Comparison of transdermal versus oral estradiol on endometrial receptivity

r Vol. 65, No.2, February 1996

FERTILITY AND STERILITY Copyright

©

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

1996 American Society for Reproductive Medicine

Comparison of transdermal versus oral estradiol on endometrial receptivity

Joel S. Krasnow, M.D.*t Bruce A. Lessey, M.D., Ph.D.* Gregory Naus, M.D.§

Lori-Linell H. Hall, M.D.* David S. Guzick, M.D., Ph.D.* Sarah L. Berga, M.D.*

University of Pittsburgh, Pittsburgh, Pennsylvania, and University of North Carolina, Chapel Hill, North Carolina

Objective: To compare the effects of oral micronized E2 with transdermal E2 on endometrial receptivity in women undergoing oocyte donation. Design: Prospective, randomized, crossover trial. Serum E2 and P concentrations were measured on cycle days 14 and 22 (luteal day +8). Endometrial biopsies were obtained on day 22 and read in a blinded fashion for histology and ,B-3-integrin expression. Setting: University-based donor oocyte program. Patients: Twenty-seven patients presenting for donor oocytes. Main Outcome Measures: Endometrial histology and ,B-3-integrin expression. Results: The endometrial glandular histology in women given oral micronized E2 was delayed by a mean of 1.6 days in comparison to that of women given transdermal E2. Seventy percent of women given oral E2 displayed a lag;;:,:4 days whereas 29.6% given transdermal E2 displayed a similar lag. Serum E2 levels were 1,194 ::!:: 108.8 pg/mL (mean::!:: SEM; conversion factor to SI unit, 3.671) in women on oral micronized E2 and 117.4 ::!:: 14.0 pg/mL in those on transdermal E 2. Conclusion: The supraphysiologic serum E2 levels associated with oral Inicronized E2 may have a deleterious impact on endometrial receptivity. The development of more physiologic hormone replacement protocols may enhance endometrial receptivity and lead to improved Fertil Steril 1996; 65:332-6 clinical pregnancy rates. Key Words: Endometrial biopsy, endometrium, luteal phase deficiency, ovum donation

Failure to achieve pregnancy after IVF-ET may be a consequence of embryonic or endometrial factors. Ovum donation offers the unique opportunity to isolate endometrial from oocyte development. Pregnancy rates from ovum donation generally are higher than those achieved through standard IVFET, even when donors and recipients share the 00cytes equally (1, 2). Part of the difference may relate to the ovarian stimulation. Ovarian stimulation in infertile couples has been reported to result in higher Received March 9, 1995; revised and accepted July 28, 1995. * Department of Obstetrics, Gynecology, and Reproductive Sciences, University of Pittsburgh School of Medicine. t Reprint requests: Joel S. Krasnow, M.D., Magee-Womens Hospital, University of Pittsburgh, 300 Halket Street, Pittsburgh, Pennsylvania 15213 (FAX: 412-641-1133). :j: Department of Obstetrics and Gynecology, University of North Carolina at Chapel Hill. § Department of Pathology, University of Pittsburgh School of Medicine. • 332

rates of pregnancy loss than in natural cycles (3). Paulson et a1. (4, 5) have proposed that the supraphysiologic E2 levels that occur in association with stimulated cycles may have an adverse impact upon endometrial receptivity. Endometrial biopsies from women treated with gonadotropins demonstrate a high rate of out-of-phase biopsies, and asynchronous development of glands and stroma (6-9). In the majority of these studies, the endometrium was sampled in the early luteal phase or in the late luteal phase after a negative serum ,a-hCG. Some authors suggest that endometrial assessment in the peri-implantation period may be more prognostic for implantation competency (10). The dose, duration, and route ofE 2 administration used in donor oocyte programs varies widely (1). The regimen used in our program from 1990 to 1992 was 2 mg oral micronized E2 (Estrace; Mead-Johnson, Princeton, NJ) three times daily. Because the ongoing pregnancy rate using this regimen was substan-

