)
A 28 (80)a 3 (22)
NAb
B
C
D
3 (8)
1 (3) 1 (8)
3 (8) 1 (8)
NA
NA 10 (83)
8 (62) NA 1 (8)
Values in parentheses are percents. NA, not applicable.
Vol. 56, No.1, July 1991
1 (8)
Table 2 Ongoing PRs per Stimulated Cycle With Different Stimulation Protocols in Patients With Pattern C Response at the Time of the Lupron Screening Test
Stimulation protocol Phase 1 (n = 34) FSH/hMG (no Lupron) Double Lupron Luteal phase Lupron Phase 2 (n = 15) Luteal phase Lupron
Clinical PR/cycle
Ongoing PR/cycle
2/12 (17)a 2/10 (2W 6/12 (50)'
2/12 (27) 1/10 (10) 5/12 (42)
5/15 (33)
5/15 (33)
Values in parentheses are percents. Includes one spontaneous abortion. , Includes one tubal pregnancy. a
b
subsequent to the Lupron screening test. All patients underwent at least one IVF cycle. Mean age ± 1 SD of the group was 35 ± 3.7 years (range 28 to 40). Indications for IVF were tubal factor (55%), endometriosis (38%), and unexplained infertility (7%). Male factor infertility couples were excluded from the study. Table 1 describes the outcome of the Lupron screening test results and the pattern exhibited at the time of the subsequent ovarian stimulation for IVF. Only the first IVF cycles immediately after Lupron screening test were included in this table. The most common pattern was pattern A, followed by pattern C. The early E2 pattern is very reproducible, and the majority of the patients showed the same pattern in subsequent IVF attempts. Subsequent patterns were not included for patients with pattern C because they underwent different stimulation protocols. Table 2 summarizes the results from our pattern C treatment protocols. Phase 1 was the prospective, randomized study. Although there was no statistical difference between the three protocols, there was a clear trend toward a higher ongoing PR in the patients in the luteal phase Lupron protocol. In addition, there was one miscarriage in the double Lupron protocol and one ectopic pregnancy in the luteal phase Lupron protocol. We elected to stop the randomization and treat the next 15 consecutive patients with the pattern C response using the luteal phase Lupron protocol in phase 2 of the study. Five patients conceived in phase 2 and have ongoing pregnancies. We used the results of our first series1 of the flareup protocol (series I) as "classic" controls to evaluate the success rate of each pattern with the different protocols (Fig. 1). Pattern A patients had very similar ongoing PRs per stimulated cycle using the flareup protocol. Pattern B showed a lower, but not staPadilla et al.
Lupron screening test for IVF
81
1 ! I
Estradiol
LU~
Pattern
A
Series I
Series II
Ra~~?8~A\e
Rate Cycle
16/41 (39%)
~
B
~-
C
2/19 (11%)
D
1/18 (6%)
2
6/16 (38%)
On~Oing
23/56 (41%) 5/23 (22%) 10/27
(37%)•
5/25 (20%)
3 4 5 Cycle Day
Figure 1 Ongoing PR per stimulated cycle of the current series (series II) compared with our previous series (series I) using the flare-up protocol. The pattern C patients had a significantly higher ongoing PR per cycle with the luteal phase Lupron protocol used in series II. *, P = 0.04.
tistically significant, ongoing PR per stimulated cycle with the flare-up protocol. We compared the PR of the patients with the pattern C response in our previous series (series I) using the flare-up protocol to the PR obtained in the luteal phase Lupron protocol ofthe current study (series II). The 27 patients in the luteal phase Lupron protocol had a significantly higher ongoing PR per stimulated cycle (11 % versus 37%, P = 0.04) when compared with the flareup protocol in series I. In patients with pattern D response, we found a higher ongoing PR when the flare-up protocol was used with high pure FSH doses, but the difference did not reach statistical significance (6% versus 20%, P = 0.11). In the current study, we had five cancellations ofthe 153 cycles (3%), and they all occurred in the pattern D patients. We analyzed the multiple PR and found that 12 of the 15 multiple pregnancies were in the patients with the pattern A response with the flare-up protocol. Of the 23 ongoing pregnancies that occurred on this pattern, there were 9 (39%) twins, 2 (9%) triplets, and 1 (4 %) quadruplet pregnancies. There was one set of twins in the pattern B response with the flare-up protocol, one set of triplets in the pure FSH/hMG protocol, and one set of triplets in the luteal phase Lupron protocol. Seventeen patients (21%) had a baseline serum FSH level;;::: 20 mIU/mL on cycle day 2 of the Lupron screening test and underwent 30 stimulated cycles. One patient (6%) had a pattern A response, one (6%) had a pattern B response, 8 (47%) had a pattern C response, and 7 (41%) had a pattern D response. Four ongoing pregnancies occurred in this group for an ongoing PR of 13% per stimulated cycle and 24% per patient. Three ofthe patients who con82
Padilla et al.
