Estradiol and progesterone replacement regimens for the induction of endometrial receptivity

FERTILITY AND STERILITY

Vol. 52, No.5, November 1989

Copyright" 1989 The American Fertility Society

Printed on acid-free paper in U.S.A.

Estradiol and progesterone replacement regimens for the induction of endometrial receptivity Kenneth Steingold, M.D.* Paul Stumpf, M.D.t David Kreiner, M.D.:j:

Hung-Ching Liu, Ph.D.§ Daniel Navot, M.D.11 Zev Rosenwaks, M.D.§

The Howard and Georgeanna Jones Institute for Reproductive Medicine, Department of Obstetrics and Gynecology, Eastern Virginia Medical School, Norfolk, Virginia

Initiation of pregnancy in premature ovarian failure patients by use of donated oocytes fertilized in vitro requires establishment of a normal endometrial environment. We compared administration of estradiol (E 2) and progesterone (P) by polysiloxane vaginal rings versus oral micronized E 2 and P vaginal suppositories in HY such patients. Serum E 2 levels were similar between groups and similar to normally-cycling controls. With vaginal administration of E 2 , a burst effect was noted, with marked elevation 1 hour after insertion. The pattern with oral administration was more consistent, although marked conversion· to estrone occurred. The P cylinder and suppositories delivered similar levels, with diminution of Pin some patients with the cylinder. Despite apparent limitations, endometrial histology was normal after each cycle; both groups achieved pregnancies. Administration of E 2 and P by polysiloxane vaginal rings achieved hormonal levels similar to oral micronized E 2 and P vaginal suppositories. Endometrial biopsies after the stimulated cycle were appropriately mature. Fertil Steril 52:756, 1989

Pioneering work in monkeys by Hodgen 1 and in humans by Lutjen et al., 2 Navot et al.,S and Rosenwaks4 established the feasibility of achieving pregnancies in agonadal subjects. The procedure requires artificial endometrial stimulation by sequential estrogen (E) and progesterone (P), and the transfer of in vitro fertilized, donated oocytes onto endometria thus rendered receptive. Although oral estradiol (E2), in the form of valerate, and either vaginal or intramuscular P has proved to be clinically successful,2•3 the optimal delivery route,

optimal dosages, and respective E and P absorption profiles have not been evaluated.4 The purpose of the present study was to assess critically two different delivery systems: vaginal polysiloxane rings and cylinders impregnated with 17{3 E 2 and P, respectively, and oral micronized E 2 with sequential P suppositories. Effectiveness of the two regimens was assessed by serum hormone levels, absorption characteristics, and the morphological changes of the endometrium. MATERIALS AND METHODS

Received February 2, 1989; revised and accepted July 7, 1989. * Present address and request for reprints: Kenneth Steingold, M.D., Box 34, Medical College of Virginia Station, Richmond, Virginia 23298. t Present address: Jersey Shore Medical Center, Neptune, New Jersey. t Present address: Long Island In Vitro Fertilization Center, Port Jefferson, New York. §Present address: Cornell Medical Center, New York City, New York. lf Present address: Mt. Sinai Medical Center, New York City, New York.

756

Steingold et al.

E 2 and P replacement regimens

Ten women with ovarian failure participated voluntarily in the study. All were amenorrheic, with low or undetectable serum E 2 levels (:o;;25 pg/ mL) and elevated serum follicle-stimulating hormone (FSH) values (>50 miU/mL). The women had a complete diagnostic work-up, as indicated in the Norfolk donor egg program. 3 Six normally menstruating women, aged 25 to 35 years, served as controls. Thirty days before they entered the study cycle, all hormonal therapy was discontinued. Five Fertility and Sterility