Krasnow et al. Comparison of transdermal versus oral E2

Fertility and Sterility

tially lower than that reported by others (2,11-14), we decided to compare it with a transdermal E2 delivery system. A histologic endpoint was selected as a marker of uterine receptivity rather than the clinical endpoint of pregnancy because the large sample size that would be required to assess clinical pregnancy rates was not feasible in our program. A proliferative phase length of 14 days was chosen based on the normal menstrual cycle data and the observation that peak pregnancy rates for ovum recipients occurred when 11 to 19 days of estrogen therapy was given before the initiation ofP supplementation (11). To minimize the impact of patient selection, all subjects served as their own controls in that each received both oral and trans dermal replacement regimens in separate mock cycles. A power analysis was performed on the basis of a pilot study in which eight subjects, each receiving one of the two treatments, were compared. Thirty subjects were calculated to be sufficient to detect a 2-day difference in histology (ex = 0.05, f3 = 0.80). To allow for patient drop out, 35 subjects were recruited. MATERIALS AND METHODS

Thirty-five women presenting for oocyte donation between January 1992 and December 1994 were recruited for this study. The protocol was approved by the institutional review board at Magee-Women's Hospital and informed consent was obtained from all women. Ovarian failure was defined as a serum FSH > 40 mIU/mL (conversion factor to SI unit, 1.00) with a serum E2 < 10 pg/mL (conversion factor to SI unit, 3.671) on two separate occasions. Individuals demonstrating any endogenous ovarian function underwent treatment with 0.1 mg SC leuprolide acetate (Lupron; Tap Pharmaceuticals, Chicago, IL) daily to suppress residual ovarian activity. Suppression was documented by E2 < 10 pg/mL, and LH and FSH < 3.5 mIU/mL (conversion factor to SI unit for LH, 1.00). To achieve physiologic serum E2 concentrations, a 0.1 mg transdermal E2 patch (Estraderm; Ciba-Geigy, Summit, NJ) was applied every 3 days. Supraphysiologic E2 concentrations were achieved by administering 2 mg of micronized E2 (Estrace; Mead-Johnson, Princeton, NJ) orally three times daily. Progesterone supplementation was initiated on the morning of the 15th day of the hormone replacement cycle by administering 100 mg P vaginal suppositories twice daily. Estradiol administration was continued throughout the secretory phase of the cycle. Hormone replacement was continued until cycle day 28. Patients were assigned randomly to receive either the transdermal or oral E2 preparation first, based Vol. 65, No.2, February 1996

on the calendar day of their visit (odd or even). Serum was obtained for sex steroid determination just before the morning Estrace dose. Endometrial biopsies were obtained on the morning of the 8th day of P supplementation (day 22) with the unimar pipelle (Prodimed, Neuilly-en TheIle, France) endometrial suction curette. Biopsies were dated by a single pathologist (G.N.) who was blinded to treatment status. The standard dating criteria of Noyes et al. (15) were used. Immunohistochemistry for f3-3-integrin was performed as described previously (16). A portion of each biopsy was snap frozen in liquid nitrogen and stored at -80°C until assayed. A semiquantitative assessment of staining, HSCORE, was determined by a single observer blind to treatment group. The HSCORE was calculated using the following equation: HSCORE = L Pi (i + 1), where i is the intensity of staining with a value of 1,2, or 3, (weak, moderate or strong, respectively) and Pi is the percentage of stained epithelial cells for each intensity, varying from 0% to 100%. Serum was obtained for E2 and P on days 14 and 22 of each cycle. Serum samples were assayed in duplicate for E2 (DPC-RIA; Diagnostics Products Corporation, Los Angeles, CA) and P (ACS 180 Chemiluminescence; Ciba Corning, East Walpole, MA) using commercially available assays. Eight patients dropped out of the study. Five patients discontinued the study after the first endometrial biopsy. One patient had a P of 11.1 nM on day 14 despite treatment with leuprolide acetate. Insufficient tissue was obtained in two biopsies. The remaining 27 paired samples form the basis for comparison. Results are expressed as mean ± SEM. Statistical analysis was performed using paired ttest, X 2 with Yates' correction, and the sign test as indicated. RESULTS