Lupron screening test for IVF
ceived had pattern C response in the Lupron screening test, and the fourth patient had a pattern D response. In contrast, 63 patients (79%) with a baseline serum FSH level < 20 mIU /mL on cycle day 2 of the Lupron screening test underwent 152 stimulated cycles. Thirty-four patients (54%) had a pattern A response, 12 (19%) had a pattern B response, 12 (19%) had a pattern C, and 5 (8%) had a pattern D response. Forty-two ongoing pregnancies occurred in this group. The ongoing PR per stimulated cycle (28%) was higher than in the high baseline FSH group, although not statistically different. The ongoing PR per patient (67%) was significantly higher in the patients with a low baseline FSH than in the high baseline FSH group, P < 0.005. DISCUSSION
Our data show that a Lupron screening test can be performed before IVF and will assist in the selection of stimulation protocols. The early serum E2 response to Lupron is reproducible in the majority of subsequent IVF attempts. Patients with pattern A response have an excellent prognosis for ovarian stimulation with the flare-up protocol. Patients with pattern B response may have a slightly lower, but still very good, ongoing PR with the flare-up protocol. Our previous series showed that the flare-up protocol was not optimal for patients with pattern C response.! The patients who show evidence of ovarian suppression by cycle day 5, i.e., patients with pattern A or B, have an excellent prognosis, which contrasts with pattern C patients who do not have the early suppression on the flare-up protocol. Although the study had small numbers to reach statistical significance, we found no improvement using a double dose of Lupron with the flare-up protocol nor using a pure FSH/hMG protocol without Lupron. An early luteal phase Lupron protocol achieves ovarian suppression before the administration of gonadotropins, which appears to be important for Lupron protocols. In this study, the pattern C patients achieved a significantly higher PR using the luteal phase Lupron protocol when compared with the flare-up protocol results from our previous series. The patients with pattern D response have the lowest E2 levels, so we doubled the dose of pure FSH to four ampules per day. We observed a higher PR with this modification but no improvement on the cancellation rate when compared with our previous series. Bernardus et al. 5 reported an improvement in oocyte recovery rates and PRs with the use of a Fertility and Sterility
I I !
! l
higher pure FSH dose to improve the exogenous FSH/LH ratio. Hofmann et al. 6 found that high doses of pure FSH (6 ampules per day) reduced cancellation rates and improved PRs in low responders. Another benefit of the Lupron screening test is that it identifies the patients who are at a higher risk for multiple pregnancy. The vast majority of the multiple pregnancies in this study occurred in patients with pattern A response. These patients should have fewer embryos transferred. Patients with pattern A response are the prime candidates for embryo cryopreservation to reduce the multipIe PRo Our current recommendation is to perform a Lupron screening test on every ovulatory patient undergoing IVF. If the patient has a favorable response, i.e., patterns A or B, she can continue the flare-up protocol and do IVF in the same cycle of the Lupron screening test. Patients with pattern C response are advised to wait until ovulation occurs and switch to the luteal phase Lupron protocol. If the patient is from out of town, the Lupron screening test can be arranged very easily with the patient's local physician. Patients with a pattern C response can be started on the luteal phase Lupron protocol by their local physician and come to our center to start the ovarian stimulation after onset of menses. Patients with pattern D are treated with a flare-up protocol and high pure FSH doses on the same cycle of the Lupron screening test. Muasher et al. 7 and Toner et al. 8 have reported that elevated FSH levels on cycle day 3 are associated with a poor response to IVF. Our data also showed a lower ongoing PR per stimulated cycle and per patient with a baseline serum FSH ~ 20 mID /mL on cycle day 2. The baseline serum FSH level on cycle day 2 should be used as part of the Lupron screening test to counsel patients on their prognosis for success in IVF. In summary, the Lupron screening test is a simple and reproducible test that allows selection of individualized ovarian stimulation protocols for IVF.
