- Email: [email protected]
High implantation rate and consequently high pregnancy rate by in vitro fertilization-embryo transfer treatment in infertile women with antisperm antibody*
FERTILITY AND STERILITY
Vol. 63, No.1, January 1995
Copyright© 1995 American Society for Reproductive Medicine
Printed on acid-free paper in U. S. A.
High implantation rate and consequently high pregnancy rate by in vitro fertilization-embryo transfer treatment in infertile women with antisperm antibody*
Toshifumi Daitoh, Ph.D. Masaharu Kamada, Ph.D.t Syuji Yamano, Ph.D. Shinji Murayama, M.D.
Tsuzuki Kobayashi, M.D. Masahiko Maegawa, Ph.D. Toshihiro Aono, Ph.D.
Department of Obstetrics and Gynecology, The University of Tokushima, School of Medicine, Tokushima, Japan
Objective: To examine the effect of antisperm immunity on postfertilization steps, such as implantation of embryos and fetal growth, in IVF-ET treatment of women with sperm-immobilizing antibodies. Design: Retrospective analysis of clinical laboratory data. Setting: The IVF-ET program of the Department of Obstetrics and Gynecology, The University of Tokushima, School of Medicine. Patients: Eighteen women with sperm-immobilizing antibodies and 122 infertile patients with nonimmune etiology as controls. Infertile couples due to a male factor and with unknown etiology were excluded. Interventions: All patients received the same IVF-ET program with GnRH agonist. Main Outcome Measures: Rates of fertilization and cleavage, implantation rate per embryo transferred and pregnancy rate (PR) in both test and comparison groups. Results: The rate of fertilization in the antisperm group (61.3%) was significantly lower than that in the comparison group (76.8%). But the implantation rate per embryo transferred (23.5%) and consequently the modified PR per oocyte recovery procedure (34.4%) in immunologically infertile women were significantly higher than those in the comparison group (7.9% and 17.8%, respectively). Conclusions: Although sperm-immobilizing antibodies prevent sperm-egg interaction, they do not seem to have any adverse effects on achievement of pregnancy. Moreover, the existence of antisperm immunity in woman with antisperm antibodies is suggested to be favorable for successful pregnancy by the IVF-ET procedure. Fertil Steril1995;63:87-91 Key Words: antisperm antibody, infertility, in vitro fertilization, implantation, immunology of pregnancy
We have reported that sperm-immobilizing antibody causes infertility by inhibiting the acrosome Received September 13, 1993; revised and accepted July 22, 1994. * Supported in part by a Grant-in-Aid for Scientific Research (No. 04671005) from the Ministry of Education, Science, and Culture of Japan, Kasumigaseki, Chiyoda-Ku, Tokyo, Japan. t Reprint requests: Masaharu Kamada, M.D., Ph.D., Department of Obstetrics and Gynecology, The University of Tokushima, School of Medicine, 3-18-15 Kuramoto-cho, Tokushima 770, Japan (FAX: 81-0886-31-2630). Vol. 63, No. 1, January 1995
reaction of sperm (1, 2). The inhibitory effect is reversible by incubation of the antibody-bound sperm in antibody-free medium (2) and there was a successful case in IVF using fetal cord serum admixtured to the medium instead of the patient's serum containing sperm-immobilizing antibodies. Therefore, we proposed that circulating antisperm antibodies in the follicular or tubal fluid may cause infertility by blocking fertilization and that IVF is a suitable measure for achieving pregnancy in patients with antisperm antibodies (1). In fact, since Daitoh et al.
IVF-ET on women with antisperm antibody
87
the first report by Ackermann et al. (3), IVF with donor serum has been widely used for the treatment of patients with immunological infertility, and some successful pregnancy cases have been reported (4-6). Another possible mechanism by which antisperm antibodies cause infertility or pregnancy failure is proposed from studies in animal models to be by their cytotoxic effects on the fertilized eggs (7, 8), because sperm antigens have been demonstrated on the surface of fertilized ova (9). Furthermore, there have been some clinical reports suggesting the relationship between a presence of antisperm antibodies and recurrent spontaneous abortion (10, 11). However, the above discussion is still controversial (12). On the contrary, maternal immune reactions to fetal antigens are known to be mandatory for maintenance of pregnancy by preventing rejection ofthe alloimmunogenic fetus or by enhancing embryo growth (13, 14). The sperm antigen(s) on the surface of the embryo (9) or trophoblast (15) may elicit favorable maternal immune responses for successful implantation of a semiallograft into the uterus. Indeed, there is a recent report (16) that the trophoblast-lymphocyte cross-reactive (TLX) antigen, which has been suggested to induce a maternal immune response and to be important for maintenance of pregnancy (13), is one of these sperm antigens. To determine the possible effects of antisperm immunity on nidation or embryo growth in women with sperm-immobilizing antibodies, we examined the outcome of IVF-ET treatment in immunologically infertile women and, as the control, infertile women due to tubal factor.