women (group 1) received the vaginal E 2 and P regimen. Polysiloxane vaginal rings were impregnated with 200 or 400 mg of E 2 • Polysiloxane vaginal cylinders were prepared with 2,000 mg of P, as previously described. 5 Briefly, the crystalline E 2 or P was mixed with fluid polysiloxane, activated with stannous octoate, and placed in appropriate molds. The 200 mg ring was inserted into the vagina on day 1 and left in situ until it was replaced by a 400 mg ring on day 9. That ring was replaced by a new 400 mg ring on day 11 and again on days 13 and 15. This last ring was left in situ until day 28. The P cylinder was placed into the vagina on day 15 and remained until removal on day 28. The second group of five women (group 2) received oral E 2 (Estrace; Mead Johnson, Evansville, IN) and P vaginal suppositories. Estradiol was given in a sequential fashion: dosage was 1 mg twice a day on days 1 through 5, increased to 2 mg twice a day on days 6 through 9. To mimic the midcycle E 2 surge, the dose was again raised to 2 mg three times a day on days 10 through 13, then decreased to 2 mg twice a day on days 14 through 28. Progesterone vaginal suppositories were given at a dosage of 25 mg three times a day on days 15 and 16, then increased to 50 mg three times a day for days 17 through 25. The dosage was reduced to 25 mg twice a day for the last three days of the stimulated cycle. Basal blood samples were obtained daily at 8:00 A.M. throughout the cycle before the morning dose of medication. In the normal controls, samples were obtained at 8:00A.M. throughout the menstrual cycle. Absorption studies were performed 1, 2, and 6 hours after ring placement, oral E 2 administration, or insertion of vaginal P cylinder or suppository. Blood was spun; serum was separated and kept frozen until assayed. E 2 , P, and estrone (E 1 ) were measured by commercially available radioimmunoassay (RIA) (E 2 and P, Pantex Corp., Santa Barbara, CA; E 1 , SCI Diagnostic, Los Angeles, CA). 6 All serum samples of a particular cycle were assayed in a single run. The interassay and intra-assay variations for E 2 , P, and E 1 were 5.7% and 5.8%, 6.6% and 5.2%, and 9.0% and 9.9%, respectively. Mean normal values and normal range (±2 standard deviation (SD)) were established by using data from the control group. Endometrial biopsies were performed with a Novak currette on days 21 and 26 during two consecutive cycles. The day of Vol. 52, No.5, November 1989

Estradiol !E2l ...... vaginal E2 o--oOral E2

Days

Figure 1 Mean± SE serum E 2 patterns in vaginal (closed circle) versus oral (open circle) administration. Shaded area denotes normal range. (See text for dosages of administration range.)

biopsy was normalized to the first day of P administration, which was designated as day 15. Endometrial biopsies were washed· with saline, then fixed with 10% buffered formaldahyde for histological dating, according to the criteria of Noyes et al. 7

RESULTS The mean ± standard error of the mean (SEM) oft he basal E 2 levels after vaginal and oral E 2 , compared with the normal controls, is shown in Figure 1. The basal pretreatment concentrations of E 2 were 15 pg/mL in both ovarian failure groups. The vaginally treated patients displayed a marked rise in E 2 on day 2 of therapy and reached a mean of 275 ± 35 pg/mL. These levels decreased until reinsertion of the 400 mg ring on day 9. With new ring insertions on days 11 and 13, a midcycle E 2 surge was mimicked, with a mean E 2 > 200 pg/mL. Because no new rings were inserted, the E 2 levels tapered off through the second half of the cycle, with a day 28 mean of 28 pg/mL. Estradiol response to· the oral administration revealed a more consistent pattern. Peak midcycle levels were 160 ± 30 pgjmL, reflecting the increased doses given on days 10 through 13. Luteal phase values remained constant between 80 and 100 pg/mL. In general, E 2 response to both treatment regimens compared favorably with the normal range established by the control group. Basal and absorption data of vaginal E 2 are Steingold et al.

E 2 and P replacement regimens

757

Ellredlol With Vaginal Ringe ~ I'

lb : I

-E2,ba181 o--<> £;z. lbiDfbllon

j\1

I

il

I1

I

::

I

: n I1 ~

'vJU (____ I

llvl

~~

-- -----------------------.' 14

16

18

20

22

24 26 26

Dey I

Figure 2 Basal and absorption E 2 levels following vaginal ring administration in one representative patient. Vaginal ring insertions and dosages are noted at the bottom of the graph.