Vol. 56, No.1, July 1991
The majority of IVF patients benefit from a flareup Lupron protocol, with the exception of the patients with a pattern C response who benefit from the luteal phase Lupron protocol. Acknowledgments. We are especially grateful to The Greater Baltimore Medical Center's IVF Laboratory for their help and support; Tap Pharmaceuticals, Chicago, Illinois, for the Lupron used for the study; Stephen Noel, Ph.D, Department of Pathology, The Greater Baltimore Medical Center for his help with the steroid assays; Theodore Baramki, M.D., Department of Gynecology, The Greater Baltimore Medical Center for his editorial guidance, and Mrs. Verna Young, IVF Coordinator, for typing this manuscript.
REFERENCES 1. Padilla, SL, Bayati J, Garcia JE: Prognostic value of the
early serum estradiol response to leuprolide acetate in in vitro fertilization. Fertil Steril 53:288, 1990 2. Padilla SL, Garcia JE: Unpublished data 3. Garcia JE, Padilla SL, Bayati J, Baramki TA: Follicular phase gonadotropin-releasing hormone agonist and human gonadotropins: a better alternative for ovulation induction in in vitro fertilization. Fertil Steril 53:302, 1990 4. Padilla SL, Garcia JE: Effect of maternal age and number of in vitro fertilization procedures on pregnancy outcome. Fertil Steril 52:270, 1989 5. Bernardus RE, Jones GS, Acosta AA, Garcia JE, Liu H-C, Jones DL, Rosenwaks Z: The significance of the ratio in follicle-stimulating hormone and luteinizing hormone in induction of multiple follicular growth. Fertil Steril43:373, 1985 6. Hofmann GE, Toner JP, Muasher SJ, Jones GS: High-dose follicle stimulating hormone (FSH) ovarian stimulation in low-responder patients for in vitro fertilization. J Vitro Fert Embryo Transfer 6:285, 1989 7. Muasher SJ, Oehninger S, Simonetti S, Matta J, Ellis LM, Liu H-C, Jones GS, Rosenwaks Z: The value of basal and/ or stimulated serum gonadotropins levels in prediction of stimulated response and in vitro fertilization outcome. Fertil Steril 50:298, 1988 8. Toner JP, Karande V, Jones GS, Muasher SJ: Basal follicle stimulating hormone (FSH) level is a better predictor of in vitro fertilization (IVF) performance than age. (Abstr. 0-003) Presented at the 46th Annual Meeting of The American Fertility Society, Washington, D.C., October 13 to 18, 1990. Published by The American Fertility Society in the Program Supplement, 1990, p 52
Padilla et al.
Lupron screening test for IVF
83
FERTILITY AND STERILITY
Printed on acid-free paper in U.S.A.