The 27 patients studied were referred for oocyte donation because of advanced age (n = 7), failure to stimulate in response to exogenously administered gonadotropin therapy (n = 11), or premature ovarian failure (POF) (n = 9). The group with POF was younger (32.6 ± 1.2 years) than those women referred for failed ovarian stimulation (37.1 ± 1.1 years; P = 0.01) or advanced age (44.6 ± 0.3 years; P = 0.0001). On cycle days 14 and 22 of an artificial cycle, serum E2 during oral E2 administration was 1,194.3 ± 108.8 and 1,276.2 ± 110.7 pg/mL compared with 117.4 ± 14.0 and 107.0 ± 8.7 pg/mL with transdermal E2 therapy (P < 0.00001). There was no significant difference in serum P concentrations on day 14 (0.3 ± 0.1 versus 0.3 ± 0.1 ng/mL; conversion factor to SI unit, 3.180) or day 22 (13.6 ± 1.5 Krasnow et al.

Comparison of transdermal versus oral E2

333

-~---~

.-...

~-<><-

-

~

... .. .. ~

~

-~--.-

21 ..-, <'-l

.§

T

'-'

..9

.-::t.s <'-l

*

18 17

16

Treatment Figure 1 Histology from paired endometrial biopsies performed on day 22 of oral micronized E2 and transdermal E2 oral micronized E 2. t:::l, oral micronized E2 glands, [[I, trans dermal E2 glands; ., oral micronized E2 stroma; 13, transdermal E2 stroma. *Statistical significance versus the transdermal E2 glands, P = 0.0009. Data are expressed as means:!: SEM.

versus 14.0 ± 1.4 ng/mL) for the oral E2 and transdermal E2 groups, respectively. No differences were noted in serum E 2, P, or endometrial histology between patients with POF, failed stimulation, or advanced age. Histologic assessment of endometrial glandular development demonstrated a mean lag of 1.6 days during treatment with oral E2 when compared with transdermal E2 (t = 3.754, P = 0.009; Fig. 1). Histologic assessment ofthe stroma did not differ between treatments. The stroma was more advanced than the glands during both the oral E2 (day 19.6 versus day 18.3; t = 3.251; P = 0.003) and transdermal E2 (day 20.5 versus day 19.9; t = 2.068; P = 0.05) treatments. Endometrial biopsies commonly are considered out of phase when there is a 3-day lag. Because endometrial P exposure occurs 1 day later in artificial cycles compared with natural cycles, the incidence of out-of-phase biopsies by 2:4 days (day 18 or less) is reported. Using this criteria, 19 (70.6%) biopsies were out of phase with oral E2 compared with 8 (29.6%) with transdermal E2 (X 2 = 7.407, P = 0.01). ,B-3-integrin expression, a marker of uterine receptivity, was 0.12 ± 0.08 with oral E2 and 0.48 ± 0.23 with transdermal E2 (P = 0.10). Assessment of the glandular histology demonstrated that the 1.6-day difference between treatments groups could be separated into a population with little or minimal difference «3 days) between treatments and another population in which larger differences (2:3 days) between treatI}lents were noted (Fig. 2). Of 12 patients demonstrating a dis334

crepancy of 2:3 days between treatments, 11 of 12 had a more advanced pattern while on transdermal therapy (P = 0.01, sign test). DISCUSSION