shown in Figure 2. This is the· typical pattern noted; the results of a single patient are depicted. A marked bolus effect was seen with insertion of the 200 mg ring, attaining a level of 800 pg/mL by 1 hour. With placement of the 400 mg ring later in the cycle, a burst effect was again seen. Subsequent insertion of the rings revealed lesser fluctuations. Absorption of the oral 1 mg pill was discernible by 1 hour, with a peak of only 40 pg/mL at 6 hours. A similar absorption pattern was seen with the 2 mg dose, reaching a peak level of 100 pg/mL by 6 hours. The maximum E 2 level noted at midcycle was 240 pg/mL. Table 1 displays the morning serum levels of E 1 in patients receiving vaginal or oral E 2 • The results are shown as single values based on the day of therapy to show the marked variation. (Data were available on only seven of the patients.) Clearly, the oral administration of micronized E 2 resulted in marked increases in serum E 1 • Levels ranged from 30 to 3,000 pg/mL. In contrast, patient no. 4 displayed minimal elevation of E 1 • As expected, vaginal E 2 administration did not significantly affect circulating E 1 levels in two of three patients. However, patient no. 5 displayed marked increase in E 1 despite the vaginal route of administration. On day 13 she exhibited an E 1 level of 238 pg/mL, whereas the E 2 level was. 84 pg/mL. The P levels were similar in the vaginal suppository group and the cylinder patients, and fell within the range of the normal controls. The vaginal suppositories produced a level of 8.5 ng/mL after 24 hours and remained fairly constant for the duration of therapy, with a peak level of 11 ng/mL. The P cylinder produced a peak of 12.5 ng/mL on 758

Steingold et al.

E 2 and P replacement regimens

day 4; however, a decline in circulating levels was noted toward the end of the cycle. Less constant levels were achieved with the cylinder than with the suppositories, and certain patients decreased to 1 to 2 ng/mL after 2 weeks of treatment with one cylinder. Absorption data of the cylinder and suppositories revealed minimal fluctuation in P levels. Absorption was gradual with the cylinder, with a steady increase to a level of 8.5 ng/mL by 24 hours; no bolus effect was evident. A similar profile was seen with the suppositories. The 25 mg dose produced a level of 7 ng/mL at 6 hours, whereas the 50 mg dose elicited peak peripheral P values of 11 ng/mL. The endometrial histology, as determined from the biopsy specimens, displayed similar maturational changes in both hormone replacement groups. In all but one patient, there was a lag in glandular maturation of 2 to 5 days on the biopsy obtained on day 21; only one patient was within one day ofthe actual date of the biopsy. However, the stroma was normal for dates in all patients. Therefore, a characteristic glandular/stromal asynchrony was prevalent in the periimplantation phase biopsy. In contrast, the day 26 biopsy revealed a glandular and stromal morphology normal for dates in all patients. DISCUSSION

Since the feasibility of pregnancy initiation and maintenance by the use of donated oocytes fertilized in vitro has been established, major questions arise concerning the optimum mode of hormonal replacement therapy. The present study assessed two different regimens in terms of hormonal profiles and endometrial histology. Although the point was not specifically addressed in this study, pregTable 1 Serum Estrone Levels in Oral Versus Vaginal E 2 Patients

Day

Oral (pg/mL)

Vaginal (pg/mL)

Patient

1

8

13

21

27

1 2 3 4

30 30 60 30

566 1324 569 30

1130 2970 872 113

447 1398 471 68

805

5 6 7

35 30 30

92 30

238 30 30

30 30 30

30

364 87

59

Fertility and Sterility

nancies have been obtained in our program through both these regimens. Therefore, when one looks at the true end result of these therapies, both are found to be capable of initiating and maintaining pregnancies in patients with premature ovarian failure. The vaginal system of delivery was initially developed for contraceptive agents. Modifications in design resulted in the potential application of delivering E 2 to hypoestrogenic women. In a previous study, a dose-response effect was achieved by varying the amount of E 2 in the vaginal ring from 100 to 400 mg. 8 Relatively stable circulating levels ofE2 were apparent for up to three months. More recently, the vaginal rings and cylinders administered to hypogonadal women were reported to mimic the physiologic levels found in premenopausal women. 9 One advantage of vaginal administration is that it avoids the first passage effects of the liver and possibly untoward hepatic effects, as reported for oral estrogen replacement therapy. 10 The ring also delivers the primary ovarian estrogen 17{j E 2 • The theoretical benefit of this remains to be determined, because the vaginal administration of ethinyl E 2 has been associated with changes in hepatic-induced binding proteins. 11 The peripheral estrogen profile with the vaginal ring system compared favorably with physiologic E patterns. However, a clear burst effect was noted upon repeated insertions of the ring system, a finding previously noted in the contraceptive vaginal rings. 12 The burst.effect appeared to diminish over time, possibly due to diminished absorptive characteristics of the vaginal mucosa under E stimulation. Therefore, it appeared that the ring system could not provide a constant, steady delivery to effect consistent serum levels of E 2 • However, the fluctuating E 2 levels did not adversely affect endometrial morphology or receptivity. The oral therapy maintained a more consistent level of hormone delivery. However, consistent with previous data, extensive intestinal metabolism of E 2 was noted with oral therapyP- 16 Although E 2 levels remained within the desired range, marked elevations ofE 1 were noted. Levels ofE 1 as high as 3,000 pg/mL werEl seen with the oral therapy. Only unconjugated E 1 was measured in this study, but it has been reported that estrone sulfate may be increased even more than E 1 , to levels 20 to 50 times over baseline, after administration of oral micronized E 2 •15 In contrast, less significant changes in E 1 with the vaginal system were seen in the present study. Previous reports described adeVol. 52, No.5, November 1989