Copyright © 1991 The American Fertility Society
The Lupron* screening test: tailoring the use of leuprolide acetate in ovarian stimulation for in vitro fertilizationt
Santiago L. Padilla, M.D. t Rosella D. Smith, M.D. Jairo E. Garcia, M.D. Women's Hospital Fertility Center, Department of Gynecology, Greater Baltimore Medical Center, Baltimore, Maryland
Study Objective: To evaluate the prognostic and therapeutic value of a Lupron (leuprolide acetate; Tap Pharmaceuticals, North Chicago, IL) screening test before ovarian stimulation for in vitro fertilization (IVF). Design: Prospective. Setting: Outpatient IVF program. Patients: Eighty patients exhibited four early estradiol (E 2) patterns. Patients with pattern A and B remained on a flare-up ovarian stimulation protocol. Patients with pattern C were randomized to three ovarian stimulation protocols. Patients with pattern D were treated with the flare-up protocol using a high pure follicle-stimulating hormone (FSH) dose. Results: Patterns A, B, C, andD occured in 44%,16%,25%, and 15% of the patients, respectively. The E2 pattern recurred in 77% of subsequent IVF cycles. Pattern A and B patients achieved a 41 % (23/56) and 22% (5/23) ongoing pregnancy rate (PR) per stimulated cycle. An early luteal phase Lupron protocol had the best ongoing PR per stimulated cycle (10/27, 37%) in patients with a pattern C response. Pattern D patients had a 20% (5/25) ongoing PR per stimulated cycle. Conclusion: The Lupron screening test allows prospective selection of stimulation protocols in ovulatory patients undergoing IVF. Early E2 patterns A and B should be treated with the flare-up protocol. Pattern C patients benefit from the luteal phase Lupron protocol and pattern D patients benefit from a high pure FSH flare-up protocol. Fertil Steril 56:79, 1991
We have reported that the early serum estradiol (E 2 ) response to Lupron (leuprolide acetate; Tap Pharmaceuticals, North Chicago, IL) during a flareup ovarian stimulation protocol for in vitro fertilization (IVF) has prognostic value. 1 A retrospective analysis revealed four distinct early E2 response patterns. Pattern A showed a prompt E2 elevation by menstrual cycle day 3, followed by a fall by cycle day 4. Pattern B showed a delayed E2 elevation by
Received December 3, 1990; revised and accepted March 12, 1991. * Tap Pharmaceuticals, North Chicago, Illinois. t Presented at the 46th Annual Meeting of The American Fertility Society, Washington, D.C., October 13 to 18, 1990. :j: Reprint requests: Santiago L. Padilla, M.D., Greater Baltimore Medical Center Physician's Pavilion, 6565 North Charles Street, Suite 207, Baltimore, Maryland 21204. Vol. 56, No.1, July 1991
cycle day 4, followed by a fall by cycle day 5 or 6. Pattern C showed persistent E2 elevation through cycle day 5. Pattern D lacked a significant early E2 response. The ongoing live-birth pregnancy rates (PRs) per stimulated cycles were extremely favorable for patterns A and B and significantly lower for patterns C and D. Pattern A is not only associated with a better PR but also with a significantly higher embryo implantation rate. 1 The vast majority of the multiple pregnancies occurred in pattern A patients. Before the current study, we did a retrospective analysis to evaluate the reproducibility of these E2 patterns. We studied 77 patients who underwent consecutive IVF attempts and found that the patterns A, B, C, and D recurred in 81 %, 62%, 76%, and 80% of subsequent IVF cycles, respectively. 2 The purpose of this prospective study is to evaluate the Padilla et al.
Lupron screening test for IVF
79
f
clinical value of a Lupron screening test before ovarian stimulation in IVF and to analyze different stimulation protocols for the patients who carried a poor prognosis in our original flare-up protocol, i.e., the patients who exhibited a pattern C or D response. MATERIALS AND METHODS
The study population consisted of infertility patients with tubal factor, endometriosis, or unexplained infertility referred for IVF from January 1 to December 1, 1989. Patients with male factor infertility were excluded to avoid variability secondary to sperm factors. All patients had regular menstrual cycles, a normal endometrial cavity documented by hysterography or hysteroscopy, a normal endometrial biopsy, and a normal semen analysis. The Lupron screening test was done by administering Lupron 0.75 mg subcutaneously (SC) for patients < 75 kg and 1 mg SC for patients 2': 75 kg. Injections were given every morning on cycle days 2, 3, and 4. A serum E2 level was obtained on cycle days 2, 3, 4, and 5. A serum follicle-stimulating hormone (FSH) level was obtained on cycle day 2. All daily serum samples were drawn before the Lupron injections. The serum was refrigerated at 4°C, and all samples from the same patient were run in the same assay. Patients with a serum E2 pattern A or B response underwent ovarian stimulation with a flare-up protocol. Briefly, Lupron was started with the identical dose as the Lupron screening test dose. Gonadotropin therapy was initiated on cycle day 5 with 75 IU of pure FSH (Metrodin; Serono Inc, Randolph, MA) and one ampule of human menopausal gonadotropin (hMG, Pergonal; Serono Inc, Randolph, MA) containing 75 IU of FSH and 75 IU of luteinizing hormone (LH) given intramuscularly (1M) at 8 A.M. and the same dose of pure FSH and hMG at 3 P.M. The dose was then adjusted in a step-down manner based on the patient's response as described earlier. 3 Patients with E2 pattern C were randomized prospectively by a randomization table to one of three protocols. Protocol 1 consisted of a combination protocol that started with 150 IU of pure FSH and two ampules of hMG on cycle days 3 and 4, followed by two ampules of hMG from cycle day 5 on. Protocol 2 was the flare-up protocol, but the Lupron dose was 1.5 mg SC for patients < 75 kg and 2 mg SC for patients 2': 75 kg. Protocol 3 was an early luteal phase Lupron protocol. Patients were monitored every 2 to 3 days starting on cycle day 15 with serum progesterone (P) levels. Lupron was started 80
Padilla et al.