MATERIALS AND METHODS
pared to those in a comparison group, namely, 273 oocyte-recovery cycles of 122 couples, whose indication for IVF-ET treatment was tubal obstruction but no immunological abnormalities. No cases showing poor quality of the husband's semen according to the criteria of the World Health Organization (17) were included. We also excluded infertile couples with unknown etiology and cases in which fertilization failure was due to anti-zona pellucida (ZP) activity in the wife's serum (18, 19) or antisperm activity (sperm immobilization) in the husband's serum. During the same 5-year period, IVF-ET procedures such as egg retrieval, culture of gametes, and ET have been performed in order of registration by the same clinical group. Ovarian Stimulation and Follicle Development
All patients were treated with GnRH agonist (GnRH-a, buserelin acetate, Hoechst Japan, Tokyo, Japan) beginning 8 days before the estimated 1st day of the menstrual cycle. From the 3rd day of menstruation or the 14th day from the start of GnRH-a administration when menstruation was missed, 225 IU of hMG (HMG Nikken; Nikken, Tokyo, Japan) or pure FSH (Fertinorm-P; Serono Japan, Tokyo, Japan) was injected daily for 4 days and subsequently 150 IU of hMG (Humegon, Nippon Organon, Tokyo, Japan) was administered daily until the mean diameter of the dominant follicle reached 18 mm as determined by ultrasonography (RT3600; Yokogawa, Tokyo, Japan). Ovulation was triggered by injection of 5,000 IU of hCG (Gonatropin; Teikoku Zohki, Tokyo, Japan) 50 hours after the last injection of hMG, and oocytes were retrieved transvaginally 35 hours later under local anesthesia using a needle-guided technique under ultrasound (US) control. Fertilization and Embryo Culture
In Vitro Fertilization-Embryo Transfer
Patients
In vitro fertilization-embryo transfer therapy was carried out from February 1989 through June 1993 on 35 oocyte recovery cycles of 18 women whose sera showed sperm-immobilizing activity (the antisperm group) at the Infertility Clinic of the Department of Obstetrics and Gynecology, Tokushima University Hospital. The rates of fertilization, cleavage, implantation, and pregnancy in the antisperm group were com88
Daitoh et al.
IVF-ET on women with antisperm antibody
Whittingham's T6 (T6) medium supplemented with donor serum was used for insemination and embryo culture in vitro. Donor sera were separated from blood of umbilical vein of fetus delivered from normal healthy mothers or nonimmunologically infertile women who underwent successful IVF-ET treatment. The donors had no history of any particular diseases, and no evidence of infection such as syphilis, hepatitis, or adult T cell leukemia was found by laboratory examinations. All donor sera were heated at 56°C for 30 minutes to inactivate human immunodeficiency virus and complement Fertility and Sterility
Table 1 Characteristics of Patients With or Without Antisperm Antibodies Antisperm group Comparison group (n = 18) (n = 122) Cycles (oocyte recovery procedure) Age* Duration of infertility (year)*
35 35.0 ± 2.7t
273 32.6 ± 3.6
9.4 ± 3.5:j:
7.7 ± 3.3
*Values are means± SD. t p < 0.01. :j: p < 0.001.
and used as mixtures with T6 medium at 7.5% (volj vol) for insemination and 15% for embryo culture. The husband's sperm was washed twice with T6 medium with 7.5% serum by centrifugation at 200 X g for 5 minutes. The supernatant was then decanted and 1 mL of medium was layered onto the sperm pellet. After incubation for 1 hour at 37°C under 5% oxygen, 5% carbon dioxide, and 90% nitrogen, motile sperm swimming up into the upper T6 medium were collected and resuspended in T6 medium and used for insemination. Mature oocytes were inseminated with 50,000 motile sperm per oocyte 6 hours after oocyte retrieval and incubated in 1 mL of T6 medium under the humidified atmosphere described above.
lization test is judged as positive. The ratio of sperm motility in the control serum to that in the test serum is calculated as the sperm immobilization value. Serum samples from five women had consistently high antibody titer (sperm immobilization value = oo, the high titer group). In the remaining 13 women, the sperm immobilization value fluctuated but did not reach oo (the low antibody group). Statistical Analysis
Data were analyzed statistically by Mann Whitney's U test, x2 test, and Fisher's direct probability test. RESULTS
Table 1 summarizes characteristics of the two groups. The antisperm group was significantly older than the comparison group (35.0 ± 2.7 versus 32.6 ± 3.6 years; mean ± SD, P < 0.01) and had a longer history of infertility (9.4 ± 3.5 versus 7.7 ± 3.3 years; P < 0.001). Table 2 shows the outcomes of the IVF-ET proTable 2 Results of IVF-ET using GnRH-a in a Long Protocol for Ovarian Stimulation Antisperm group* (n = 18)
Embryo Growth and Transfer
Inseminated oocytes were examined under a dissecting microscope approximately 16 hours after insemination and the presence of two pronuclei was taken as evidence of normal fertilization. Fertilized oocytes were transferred to growth medium, and their cleavage was examined 24 hours later. Healthy embryos were replaced transcervically in the uterus under US guidance. Patients were given 1,500 IU of hCG on the 2nd, 4th, 6th, and 8th day after ET. Sperm-Immobilization Test
Antisperm activity of sera was measured by the method of Isojima et al. (20). In this test, a mixture of0.025 mL of washed sperm suspension (40 X 106 / mL), 0.05 mL of guinea pig serum as complement and 0.25 mL of test serum is incubated at 32°C for 60 minutes, and then the proportion of motile sperm is estimated under a microscope. When the percentage of motile sperm in the patient's serum is <50% of that in control serum, the sperm immobiVol. 63, No. 1, January 1995
No. of cycles per patient Mature ova No. of mature ova per cycle Fertilized ova§ Developed ova11 Transferred embryos No. of implanted embryos Implantation rate per embryo transferred(%) Pregnancies** Pregnancy rate per cycle(%) Modified pregnancy rate (%)tt Abortion rate
1.9 ± LOt 217 6.2 ± 3.6:j: 133 (61.3) II 144 (85.7):j: 85 20 23.511 13 (3)
Comparison group* (n = 122) 2.2 ± 1.7 1,661 6.1 ± 3.7 1,265 (76.8) 1,057 (83.5) 749 59 7.9 49 (9)
37.1
17.9
34.4 (11/32):j::j: 23.lt
17.8 (44/247) 18.4
* Couples with a male factor and unknown etiology were excluded in this survey. t Not significant. :j: Values are means± SD. § Values in parentheses are percentage of mature ova. II P < 0.01. 11 Values in parentheses are percentage of fertilized ova. ** Total number of pregnancies with number of abortions in parentheses. tt Values in parentheses are number of first pregnancies divided by number of cycles that were tried until achievement of the first pregnancy. :j::j: p < 0.05. Daitoh et al.