quate vaginal delivery of E 2 by administration of micronized E 2 in solution17 or by use of the vaginal rings 18 similar to the present study. In both these studies, the E 2 elevation was significantly greater than E 1 • In the present study, other hepatic effects, such as binding globulins and renin substrate, were not measured. However, in a study comparing oral and transdermal E 2 delivery in doses effective for donor oocyte programs, whereas oral E 2 predictably elevated certain hepatic parameters such as renin substrate, thyroid-binding globulin, and sex hormone-binding globulin, these markers were unaffected by the transdermal E 2 •19 Since very high doses of estrogens are used in donor oocyte programs, caution is advised until these adverse effects are completely understood. The P regimens were fairly similar in their ability to achieve adequate blood levels. Marked fluctuations were not noted with the suppositories or the cylind':r. The only concern with the cylinder was that levels tended to diminish toward the end of therapy in certain patients. It is possible to deliver higher levels of P by the vaginal route. In a previous study using 200 or 400 mg P suppositories in a wax base, levels up to 30 ng/mL were achieved. 20 However, as previously advised, 2 •3 dosage should be changed to intramuscular P supplementation (25 to 50 mg/d) after embryo transfer is performed. Despite fluctuating or sometimes diminishing levels of E 2 and P throughout the stimulated cycle, endometrial histology was adequately matured by day 26. All patients had normal endometrial morphology at the late luteal phase biopsy. In contrast, marked glandular/stromal asynchrony was noted on the day-21 biopsies in six of seven patients. These findings are in contrast with those of Navot et al., 3 who found close correlation of patients' biopsies on the day oftheir cycle. It is difficult to pinpoint the exact cause of this asynchrony, whether it is an E or a P effect. In a similar study using the vaginal ring system, the endometrial content of E 2 and P receptors was noted to be normal for women in the late luteal phase. 9 Endometrial histology in that study was done at the end of the cycle and also correlated with dates. It is possible that the method of P delivery is not subtle enough to mimic the normal cycle. There was no gradual increase in P levels, as noted at midcycle in normal patients. 21 Possibly this is important in helping to advance the glandular components, which appear to be lagging. Both regimens could deliver adequate levels of Steingold et al. E 2 and P replacement regimens

759

E 2 and P to stimulate a normal menstrual cycle. The other parameter measured, endometrial histology, also revealed normal maturation by day 26. Most importantly, ongoing pregnancies have been achieved with both methods of hormone replacement therapy. The sequelae of this high-dose medication-beyond the known factors of absorption fluctuation, intestinal metabolism, and hepatic effects-are presently unknown; therefore, the optimum mode of therapy remains to be determined. Acknowledgments. We thank Ms. Beverly Cox, Ms. Shirley Robinson, and Ms. Susan Alcorn for their excellent technical assistance with the laboratory work in this study and Charlotte Schrader, Ph.D., for editorial assistance.