T

2 0

>. 1 9 01)

r

Krasnow et aI. Comparison of trans dermal versus oral E2

The oocyte donation model allows endometrial maturation to be studied independent of ovarian function. The serum E2 achieved with transdermal E2 was similar to the mean E2 concentration during the follicular phase of eumenorrheic women (17). Administration of 6 mg/d of oral micronized E2 resulted in supraphysiologic E2 concentrations that are comparable to those seen with gonadotropin therapy. However, both treatments maintain serum E2 concentrations seen in natural cycles. The midluteallag in endometrial maturation of 2 to 4 days that was seen with both treatments in this study has been observed previously in oocyte recipients treated with exogenous E2 and P (11-14, 18). This occurred whether the E2 priming was performed in a fixed (11-14, 18) or incremental dose (14). However, all the incremental E2 regimens reported used supraphysiologic E2 concentrations. Therefore, it remains to be seen whether incremental E2 rises mimicking the physiologic E2 rises of a natural cycle can enhance further endometrial maturation in artificial cycles. The midluteallag in endometrial histology may be explained partially by a 1-

6

5-

>.

5 & J:

»:

4-

»: =::::

'.

»:

3-

»:

21- m

.... ....

0

"»'

I I

I

I

I

"t

<'(

N

:: :: :

r::::

:

I

I

'.

:

:::::

:.: . .... .... .:.: lI'l

:

I

I

I

0

I

N

I (f'l

I ~

I

lI'l

Glandular Histology

(Days) Figure 2 A comparison of paired endometrial biopsies on day 22 by individual subject. The data are illustrated as a frequency distribution of the endometrial glandular dating expressed as transdermal E2 minus oral micronized E2 (t:::l, trans dermal E2 - oral micronized E 2.)

Fertility and Sterility

day delay in endometrial P exposure compared with the natural cycle. Supraphysiologic follicular phase E2 levels also have been implicated in this maturational delay (11). Our finding that sustained follicular phase E2 levels > 1,000 pg/mL results in a 1.6day lag in glandular maturation compared with physiologic E2 concentrations supports this contention. The enhanced glandular maturation seen with transdermal E2 could be a consequence of a more physiologic E2 milieu or, alternatively, could be a reflection of the route of E2 administration. When compared with transdermal E 2, oral E2 also results in higher serum levels of estrone (E l ) and its conjugates because of hepatic metabolism ofE 2 to El (19). With trans dermal administration, the effects of gastrointestinal absorption are avoided and E2:El ratios approximate 1.0 (20). To separate the effect of serum E2 level from route of administration, it will be necessary to compare trans dermal to oral routes of administration within the physiologic E2 range. The wide biologic variation in endometrial responses seen between treatments in our cohort suggests that the choice of hormonal treatment for endometrial priming in preparation for donor oocytes will be more important for some patients than for others. The association between endometrial histology and pregnancy outcome remains to be determined. Nonetheless, it is intriguing that, in the review by Benshushan and Schenker (21), the uterine preparation protocol using transdermal E2 had a 55.4% pregnancy rate per transfer compared with a 29.4% pregnancy rate per transfer for those cycles with oral E2 valerate. However, when parenteral E2 valerate was administered biweekly and titrated to achieve a serum E2 level of 400 to 700 pg/mL, the clinical pregnancy rate was 61.3% (22). Both parental E2 valerate and transdermal E2 results in an E2:El ratio approximating 1.0 (19, 20). The possibility that E2:El ratios :s; 1.0 or that high El concentrations are detrimental to implantation awaits further investigation. The clinical significance of a 1.6-day lag in endometrial glandular maturation is debatable. Davies et aL (23) reported an increase in clinical pregnancy rates in ovum recipients when endometrial biopsies in mock cycles were within 2 days ofthe chronological biopsy date. Biopsies that are out of phase by ~3 days in natural cycles are associated with a reduction in clinical pregnancy rates (24). Therefore, a comparison of the number of out-of-phase biopsies in each group may be a relatively good reflection of endometrial receptivity. The number of biopsies that were out of phase using our criteria of ~4 days for artificial cycles was greater (P = 0.01) in the oral E2 group, which supports the notion that supraphysioVol. 65, No.2, February 1996