IVF-ET on women with antisperm antibody
89
Table 3 Results of IVF-ET Treatment of Women with Sperm-Immobilizing Antibodies Using Donor Serum
No. of patients No. of cycles Mature ova collected Fertilized ova:j: Developed ova II Pregnancies1f Pregnancy rate per cycle(%) Modified pregnancy rate(%)**
High titer group*
Low titer groupt
5 9 63 44 (69.8) 38 (86.4)
13 26 154 89 (57.8)§ 76 (85.4)§
4 (2)
9 (1)
44.4 33.3 (2/6)§
34.6 34.6 (9/26)
*Sperm immobilization value was consistently high (oo). t Sperm immobilization value was fluctuating but did not reach oo. II Values in parentheses are percentage of fertilized ova. § Not significant. :j: Values in parentheses are percentage of mature ova. 1f Total number of pregnancies with number of abortions in parentheses. Spontaneous abortion occured twice in the same patient with a high titer of antisperm antibodies. **Values in parentheses are number of first pregnancies divided by number of cycles that were tried until achievement of the first pregnancy.
gram in the two groups. In the antisperm group, in which donor serum was added to the culture medium instead of patient's serum, the fertilization rate (61.3%) was significantly lower than that in the comparison group (76.8%, P < 0.01), but the rates of development to cleaved ova were similar in the two groups (85.7% in the antisperm group versus 83.5% in the comparison group). The implantation rate per embryo transferred (23.5%) and consequently the multiple pregnancy rate per embryo transfer (~3 embryos) cycle (25.0%) were significantly (P < 0.01) higher in the antisperm group than in the comparison group (7.9% and 3.3%, respectively). The modified PR in the antisperm group was also significantly higher than that in the comparison group (34.4% versus 17.8%, P < 0.05). To avoid misassessment by entry of cases who achieved pregnancy more than one cycle, modified PRs in both groups were calculated by number of first pregnancies divided by number of oocyte recovery cycles that were tried until achievement of the first pregnancy. The results of IVF-ET treatment were not affected by antibody titer (Table 3). Spontaneous abortion occurred in three cycles of two patients with antisperm antibodies, but the abortion rate (23.1 %, 3 of 13) was similar to that (18.4%, 9 of 49) of the comparison group (Table 2). In the case in which miscarriage occurred twice (the 1st and 3rd IVF -ET treatment), the antibody activity of the serum was high (the sperm immobiliza90
Daitoh et al.
IVF-ET on women with antisperm antibody
tion value was consistently >100). However, conception was successful on a 4th series of IVF-ET therapy resulted in delivery of a normal healthy infant. Including this infant, none of the 16 infants delivered showed any abnormalities. DISCUSSION
Previously, in a human zona penetration test with human ZP matured in vitro, we found that sperm-immobilizing antibodies blocked fertilization and suggested that in IVF-ET treatment, donor serum should be used instead of the serum of immunologically infertile women (1). This IVF-ET procedure can overcome the blocking effect of antisperm antibodies on fertilization, but would not overcome implantation failure of fertilized eggs or miscarriage if such antibodies or maternal immunities against sperm have cytotoxic effects on the embryo. In fact, cytotoxic effects of antisperm antibodies on fertilized eggs have been observed in animal models (7, 8). Furthermore, there are reports suggesting an association between antisperm antibodies and recurrent spontaneous abortion (10, 11). However, it is unlikely that sperm-immobilizing antibodies are cytotoxic to fertilized eggs, at least in humans. This conclusion is supported by previous reports of others (3-6) and by the present findings that the rate of miscarriage in women with antisperm antibodies was not high and no abnormalities were observed in any of the 16 infants conceived after IVF-ET therapy. Contrary to our expectation of possible risk such as inhibition of nidation or of fetal growth, a very high PR per cycle and implantation rate per embryo transferred were demonstrated in the antisperm group, although this group was older and had a longer history of infertility than the comparison group. The high implantation rate in the antisperm group seems to be related primarily to maternal receptivity of the fetus, not to the quality of the embryo because only mature ova with normal morphology were inseminated in both groups. In addition, the mean numbers of matured ova collected per cycle and the cleavage rates were similar in the two groups. Because the low level of spermimmobilizing antibodies remaining even after washing the oocyte may affect sperm-egg interaction, it is difficult to compare the fertilization rates in the two groups, but the fertilization rate in the antisperm group was rather lower than that in the comparison group. It is unlikely that sperm-immobilizing antibodies was a direct cause of favorable Fertility and Sterility
outcomes of IVF-ET treatment in the anti-sperm group, because the implantation rates and PRs were similar in the high and low antibody groups. These results strongly suggest not only that antisperm antibodies are not cytotoxic to embryos, but also that the presence of a cellular and/or humoral immune reaction against sperm antigen(s) established in women in whom such antibodies are detectable, is advantageous to achieve pregnancy. This possibility is supported by the findings that a maternal response to inseminated sperm increased the implantation rate and subsequent PR in IVF-ET (21) and GIFT (22). Namely, as described in introduction, maternal immune reactions against sperm antigen(s) on the surface of embryo or trophoblast may be favorable. Thaler et al. (23) suggested that the Fe receptor for immunoglobulin G (FeR) in seminal plasma was located on the trophoblast and could play a role in regulating maternal immunity after insemination. In fact, various FeR-related proteins have been found m human seminal plasma (23-25). Acknowledgments. We thank Nanae Furumoto, Ph.D., Department of Data Processing and Statistics, Tokushima Bunri University, Tokushima, Japan, for statistical analyses.