Lupron screening test for IVF
when ovulation was detected, as evident by a serum P > 3 ng/mL. Patients continued the Lupron until menses. If a serum E2 was 250 pg/mL, ultrasound (US) follicular monitoring was started. Patients on the FSH/hMG (protocol 1) had daily LH monitoring starting with US monitoring. Ovarian stimulation was continued until two follicles were at least 16 mm in size for all Lupron protocols and 15 mm in size for the pure FSH/hMG group without Lupron. That evening 10,000 U of hCG were injected 1M between 9 P.M. and 1 A.M. A transvaginal US-guided oocyte retrieval was scheduled 35 hours after the hCG injection. A maximum of five oocytes were inseminated with 150 X 103 washed motile sperm/mL of insemination media 3 to 26 hours after aspiration, depending on the degree of maturation. If cryopreservation was elected, up to 16 oocytes were inseminated, and a maximum of four embryos were transferred fresh. The embryo laboratory details have been described earlier. 4 Briefly, insemination media consisted of Ham's F-10 medium (Gibco, Grand Island, NY) with 7.5% heated-inactivated fetal cord serum. On day 1 after insemination, the cumulus was removed mechanically to look for signs of fertilization. If fertilization did not occur, the oocytes were reinseminated. Fertilized oocytes with three or more pronuclei were discarded. After fertilization, the embryos were transferred to growth media that consisted of Ham's F-lO with 15% fetal cord serum. Oocytes and embryos were incubated in 5% CO 2 in air at 37°C. Embryo transfer (ET) was done 72 hours after oocyte retrieval. All patients who underwent ET received supplemental P throughout the luteal phase. If pregnancy was established, the patient was given 250 mg of 17hydroxyprogesterone caproate 1M every week until the 12th week of gestation. Transvaginal US was done 3 to 4 weeks after ET for determination of Fertility and Sterility
fetal heart activity and detection of multiple pregnancies. Clinical pregnancies were established when two serum {3-hCG titers, drawn at least 10 days after ET, were rising, with US and/or histologic evidence of pregnancy. An ongoing pregnancy was defined as an intrauterine pregnancy of at least 14 weeks of gestation with presence of fetal heart activity. Serum E2 and FSH were measured in duplicate. Serum E2 was assayed with a commercial doubleantibody radioimmunoassay (RIA) kit available from Diagnostic Products Inc, (Los Angeles, California). Estradiol (1251 labeled) was used, and six E2 standards were run with concentrations between 20 and 3,600 pg/mL. Interassay coefficient of variation (CV) was between 4.6% and 13.7% depending on the level of hormone measured. The intra-assay CV depended on the level of hormone and ranged between 2.8% and 11.4%. Both of the high CVs were associated with the low values of E 2. The E2 antiserum was highly specific. The highest cross-reactivity to other natural steroids is with estrone (2.4%) and estriol (0.5%). Serum FSH was measured with a double-antibody RIA from Amersham Corporation (Arlington Heights, IL). Follicle-stimulating hormone 25 I labeled) and six standards between 0 and 145 mIU/mL were used for the assay. Intra-assay CV was between 3.1 % and 5.2%, and interassay CV was between 6% and 10.7%. The cross-reactivity of the FSH antiserum was 0.9% with thyroid-stimulating hormone, 0.08% with LH, and 0.06% with hCG. All samples were repeated when the intra-assay CV between duplicates was> 10%. Statistical evaluation was performed with Fisher's exact test.