logic E2 concentrations may have a deleterious impact upon implantation. Endometrial biopsies performed in the midluteal phase show a higher incidence of out-of-phase biopsies than late luteal phase biopsies. This has been seen in both natural cycles and hormonal replacement cycles in preparation for donor oocytes (10, 12). The mechanism by which these biopsies "catch up" is not known. Because the dating of late luteal biopsies is dependent primarily upon morphological characteristics within the stroma, it is possible that peri-implantation abnormalities in glandular histology will go unrecognized. This was the rationale behind our investigation of specific implantation markers in the midluteal phase. To investigate the mechanism by which these two treatment regimens may impact upon endometrial receptivity, we studied ,B-3-integrin expression, which is a marker of uterine receptivity. The absence of ,B-3 integrin expression has been associated with lower pregnancy rates in various clinical conditions (25). In natural cycles, in-phase midluteal biopsies almost always have an HSCORE > 0.7 (25). In our cohort, the majority of subjects had an HSCORE < 0.7. This is consistent with the data in which the mean histologic day is <20 with both treatments, because ,B-3-integrin expression usually is not detected until day 20 (16, 25). Despite the relatively high incidence of out-of-phase biopsies and low or absent ,B-3-integrin expression, the recipients in donor oocyte programs have a high rate of conception. One interpretation is that histologic lag or absence of ,B-3-integrin expression may not be critical to the implantation process during oocyte donation. Younger oocytes may result in embryos that are better able to overcome the delay or deficits in endometrial maturation. However, protocols of hormone replacement that induce normal histology and ,B-3integrin expression theoretically should result in even greater conception rates when donor oocytes are used. In summary, endometrial biopsy and ,B-3-integrin expression serve as imperfect estimates of uterine receptivity. In this study, transdermal E2 therapy results in more physiologic circulating E2 levels and improved endometrial histology compared with 6 mg daily of micronized E 2. To assess whether the differences in histology are a function ofE 2 levei or route of administration, further studies comparing oral and transdermal routes at physiologic E2 levels are required. Acknowledgments. We thank Ms. T. Babyak for her assistance with patient recruitment and scheduling; Ms. K. Shell and J. Sun, M.D., for their assistance with the ,B-3-integrin analysis; and Tammy Daniels, B.S., for expert laboratory assistance.

Krasnow et al. Comparison of transdermal versus oral E2

335

REFERENCES 1. Varon Y, Botchan A, Amit A, Peyser MR, David MP, Lessing JB. Endometrial receptivity in the light of modern assisted reproductive technologies. Fertil SteriI1994;62:22532. 2. Check JH, Nowroozi K, Chase J. Comparison of pregnancy rates following IVF-ET between donors and recipients in a donor oocyte program. J Assist Reprod Genet 1992;9: 248-50. 3. Liu H-C, Jones HW Jr, Rosenwaks Z. The efficiency of human reproduction after in vitro fertilization and embryo transfer. Fertil Steril 1988;49:649-53. 4. Paulson RJ, Sauer MV, Lobo RA. Embryo implantation after human in vitro fertilization: importance of endometrial receptivity. Fertil Steril 1990;53:870-4. 5. Paulson RJ, Sauer MV, Francis MM, Macaso TM, Lobo RA. In vitro fertilization in unstimulated cycles: the University of Southern California experience. Fertil Steril1992;57:2903. 6. Ben-Nun I, Jaffe R, Fejgin MD, Beyth Y. Therapeutic maturation of endometrium in in vitro fertilization and embryo transfer. Fertil Steril 1992;57:953-62. 7. Forman RG, Eychenne B, Nessmann C, Frydman R, Robel P. Assessing the early luteal phase in in vitro fertilization cycles: relationships between plasma steroids, endometrial receptors and endometrial histology. Fertil Steril 1989; 51:310-6. 8. Metzger DA. Luteal phase following ovulation induction. Endometrial pathology and luteal support. Assist Reprod Rev 1992;2:62-77. 9. Benadiva CA, Metzger DA. Superovulation with human menopausal gonadotropins is associated with endometrial gland-stroma dyssynchrony. Fertil Steril 1994;61:700-4. 10. Castelbaum AJ, Wheeler J, Coutifaris CB, Mastroianni L Jr, Lessey BA. Timing of the endometrial biopsy may be critical for the accurate diagnosis of luteal phase deficiency. Fertil SteriI1993;61:443-7. 11. Younis JS, Mordel N, Newin A, Simon, Schenker JC, Laufer N. Artificial endometrial preparation for oocyte donation: the effect of estrogen stimulation on clinical outcome. J Assist Reprod Genet 1992;9:222-7. 12. Rosenwaks Z, Navot D, Veeck L, Lie HC, Steingold K, Kreinger D, et al. Oocyte donation. The Norfolk Program. Department of Obstetrics and Gynecology, Eastern Virginia Medical School, Norfolk. Ann NY Acad Sci 1988;541:728-41.