REFERENCES 1. Kamada M, Daitoh T, Hasebe H, Irahara M, Yamano S, Mori T. Blocking of human fertilization in vitro by sera with sperm immobilizing antibodies. Am J Obstet Gynecol 1985;153:328-31. 2. Bandoh R, Yamano S, Kamada M, Daitoh T, Aono T. Effect of sperm-immobilizing antibodies on the acrosome reaction of human spermatozoa. Fertil Steril 1992;57:387-92. 3. Ackermann SB, Graff D, van Uem JFHM, Swanson RJ, Veeck LL, Acosta AA, eta!. Immunologic infertility and in vitro fertilization. Fertil Steril1984;42:474-7. 4. Yovich J, Stanger JD, Kay D, Boettcher B. In-vitro fertilization of oocytes from women with serum antisperm antibodies. Lancet 1984;18:369-70. 5. Pexieder T, Boillat E, Janecek P. Antisperm antibodies and in vitro fertilization failure. J In Vitro Fertil Embryo Transfer 1985;2:229-32. 6. Clark GN, Lopata A, Johnston WIH. Effect of sperm antibodies in females on human in vitro fertilization. Fertil Steril 1986;46:435-41. 7. Menge AC. Effect of isoimmunization and isoantisera against seminal antigens on fertility processes in female rabbits. Bioi Reprod 1971;4:137-44. 8. Koyama K, Hasegawa A, Isojima S. Effect of anti-sperm antibody on the in vitro development of rat embryos. Gamete Res 1984;10:143-52.
Vol. 63, No. 1, January 1995
9. Menge AC, Fleming CH. Detection of sperm antigens on mouse ova and early embryos. Dev Biol1978;63:111-7. 10. Menge AC, Medley NE, Mangione CM, Dietrich JW. The incidence and influence of antisperm antibodies in infertile human couples on sperm-cervical mucus interaction and subsequent fertility. Fertil Steril 1982;38:439-46. 11. Witkin SS, Chaudry A. Association between recurrent spontaneous abortions and circulating lgG antibodies to sperm tails in women. J Reprod lmmunol1989;15:151-8. 12. Clarke GN, Baker HWG. Lack of association between sperm antibodies and recurrent spontaneous abortion. Fertil Steril 1993;59:463-4. 13. Faulk WP, Coulam CB, Mcintyre JA. Reproductive immunology: Biomarkers of compromised pregnancies. Environmental Health Perspectives 1987;74:119-27. 14. Wegmann TG. Placental immunotrophism: Maternal T cells enhance placental growth and function. Am J Reprod Immunol Microbiol1987;15:67-70. 15. Tsuji Y, Fukuda H, Iuchi A, Ishizuka I, Isojima S. Sperm immobilizing antibodies react to the 3-0 sulfated galactose residue of seminolipid on human sperm. J Reprod lmmunol 1992;22:225-36. 16. Purcell DFJ, McKenzie IFC, Lublin DM, Johnson PM, Atkinson JP, Oglesby TJ, eta!. The human cell-surface glycoprotein Huly-m5, membrane co-factor protein (MCP) of the complement system, and trophoblast leukocyte-common (TLX) antigen, are CD46. Immunology 1990;70:155-61. 17. World Health Organization. WHO laboratory manual for the examination of human semen and semen-cervical mucus interaction. Cambridge: The Press Syndicate of the University of Cambridge, 1987:27. 18. Kamada M, Hasebe H, Irahara M, Kinoshita T, Naka 0, Mori T. Detection of anti-zona pellucida activities in human sera by the passive hemagglutination reaction. Fertil Steril1984;41:901-6. 19. Kamada M, Daitoh T, Mori K, Maeda N, Hirano K, Irahara M, et a!. Etiological implication of autoantibodies to zona pellucida in human female infertility. Am J Reprod Immunol 1992;28:104-9. 20. Isojima S, Tsuchiya K, Koyama K, Naka 0, Adachi H. Further studies on sperm-immobilizing antibody found in sera of unexplained cases of sterility in women. Am J Obstet Gynecol 1972;112:199-207. 21. Bellinge BS, Copeland CM, Thomas TD, Mazzucchelli RE, O'Neil G, Cohen MJ. The influence of patient insemination on the implantation rate in an in vitro fertilization and embryo transfer program. Fertil Steril 1986;46:252-6. 22. Marconi G, Auge L, Oses R, QuintanaR, Raffo F, Young E. Does sexual intercourse improve pregnancy rates in gamete intrafallopian transfer? Fertil Steril1989;51:357-8. 23. Thaler CJ, Mcintyre JA, Faulk WP. Fe receptor and trophoblast antigens in seminal plasma: a potential stimulus to prime mothers for implantation. Immunol Lett 1990;26: 145-52. 24. Witkin SS, RichardJM, Bongiovanni AN, Zelikovski G. An lgG-Fc binding protein in seminal plasma. Am J Reprod lmmunol1983;3:23-7. 25. Kamada M, Liang ZG, Koide SS. Identification of lgG and Fc-binding proteins in human seminal plasma and sperm. Arch Androl1991;27:1-7.