e
RESULTS Eighty patients had the Lupron screening test. These patients underwent a total of 153 IVF cycles
Table 1 Frequency of the Early Serum E2 Pattern During the Lupron Screening Test and Reproducibility in the Subsequent Flare-Up Lupron Ovarian Stimulation for IVF Subsequent IVF cycles pattern Lupron screening pattern A B C D a b
(n = 35) (n = 13) (n = 20) (n = 12)
A 28 (80)a 3 (22)
NAb
B
C
D
3 (8)
1 (3) 1 (8)
3 (8) 1 (8)
NA
NA 10 (83)
8 (62) NA 1 (8)
Values in parentheses are percents. NA, not applicable.
Vol. 56, No.1, July 1991
1 (8)
Table 2 Ongoing PRs per Stimulated Cycle With Different Stimulation Protocols in Patients With Pattern C Response at the Time of the Lupron Screening Test
Stimulation protocol Phase 1 (n = 34) FSH/hMG (no Lupron) Double Lupron Luteal phase Lupron Phase 2 (n = 15) Luteal phase Lupron
Clinical PR/cycle
Ongoing PR/cycle
2/12 (17)a 2/10 (2W 6/12 (50)'
2/12 (27) 1/10 (10) 5/12 (42)
5/15 (33)
5/15 (33)
Values in parentheses are percents. Includes one spontaneous abortion. , Includes one tubal pregnancy. a
b
subsequent to the Lupron screening test. All patients underwent at least one IVF cycle. Mean age ± 1 SD of the group was 35 ± 3.7 years (range 28 to 40). Indications for IVF were tubal factor (55%), endometriosis (38%), and unexplained infertility (7%). Male factor infertility couples were excluded from the study. Table 1 describes the outcome of the Lupron screening test results and the pattern exhibited at the time of the subsequent ovarian stimulation for IVF. Only the first IVF cycles immediately after Lupron screening test were included in this table. The most common pattern was pattern A, followed by pattern C. The early E2 pattern is very reproducible, and the majority of the patients showed the same pattern in subsequent IVF attempts. Subsequent patterns were not included for patients with pattern C because they underwent different stimulation protocols. Table 2 summarizes the results from our pattern C treatment protocols. Phase 1 was the prospective, randomized study. Although there was no statistical difference between the three protocols, there was a clear trend toward a higher ongoing PR in the patients in the luteal phase Lupron protocol. In addition, there was one miscarriage in the double Lupron protocol and one ectopic pregnancy in the luteal phase Lupron protocol. We elected to stop the randomization and treat the next 15 consecutive patients with the pattern C response using the luteal phase Lupron protocol in phase 2 of the study. Five patients conceived in phase 2 and have ongoing pregnancies. We used the results of our first series1 of the flareup protocol (series I) as "classic" controls to evaluate the success rate of each pattern with the different protocols (Fig. 1). Pattern A patients had very similar ongoing PRs per stimulated cycle using the flareup protocol. Pattern B showed a lower, but not staPadilla et al.
Lupron screening test for IVF
81
1 ! I
Estradiol
LU~
Pattern
A
Series I
Series II
Ra~~?8~A\e
Rate Cycle
16/41 (39%)
~
B
~-
C
2/19 (11%)
D
1/18 (6%)
2
6/16 (38%)
On~Oing
23/56 (41%) 5/23 (22%) 10/27
(37%)•
5/25 (20%)
3 4 5 Cycle Day
Figure 1 Ongoing PR per stimulated cycle of the current series (series II) compared with our previous series (series I) using the flare-up protocol. The pattern C patients had a significantly higher ongoing PR per cycle with the luteal phase Lupron protocol used in series II. *, P = 0.04.