336

Krasnow et al.

Comparison of transdermal versus oral E2

13. Younis JS, Mordel N, Ligovetzky GA, Lewin A, Schenker JG, Laufer N. The effect of prolonged artificial follicular phase on endometrium development in an oocyte donation program. J In Vitro Fert Embryo Transf 1991;8:84-8. 14. Navot D, Anderson TL, Droesch K, Scott RT, Kreiner D, Rosenwaks Z. Hormonal manipulation of endometrial maturation. J Clin Endocrinol Metab 1989;68:801-7. 15. Noyes RW, Hertig AT, Rock J. Dating the endometrial biopsy. Fertil Steril 1950; 1:3-25. 16. Lessey BA, Damjanovich L, Coutifaris C, Castelbaum A, AlbeIda SM, Buck CA. Integrin adhesion molecules in the human endometrium. Correlation with the normal and abnormal menstrual cycle. J Clin Invest 1992;90:188-95. 17. Powers MS, Schenkel L, Darley PE, Good WR, Balestra JC, Place VA. Pharmacokinetics and pharmacodynamics of transdermal dosage forms of 17,B-estradiol: comparison with conventional oral estrogens used for hormone replacement. Am J Obstet Gynecol 1985; 152: 1099-106. 18. Younis JS, Ezra Y, Sherman Y, Simon A, Schenker JG, Laufer N. The effect of estradiol depletion during the luteal phase on endometrial development. Fertil Steril 1994; 62:103-17. 19. Dusterberg B, Schmidt-Gollwitzer M, Humpel M. Pharmacokinetics and biotransformation of oestradiol valerate in ovariectomized women. Horm Res 1985;21:145-54. 20. Chetkowski RJ, Meldrum DR, Steingold K. Biologic effects of transdermal oestradiol. N Engl J Med 1986;314:1615-20. 21. Benshushan A, Schenker JG. Reproductive health care policies around the world. Ovum donation-An overview. J Assist Reprod Genet 1993;10:105-11. 22. Feinman MF, Sher G, Massaranni G, Vaught L, Andreyko J, Salen R, et al. High fecundity rates in donor oocyte recipients and in-vitro fertilization surrogates using parenteral oestradiol valerate. Hum Reprod 1993;8:1145-7. 23. Davies MC, Anderson MC, Mason BA, Jacobs HS. Oocyte donation: the role of endometrial receptivity. Hum Reprod 1990;5:862-9. 24. Klentzeris LD, Li TC, Dockery P, Cooke ID. The endometrial biopsy as a predictive factor of pregnancy rate in women with unexplained infertility. Eur J Obstet Gynecol Reprod BioI 1992;45:119-24. 25. Lessey BA, Castelbaum AJ, Sawin SW, Buck CA, Schinnar R, Bilker W, et al. Aberrant integrin expression in the endometrium of women with endometriosis. J Clin Endocrinol Metab 1994;79:643-9.

Fertility and Sterility