Daitoh et al.
IVF-ET on women with antisperm antibody
91
Vol. 63, No.1, January 1995
Copyright© 1995 American Society for Reproductive Medicine
Printed on acid-free paper in U. S. A.
High implantation rate and consequently high pregnancy rate by in vitro fertilization-embryo transfer treatment in infertile women with antisperm antibody*
Toshifumi Daitoh, Ph.D. Masaharu Kamada, Ph.D.t Syuji Yamano, Ph.D. Shinji Murayama, M.D.
Tsuzuki Kobayashi, M.D. Masahiko Maegawa, Ph.D. Toshihiro Aono, Ph.D.
Department of Obstetrics and Gynecology, The University of Tokushima, School of Medicine, Tokushima, Japan
Objective: To examine the effect of antisperm immunity on postfertilization steps, such as implantation of embryos and fetal growth, in IVF-ET treatment of women with sperm-immobilizing antibodies. Design: Retrospective analysis of clinical laboratory data. Setting: The IVF-ET program of the Department of Obstetrics and Gynecology, The University of Tokushima, School of Medicine. Patients: Eighteen women with sperm-immobilizing antibodies and 122 infertile patients with nonimmune etiology as controls. Infertile couples due to a male factor and with unknown etiology were excluded. Interventions: All patients received the same IVF-ET program with GnRH agonist. Main Outcome Measures: Rates of fertilization and cleavage, implantation rate per embryo transferred and pregnancy rate (PR) in both test and comparison groups. Results: The rate of fertilization in the antisperm group (61.3%) was significantly lower than that in the comparison group (76.8%). But the implantation rate per embryo transferred (23.5%) and consequently the modified PR per oocyte recovery procedure (34.4%) in immunologically infertile women were significantly higher than those in the comparison group (7.9% and 17.8%, respectively). Conclusions: Although sperm-immobilizing antibodies prevent sperm-egg interaction, they do not seem to have any adverse effects on achievement of pregnancy. Moreover, the existence of antisperm immunity in woman with antisperm antibodies is suggested to be favorable for successful pregnancy by the IVF-ET procedure. Fertil Steril1995;63:87-91 Key Words: antisperm antibody, infertility, in vitro fertilization, implantation, immunology of pregnancy
We have reported that sperm-immobilizing antibody causes infertility by inhibiting the acrosome Received September 13, 1993; revised and accepted July 22, 1994. * Supported in part by a Grant-in-Aid for Scientific Research (No. 04671005) from the Ministry of Education, Science, and Culture of Japan, Kasumigaseki, Chiyoda-Ku, Tokyo, Japan. t Reprint requests: Masaharu Kamada, M.D., Ph.D., Department of Obstetrics and Gynecology, The University of Tokushima, School of Medicine, 3-18-15 Kuramoto-cho, Tokushima 770, Japan (FAX: 81-0886-31-2630). Vol. 63, No. 1, January 1995
reaction of sperm (1, 2). The inhibitory effect is reversible by incubation of the antibody-bound sperm in antibody-free medium (2) and there was a successful case in IVF using fetal cord serum admixtured to the medium instead of the patient's serum containing sperm-immobilizing antibodies. Therefore, we proposed that circulating antisperm antibodies in the follicular or tubal fluid may cause infertility by blocking fertilization and that IVF is a suitable measure for achieving pregnancy in patients with antisperm antibodies (1). In fact, since Daitoh et al.
IVF-ET on women with antisperm antibody
87
the first report by Ackermann et al. (3), IVF with donor serum has been widely used for the treatment of patients with immunological infertility, and some successful pregnancy cases have been reported (4-6). Another possible mechanism by which antisperm antibodies cause infertility or pregnancy failure is proposed from studies in animal models to be by their cytotoxic effects on the fertilized eggs (7, 8), because sperm antigens have been demonstrated on the surface of fertilized ova (9). Furthermore, there have been some clinical reports suggesting the relationship between a presence of antisperm antibodies and recurrent spontaneous abortion (10, 11). However, the above discussion is still controversial (12). On the contrary, maternal immune reactions to fetal antigens are known to be mandatory for maintenance of pregnancy by preventing rejection ofthe alloimmunogenic fetus or by enhancing embryo growth (13, 14). The sperm antigen(s) on the surface of the embryo (9) or trophoblast (15) may elicit favorable maternal immune responses for successful implantation of a semiallograft into the uterus. Indeed, there is a recent report (16) that the trophoblast-lymphocyte cross-reactive (TLX) antigen, which has been suggested to induce a maternal immune response and to be important for maintenance of pregnancy (13), is one of these sperm antigens. To determine the possible effects of antisperm immunity on nidation or embryo growth in women with sperm-immobilizing antibodies, we examined the outcome of IVF-ET treatment in immunologically infertile women and, as the control, infertile women due to tubal factor.