tistically significant, ongoing PR per stimulated cycle with the flare-up protocol. We compared the PR of the patients with the pattern C response in our previous series (series I) using the flare-up protocol to the PR obtained in the luteal phase Lupron protocol ofthe current study (series II). The 27 patients in the luteal phase Lupron protocol had a significantly higher ongoing PR per stimulated cycle (11 % versus 37%, P = 0.04) when compared with the flareup protocol in series I. In patients with pattern D response, we found a higher ongoing PR when the flare-up protocol was used with high pure FSH doses, but the difference did not reach statistical significance (6% versus 20%, P = 0.11). In the current study, we had five cancellations ofthe 153 cycles (3%), and they all occurred in the pattern D patients. We analyzed the multiple PR and found that 12 of the 15 multiple pregnancies were in the patients with the pattern A response with the flare-up protocol. Of the 23 ongoing pregnancies that occurred on this pattern, there were 9 (39%) twins, 2 (9%) triplets, and 1 (4 %) quadruplet pregnancies. There was one set of twins in the pattern B response with the flare-up protocol, one set of triplets in the pure FSH/hMG protocol, and one set of triplets in the luteal phase Lupron protocol. Seventeen patients (21%) had a baseline serum FSH level;;::: 20 mIU/mL on cycle day 2 of the Lupron screening test and underwent 30 stimulated cycles. One patient (6%) had a pattern A response, one (6%) had a pattern B response, 8 (47%) had a pattern C response, and 7 (41%) had a pattern D response. Four ongoing pregnancies occurred in this group for an ongoing PR of 13% per stimulated cycle and 24% per patient. Three ofthe patients who con82
Padilla et al.
Lupron screening test for IVF
ceived had pattern C response in the Lupron screening test, and the fourth patient had a pattern D response. In contrast, 63 patients (79%) with a baseline serum FSH level < 20 mIU /mL on cycle day 2 of the Lupron screening test underwent 152 stimulated cycles. Thirty-four patients (54%) had a pattern A response, 12 (19%) had a pattern B response, 12 (19%) had a pattern C, and 5 (8%) had a pattern D response. Forty-two ongoing pregnancies occurred in this group. The ongoing PR per stimulated cycle (28%) was higher than in the high baseline FSH group, although not statistically different. The ongoing PR per patient (67%) was significantly higher in the patients with a low baseline FSH than in the high baseline FSH group, P < 0.005. DISCUSSION
Our data show that a Lupron screening test can be performed before IVF and will assist in the selection of stimulation protocols. The early serum E2 response to Lupron is reproducible in the majority of subsequent IVF attempts. Patients with pattern A response have an excellent prognosis for ovarian stimulation with the flare-up protocol. Patients with pattern B response may have a slightly lower, but still very good, ongoing PR with the flare-up protocol. Our previous series showed that the flare-up protocol was not optimal for patients with pattern C response.! The patients who show evidence of ovarian suppression by cycle day 5, i.e., patients with pattern A or B, have an excellent prognosis, which contrasts with pattern C patients who do not have the early suppression on the flare-up protocol. Although the study had small numbers to reach statistical significance, we found no improvement using a double dose of Lupron with the flare-up protocol nor using a pure FSH/hMG protocol without Lupron. An early luteal phase Lupron protocol achieves ovarian suppression before the administration of gonadotropins, which appears to be important for Lupron protocols. In this study, the pattern C patients achieved a significantly higher PR using the luteal phase Lupron protocol when compared with the flare-up protocol results from our previous series. The patients with pattern D response have the lowest E2 levels, so we doubled the dose of pure FSH to four ampules per day. We observed a higher PR with this modification but no improvement on the cancellation rate when compared with our previous series. Bernardus et al. 5 reported an improvement in oocyte recovery rates and PRs with the use of a Fertility and Sterility
I I !