REFERENCES 1. Hodgen GD: Surrogate embryo transfer combined with estrogen-progesterone therapy in monkeys. JAMA 250:2167, 1983 2. Lutjen P, Trounson A, Leeton J, Findlay J, Wood C, Renou P: Establishment and maintenance of pregnancy using in vitro fertilization and embryo donation in a patient with primary ovarian failure. Nature 307:174, 1984 3. Navot D, Laufer N, Kopolovic J, Rabinowitz R, Birkenfeld A, Lewin A, Granat M, Margalioth EJ, Schenken JG: Artificially induced endometrial cycles and establishment of pregnancies in the absence of ovaries. N Eng J Med 314: 806,1986 4. Rosenwaks Z: Donor eggs: their application in modern reproductive technologies. Fertil Steril 4 7:895, 1987 5. Stumpf PG, Maruca J, Santen RJ, Demers LM: Development of a vaginal ring for achieving physiologic levels of 17 -{3 estradiol in hypoestrogenic women. J Clin Endocrinol Metab 54:208, 1982 6. Jones GS, Acosta AA, Garcia JE, Bernardus RE, Rosenwaks Z: The effect of follicle stimulating hormone without additional luteinizing hormone on follicular stimulation and oocyte development in normal ovulatory women. Fertil Steril43:696, 1985 7. Noyes RW, Hertig AT, Rock J: Dating the endometrial biopsy. Fertil Steril1:3, 1950 8~ Stumpf PG: Selecting constant serum estradiol levels achieved by vaginal rings. Obstet Gynecol67:91, 1986

760

Steingold et al.

E 2 and P replacement regimens

9. Simon JA, Rodi IA, Stumpf PG, Sauer MV, Cohen SW, Ford LC, Bustillo M, Buster JE: Polysiloxane vaginal rings and cylinders for physiologic endometrial priming in functionally agonadal women. Fertil Steril 46:619, 1986 10. Geola FL, Fumar AM, Tataryn IV, Lu KH, Hershman JM, Eggena P, Sambhi MP, Judd HL: Biological effects of various doses of equine estrogens in postmenopausal women. J Clin Endocrinol Metab 51:620, 1980 11. Goebelsmann U, Maschak CA, Mishell DR: Comparison of hepatic impact of oral and vaginal administration of ethinyl estradiol. Am J Obstet Gynecol151:868, 1985 12. Mishell DR, Moore DE, Roy S, Brenner PF, Pate MA: Clinical performances and endocrine profiles with contraceptive vaginal rings containing a combination of estradiol and d-norgestrel. Am J Obstet Gynecol130:55, 1978 13. Yen SSC, Martin PL, Burnier AM, Czekala NM, Greaney MO, Callan tine MR: Circulating estradiol, estrone, and gonadotropin levels following the administration of orally active 17 -{3 estradiol in postmenopausal women. J Clin Endocrinol Metab 40:578, 1975 14. Adlercreutz H, Martin F, Pulkkinen M, Dencker H, Rimer U, Sjoberg N-0, Tikkanen MJ: Intestinal metabolism of estrogens. J Clin Endocrinol Metab 43:497, 1976 15. Dada OA, Laumas V, Landgren B-M, Cekan SZ, Diczfalusy E: Effect of graded oral doses of estradiol on circulating hormonal levels. Acta Endocrinol (Copenh) 88:754, 1978 16. Diczfalusy E, Franksson C, Lisboa BP, Martinsen B: Formation of estrone glucosiduronate by the human intestinal tract. Acta Endocrinol (Copenh) 40:537, 1962 17. Schiff I, Tulchinsky D, Ryan KJ: Vaginal absorption of estrone and 17{3 estradiol. Fertil Steril28:1063, 1977 18. Veldhuis JD, Samojlik E, Evans WS, Rogol AD, Ridgeway CE, Crowley WF, Kolp L, Checinska E, Kirschner MA, Thorner M 0, Stumpf P: Endocrine impact of pure estradiol replacement in postmenopausal women: alterations in anterior pituitary hormone release and circulating sex steroid hormone concentrations. Am J Obstet Gynecol 155:334, 1986 19. Steingold K, DeZiegler D, Reznikov S, Fratkin M, Matt D: Comparison oftransdermal to oral estradiol administration on hepatic biochemical markers in women with premature ovarian failure. Society for Gynecologic Investigation, 161 (March 1988) 20. Glazener CMA, Bailey I, Hull MGR: Effectiveness of vaginal administration of progesterone. Br J Obstet Gynaecol 92:364, 1985 21. Hoff JD, Quigley ME, Yen SSC: Hormonal dynamics at midcycle: a reevaluation. J Clin Endocrinol Metab 57:792, 1983

Fertility and Sterility