MATERIALS AND METHODS
pared to those in a comparison group, namely, 273 oocyte-recovery cycles of 122 couples, whose indication for IVF-ET treatment was tubal obstruction but no immunological abnormalities. No cases showing poor quality of the husband's semen according to the criteria of the World Health Organization (17) were included. We also excluded infertile couples with unknown etiology and cases in which fertilization failure was due to anti-zona pellucida (ZP) activity in the wife's serum (18, 19) or antisperm activity (sperm immobilization) in the husband's serum. During the same 5-year period, IVF-ET procedures such as egg retrieval, culture of gametes, and ET have been performed in order of registration by the same clinical group. Ovarian Stimulation and Follicle Development
All patients were treated with GnRH agonist (GnRH-a, buserelin acetate, Hoechst Japan, Tokyo, Japan) beginning 8 days before the estimated 1st day of the menstrual cycle. From the 3rd day of menstruation or the 14th day from the start of GnRH-a administration when menstruation was missed, 225 IU of hMG (HMG Nikken; Nikken, Tokyo, Japan) or pure FSH (Fertinorm-P; Serono Japan, Tokyo, Japan) was injected daily for 4 days and subsequently 150 IU of hMG (Humegon, Nippon Organon, Tokyo, Japan) was administered daily until the mean diameter of the dominant follicle reached 18 mm as determined by ultrasonography (RT3600; Yokogawa, Tokyo, Japan). Ovulation was triggered by injection of 5,000 IU of hCG (Gonatropin; Teikoku Zohki, Tokyo, Japan) 50 hours after the last injection of hMG, and oocytes were retrieved transvaginally 35 hours later under local anesthesia using a needle-guided technique under ultrasound (US) control. Fertilization and Embryo Culture
In Vitro Fertilization-Embryo Transfer
Patients
In vitro fertilization-embryo transfer therapy was carried out from February 1989 through June 1993 on 35 oocyte recovery cycles of 18 women whose sera showed sperm-immobilizing activity (the antisperm group) at the Infertility Clinic of the Department of Obstetrics and Gynecology, Tokushima University Hospital. The rates of fertilization, cleavage, implantation, and pregnancy in the antisperm group were com88
Daitoh et al.
IVF-ET on women with antisperm antibody
Whittingham's T6 (T6) medium supplemented with donor serum was used for insemination and embryo culture in vitro. Donor sera were separated from blood of umbilical vein of fetus delivered from normal healthy mothers or nonimmunologically infertile women who underwent successful IVF-ET treatment. The donors had no history of any particular diseases, and no evidence of infection such as syphilis, hepatitis, or adult T cell leukemia was found by laboratory examinations. All donor sera were heated at 56°C for 30 minutes to inactivate human immunodeficiency virus and complement Fertility and Sterility
Table 1 Characteristics of Patients With or Without Antisperm Antibodies Antisperm group Comparison group (n = 18) (n = 122) Cycles (oocyte recovery procedure) Age* Duration of infertility (year)*
35 35.0 ± 2.7t
273 32.6 ± 3.6
9.4 ± 3.5:j:
7.7 ± 3.3
*Values are means± SD. t p < 0.01. :j: p < 0.001.
and used as mixtures with T6 medium at 7.5% (volj vol) for insemination and 15% for embryo culture. The husband's sperm was washed twice with T6 medium with 7.5% serum by centrifugation at 200 X g for 5 minutes. The supernatant was then decanted and 1 mL of medium was layered onto the sperm pellet. After incubation for 1 hour at 37°C under 5% oxygen, 5% carbon dioxide, and 90% nitrogen, motile sperm swimming up into the upper T6 medium were collected and resuspended in T6 medium and used for insemination. Mature oocytes were inseminated with 50,000 motile sperm per oocyte 6 hours after oocyte retrieval and incubated in 1 mL of T6 medium under the humidified atmosphere described above.
lization test is judged as positive. The ratio of sperm motility in the control serum to that in the test serum is calculated as the sperm immobilization value. Serum samples from five women had consistently high antibody titer (sperm immobilization value = oo, the high titer group). In the remaining 13 women, the sperm immobilization value fluctuated but did not reach oo (the low antibody group). Statistical Analysis
Data were analyzed statistically by Mann Whitney's U test, x2 test, and Fisher's direct probability test. RESULTS
Table 1 summarizes characteristics of the two groups. The antisperm group was significantly older than the comparison group (35.0 ± 2.7 versus 32.6 ± 3.6 years; mean ± SD, P < 0.01) and had a longer history of infertility (9.4 ± 3.5 versus 7.7 ± 3.3 years; P < 0.001). Table 2 shows the outcomes of the IVF-ET proTable 2 Results of IVF-ET using GnRH-a in a Long Protocol for Ovarian Stimulation Antisperm group* (n = 18)
Embryo Growth and Transfer
Inseminated oocytes were examined under a dissecting microscope approximately 16 hours after insemination and the presence of two pronuclei was taken as evidence of normal fertilization. Fertilized oocytes were transferred to growth medium, and their cleavage was examined 24 hours later. Healthy embryos were replaced transcervically in the uterus under US guidance. Patients were given 1,500 IU of hCG on the 2nd, 4th, 6th, and 8th day after ET. Sperm-Immobilization Test
Antisperm activity of sera was measured by the method of Isojima et al. (20). In this test, a mixture of0.025 mL of washed sperm suspension (40 X 106 / mL), 0.05 mL of guinea pig serum as complement and 0.25 mL of test serum is incubated at 32°C for 60 minutes, and then the proportion of motile sperm is estimated under a microscope. When the percentage of motile sperm in the patient's serum is <50% of that in control serum, the sperm immobiVol. 63, No. 1, January 1995
No. of cycles per patient Mature ova No. of mature ova per cycle Fertilized ova§ Developed ova11 Transferred embryos No. of implanted embryos Implantation rate per embryo transferred(%) Pregnancies** Pregnancy rate per cycle(%) Modified pregnancy rate (%)tt Abortion rate
1.9 ± LOt 217 6.2 ± 3.6:j: 133 (61.3) II 144 (85.7):j: 85 20 23.511 13 (3)
Comparison group* (n = 122) 2.2 ± 1.7 1,661 6.1 ± 3.7 1,265 (76.8) 1,057 (83.5) 749 59 7.9 49 (9)
37.1
17.9
34.4 (11/32):j::j: 23.lt
17.8 (44/247) 18.4
* Couples with a male factor and unknown etiology were excluded in this survey. t Not significant. :j: Values are means± SD. § Values in parentheses are percentage of mature ova. II P < 0.01. 11 Values in parentheses are percentage of fertilized ova. ** Total number of pregnancies with number of abortions in parentheses. tt Values in parentheses are number of first pregnancies divided by number of cycles that were tried until achievement of the first pregnancy. :j::j: p < 0.05. Daitoh et al.