! l
higher pure FSH dose to improve the exogenous FSH/LH ratio. Hofmann et al. 6 found that high doses of pure FSH (6 ampules per day) reduced cancellation rates and improved PRs in low responders. Another benefit of the Lupron screening test is that it identifies the patients who are at a higher risk for multiple pregnancy. The vast majority of the multiple pregnancies in this study occurred in patients with pattern A response. These patients should have fewer embryos transferred. Patients with pattern A response are the prime candidates for embryo cryopreservation to reduce the multipIe PRo Our current recommendation is to perform a Lupron screening test on every ovulatory patient undergoing IVF. If the patient has a favorable response, i.e., patterns A or B, she can continue the flare-up protocol and do IVF in the same cycle of the Lupron screening test. Patients with pattern C response are advised to wait until ovulation occurs and switch to the luteal phase Lupron protocol. If the patient is from out of town, the Lupron screening test can be arranged very easily with the patient's local physician. Patients with a pattern C response can be started on the luteal phase Lupron protocol by their local physician and come to our center to start the ovarian stimulation after onset of menses. Patients with pattern D are treated with a flare-up protocol and high pure FSH doses on the same cycle of the Lupron screening test. Muasher et al. 7 and Toner et al. 8 have reported that elevated FSH levels on cycle day 3 are associated with a poor response to IVF. Our data also showed a lower ongoing PR per stimulated cycle and per patient with a baseline serum FSH ~ 20 mID /mL on cycle day 2. The baseline serum FSH level on cycle day 2 should be used as part of the Lupron screening test to counsel patients on their prognosis for success in IVF. In summary, the Lupron screening test is a simple and reproducible test that allows selection of individualized ovarian stimulation protocols for IVF.
Vol. 56, No.1, July 1991
The majority of IVF patients benefit from a flareup Lupron protocol, with the exception of the patients with a pattern C response who benefit from the luteal phase Lupron protocol. Acknowledgments. We are especially grateful to The Greater Baltimore Medical Center's IVF Laboratory for their help and support; Tap Pharmaceuticals, Chicago, Illinois, for the Lupron used for the study; Stephen Noel, Ph.D, Department of Pathology, The Greater Baltimore Medical Center for his help with the steroid assays; Theodore Baramki, M.D., Department of Gynecology, The Greater Baltimore Medical Center for his editorial guidance, and Mrs. Verna Young, IVF Coordinator, for typing this manuscript.
REFERENCES 1. Padilla, SL, Bayati J, Garcia JE: Prognostic value of the
early serum estradiol response to leuprolide acetate in in vitro fertilization. Fertil Steril 53:288, 1990 2. Padilla SL, Garcia JE: Unpublished data 3. Garcia JE, Padilla SL, Bayati J, Baramki TA: Follicular phase gonadotropin-releasing hormone agonist and human gonadotropins: a better alternative for ovulation induction in in vitro fertilization. Fertil Steril 53:302, 1990 4. Padilla SL, Garcia JE: Effect of maternal age and number of in vitro fertilization procedures on pregnancy outcome. Fertil Steril 52:270, 1989 5. Bernardus RE, Jones GS, Acosta AA, Garcia JE, Liu H-C, Jones DL, Rosenwaks Z: The significance of the ratio in follicle-stimulating hormone and luteinizing hormone in induction of multiple follicular growth. Fertil Steril43:373, 1985 6. Hofmann GE, Toner JP, Muasher SJ, Jones GS: High-dose follicle stimulating hormone (FSH) ovarian stimulation in low-responder patients for in vitro fertilization. J Vitro Fert Embryo Transfer 6:285, 1989 7. Muasher SJ, Oehninger S, Simonetti S, Matta J, Ellis LM, Liu H-C, Jones GS, Rosenwaks Z: The value of basal and/ or stimulated serum gonadotropins levels in prediction of stimulated response and in vitro fertilization outcome. Fertil Steril 50:298, 1988 8. Toner JP, Karande V, Jones GS, Muasher SJ: Basal follicle stimulating hormone (FSH) level is a better predictor of in vitro fertilization (IVF) performance than age. (Abstr. 0-003) Presented at the 46th Annual Meeting of The American Fertility Society, Washington, D.C., October 13 to 18, 1990. Published by The American Fertility Society in the Program Supplement, 1990, p 52
Padilla et al.
Lupron screening test for IVF
83