IVF-ET on women with antisperm antibody
89
Table 3 Results of IVF-ET Treatment of Women with Sperm-Immobilizing Antibodies Using Donor Serum
No. of patients No. of cycles Mature ova collected Fertilized ova:j: Developed ova II Pregnancies1f Pregnancy rate per cycle(%) Modified pregnancy rate(%)**
High titer group*
Low titer groupt
5 9 63 44 (69.8) 38 (86.4)
13 26 154 89 (57.8)§ 76 (85.4)§
4 (2)
9 (1)
44.4 33.3 (2/6)§
34.6 34.6 (9/26)
*Sperm immobilization value was consistently high (oo). t Sperm immobilization value was fluctuating but did not reach oo. II Values in parentheses are percentage of fertilized ova. § Not significant. :j: Values in parentheses are percentage of mature ova. 1f Total number of pregnancies with number of abortions in parentheses. Spontaneous abortion occured twice in the same patient with a high titer of antisperm antibodies. **Values in parentheses are number of first pregnancies divided by number of cycles that were tried until achievement of the first pregnancy.
gram in the two groups. In the antisperm group, in which donor serum was added to the culture medium instead of patient's serum, the fertilization rate (61.3%) was significantly lower than that in the comparison group (76.8%, P < 0.01), but the rates of development to cleaved ova were similar in the two groups (85.7% in the antisperm group versus 83.5% in the comparison group). The implantation rate per embryo transferred (23.5%) and consequently the multiple pregnancy rate per embryo transfer (~3 embryos) cycle (25.0%) were significantly (P < 0.01) higher in the antisperm group than in the comparison group (7.9% and 3.3%, respectively). The modified PR in the antisperm group was also significantly higher than that in the comparison group (34.4% versus 17.8%, P < 0.05). To avoid misassessment by entry of cases who achieved pregnancy more than one cycle, modified PRs in both groups were calculated by number of first pregnancies divided by number of oocyte recovery cycles that were tried until achievement of the first pregnancy. The results of IVF-ET treatment were not affected by antibody titer (Table 3). Spontaneous abortion occurred in three cycles of two patients with antisperm antibodies, but the abortion rate (23.1 %, 3 of 13) was similar to that (18.4%, 9 of 49) of the comparison group (Table 2). In the case in which miscarriage occurred twice (the 1st and 3rd IVF -ET treatment), the antibody activity of the serum was high (the sperm immobiliza90
Daitoh et al.
IVF-ET on women with antisperm antibody
tion value was consistently >100). However, conception was successful on a 4th series of IVF-ET therapy resulted in delivery of a normal healthy infant. Including this infant, none of the 16 infants delivered showed any abnormalities. DISCUSSION
Previously, in a human zona penetration test with human ZP matured in vitro, we found that sperm-immobilizing antibodies blocked fertilization and suggested that in IVF-ET treatment, donor serum should be used instead of the serum of immunologically infertile women (1). This IVF-ET procedure can overcome the blocking effect of antisperm antibodies on fertilization, but would not overcome implantation failure of fertilized eggs or miscarriage if such antibodies or maternal immunities against sperm have cytotoxic effects on the embryo. In fact, cytotoxic effects of antisperm antibodies on fertilized eggs have been observed in animal models (7, 8). Furthermore, there are reports suggesting an association between antisperm antibodies and recurrent spontaneous abortion (10, 11). However, it is unlikely that sperm-immobilizing antibodies are cytotoxic to fertilized eggs, at least in humans. This conclusion is supported by previous reports of others (3-6) and by the present findings that the rate of miscarriage in women with antisperm antibodies was not high and no abnormalities were observed in any of the 16 infants conceived after IVF-ET therapy. Contrary to our expectation of possible risk such as inhibition of nidation or of fetal growth, a very high PR per cycle and implantation rate per embryo transferred were demonstrated in the antisperm group, although this group was older and had a longer history of infertility than the comparison group. The high implantation rate in the antisperm group seems to be related primarily to maternal receptivity of the fetus, not to the quality of the embryo because only mature ova with normal morphology were inseminated in both groups. In addition, the mean numbers of matured ova collected per cycle and the cleavage rates were similar in the two groups. Because the low level of spermimmobilizing antibodies remaining even after washing the oocyte may affect sperm-egg interaction, it is difficult to compare the fertilization rates in the two groups, but the fertilization rate in the antisperm group was rather lower than that in the comparison group. It is unlikely that sperm-immobilizing antibodies was a direct cause of favorable Fertility and Sterility
outcomes of IVF-ET treatment in the anti-sperm group, because the implantation rates and PRs were similar in the high and low antibody groups. These results strongly suggest not only that antisperm antibodies are not cytotoxic to embryos, but also that the presence of a cellular and/or humoral immune reaction against sperm antigen(s) established in women in whom such antibodies are detectable, is advantageous to achieve pregnancy. This possibility is supported by the findings that a maternal response to inseminated sperm increased the implantation rate and subsequent PR in IVF-ET (21) and GIFT (22). Namely, as described in introduction, maternal immune reactions against sperm antigen(s) on the surface of embryo or trophoblast may be favorable. Thaler et al. (23) suggested that the Fe receptor for immunoglobulin G (FeR) in seminal plasma was located on the trophoblast and could play a role in regulating maternal immunity after insemination. In fact, various FeR-related proteins have been found m human seminal plasma (23-25). Acknowledgments. We thank Nanae Furumoto, Ph.D., Department of Data Processing and Statistics, Tokushima Bunri University, Tokushima, Japan, for statistical analyses.
REFERENCES 1. Kamada M, Daitoh T, Hasebe H, Irahara M, Yamano S, Mori T. Blocking of human fertilization in vitro by sera with sperm immobilizing antibodies. Am J Obstet Gynecol 1985;153:328-31. 2. Bandoh R, Yamano S, Kamada M, Daitoh T, Aono T. Effect of sperm-immobilizing antibodies on the acrosome reaction of human spermatozoa. Fertil Steril 1992;57:387-92. 3. Ackermann SB, Graff D, van Uem JFHM, Swanson RJ, Veeck LL, Acosta AA, eta!. Immunologic infertility and in vitro fertilization. Fertil Steril1984;42:474-7. 4. Yovich J, Stanger JD, Kay D, Boettcher B. In-vitro fertilization of oocytes from women with serum antisperm antibodies. Lancet 1984;18:369-70. 5. Pexieder T, Boillat E, Janecek P. Antisperm antibodies and in vitro fertilization failure. J In Vitro Fertil Embryo Transfer 1985;2:229-32. 6. Clark GN, Lopata A, Johnston WIH. Effect of sperm antibodies in females on human in vitro fertilization. Fertil Steril 1986;46:435-41. 7. Menge AC. Effect of isoimmunization and isoantisera against seminal antigens on fertility processes in female rabbits. Bioi Reprod 1971;4:137-44. 8. Koyama K, Hasegawa A, Isojima S. Effect of anti-sperm antibody on the in vitro development of rat embryos. Gamete Res 1984;10:143-52.
Vol. 63, No. 1, January 1995
9. Menge AC, Fleming CH. Detection of sperm antigens on mouse ova and early embryos. Dev Biol1978;63:111-7. 10. Menge AC, Medley NE, Mangione CM, Dietrich JW. The incidence and influence of antisperm antibodies in infertile human couples on sperm-cervical mucus interaction and subsequent fertility. Fertil Steril 1982;38:439-46. 11. Witkin SS, Chaudry A. Association between recurrent spontaneous abortions and circulating lgG antibodies to sperm tails in women. J Reprod lmmunol1989;15:151-8. 12. Clarke GN, Baker HWG. Lack of association between sperm antibodies and recurrent spontaneous abortion. Fertil Steril 1993;59:463-4. 13. Faulk WP, Coulam CB, Mcintyre JA. Reproductive immunology: Biomarkers of compromised pregnancies. Environmental Health Perspectives 1987;74:119-27. 14. Wegmann TG. Placental immunotrophism: Maternal T cells enhance placental growth and function. Am J Reprod Immunol Microbiol1987;15:67-70. 15. Tsuji Y, Fukuda H, Iuchi A, Ishizuka I, Isojima S. Sperm immobilizing antibodies react to the 3-0 sulfated galactose residue of seminolipid on human sperm. J Reprod lmmunol 1992;22:225-36. 16. Purcell DFJ, McKenzie IFC, Lublin DM, Johnson PM, Atkinson JP, Oglesby TJ, eta!. The human cell-surface glycoprotein Huly-m5, membrane co-factor protein (MCP) of the complement system, and trophoblast leukocyte-common (TLX) antigen, are CD46. Immunology 1990;70:155-61. 17. World Health Organization. WHO laboratory manual for the examination of human semen and semen-cervical mucus interaction. Cambridge: The Press Syndicate of the University of Cambridge, 1987:27. 18. Kamada M, Hasebe H, Irahara M, Kinoshita T, Naka 0, Mori T. Detection of anti-zona pellucida activities in human sera by the passive hemagglutination reaction. Fertil Steril1984;41:901-6. 19. Kamada M, Daitoh T, Mori K, Maeda N, Hirano K, Irahara M, et a!. Etiological implication of autoantibodies to zona pellucida in human female infertility. Am J Reprod Immunol 1992;28:104-9. 20. Isojima S, Tsuchiya K, Koyama K, Naka 0, Adachi H. Further studies on sperm-immobilizing antibody found in sera of unexplained cases of sterility in women. Am J Obstet Gynecol 1972;112:199-207. 21. Bellinge BS, Copeland CM, Thomas TD, Mazzucchelli RE, O'Neil G, Cohen MJ. The influence of patient insemination on the implantation rate in an in vitro fertilization and embryo transfer program. Fertil Steril 1986;46:252-6. 22. Marconi G, Auge L, Oses R, QuintanaR, Raffo F, Young E. Does sexual intercourse improve pregnancy rates in gamete intrafallopian transfer? Fertil Steril1989;51:357-8. 23. Thaler CJ, Mcintyre JA, Faulk WP. Fe receptor and trophoblast antigens in seminal plasma: a potential stimulus to prime mothers for implantation. Immunol Lett 1990;26: 145-52. 24. Witkin SS, RichardJM, Bongiovanni AN, Zelikovski G. An lgG-Fc binding protein in seminal plasma. Am J Reprod lmmunol1983;3:23-7. 25. Kamada M, Liang ZG, Koide SS. Identification of lgG and Fc-binding proteins in human seminal plasma and sperm. Arch Androl1991;27:1-7.
Daitoh et al.
IVF-ET on women with antisperm antibody
91