Cryopreservation of cleaving embryos and expanded blastocysts in the human: a comparative study

Transcription

1 FERTILITY AND STERILITY Copyright" 1985 The American Fertility Society Vol. 44, No.5, November 1985 Printed in U.s.A. Cryopreservation of cleaving embryos and expanded blastocysts in the human: a comparative study Carole B. Fehilly, Ph.D.* Jacques Cohen, Ph.D.*t Roger F. Simons* Simon B. Fishel, Ph.D.* Robert G. Edwards, Ph.D., D.Sc.*:j: Bourn Hall Clinic, Bourn, and Physiological Laboratory, Cambridge University, Cambridge, United Kingdom The survival and implantation capacity of cryopreserved cleaving (5-cell to 10-cell) human embryos and expanded blastocysts was compared. Twice as many cleaving embryos were frozen as were expanding blastocysts because of the low developmental potential of human embryos in vitro. However, significantly more expanded blastocysts survived cryopreservation than cleaving embryos, and relatively more pregnancies were established by the replacement of thawed blastocysts than by the replacement of thawed cleaving embryos. Cleaving embryos from 26 women were thawed; 17 had thawed embryos replaced, and 4 subsequently became pregnant. Expanded blastocysts were thawed from 23 other women; 15 had thawed blastocysts replaced, and 8 subsequently became pregnant. The pregnancy of one patient in each group aborted; both patients were over 40 years of age. It is estimated that by maintaining the current policy of replacing three fresh embryos and freezing any remaining embryos when they reach blastocyst stage, the total incidence of pregnancy would increase by 3%. Fertil Steril 44:638, 1985 Pregnancies have been established in several clinics after the replacement of frozen-thawed cleaving embryosl,2 or expanded blastocysts. 3 Embryo freezing allows replacement during a cycle subsequent to the in vitro fertilization (lvf) cycle and avoids the pressure to replace large numbers of fresh embryos with the attendant risk of multipregnancy. It may also allow better synchrony between the embryo and the uterus, thus improving the chance of implantation. More Received March 19, 1985; revised and accepted August 12, *Bourn Hall Clinic. treprint requests: Jacques Cohen, Ph.D., Department of Physiology and Pharmacology, College of Veterinary Medicine, University of Georgia, Athens, Georgia :J:Physiological Laboratory, University of Cambridge. 638 Fehilly et a1. Cryopreservation of human embryos cleaving embryos are available for freezing than blastocysts because many embryos die during cleavage, compaction, and blastulation in vitro. Nevertheless, blastocysts may offer an advantage in freezing and also allow for better synchrony between the patient's uterus and the thawed replaced blastocyst, as practiced in farm animals. 4 The aim of the present report was (1) to assess the developmental potential of fresh embryos not replaced during the treatment cycle for IVF and (2) to compare the viability of cleaving embryos and blastocysts after cryopreservation. MATERIALS AND METHODS The work began after ethical guidelines were issued by the Royal College of Obstetricians and

2 Gynecologists, the Medical Research Council, the British Medical Association, and the Ethical Committee of Bourn Hall Clinic. The cryopreservation of human embryos was also accepted by the Warnock Committee. 5 IN VITRO FERTILIZATION CYCLE Follicular stimulation was induced by 100 mg of clomiphene citrate (CC; Serophene, Serono, Rome, Italy) on days 2 to 6, plus human menopausal gonadotropin (Serono) in variable dosage from day 5. Urinary estrogens and luteinizing hormone (LH) were assessed by methods described previously.6 Ultrasonography was also used in assessing follicular growth. Laparoscopy for oocyte recovery occurred in most patients 34 hours after 5000 IU human chorionic gonadotropin (hcg; Pregnyl, Organon, Oss, The Netherlands) was given to induce follicular maturation or 26 hours after the onset of an endogenous LH surge in the remainder. Details of the methods for laparoscopy, oocyte collection, insemination, and replacement of embryos are given in detail elsewhere.6 FROZEN EMBRYO REPLACEMENT CYCLE Patients who normally had regular menstrual cycles received no follicular stimulation. Patients with irregular menstrual cycles received 100 mg CC on days 2 to 6. All patients were monitored for urinary estrogens, LH, and plasma progesterone, and ultrasound assessments were made offollicular growth. All patients on a natural cycle had an endogenous LH surge, whereas in the CC-stimulated group, some patients had an endogenous LH surge and some received 5000 IU hcg. EMBRYO CULTURE Oocytes and embryos were cultured at 37 C in drops of Earle's medium containing sodium pyruvate and 15% inactivated human serum, under paraffin oil, with a gas phase of 5% CO2, 5% O2, and 90% N2. Embryos growing in culture were examined once or twice daily from fertilization onward with a Wild dissecting microscope (Wild, Heerbrugg, Switzerland) (magnification, x 60 to x 500) fitted with a light green filter. Embryos were assessed for cleavage, blastomere regularity, the presence and volume of cytoplasmic fragments, compaction, and blastulation. Cleaving embryos were not frozen if they were retarded more than 24 hours in their development and/or appeared degenerate and/or had cytoplasmic fragments constituting more than about 35% of the embryo's mass. Blastocysts were not frozen if they lacked a distinct inner cell mass and/or had several cavities instead of only one and/or appeared partially degenerate. After thawing, embryos were cultured under the same conditions as before freezing and examined at 2- to 3-hour intervals for assessment of their survival and growth. Survival of thawed cleaving embryos was confirmed by a general morphologic examination, continued cleavage, and the occurrence of compaction. Survival of thawed blastocysts was confirmed by the reexpansion of the blastocoele, general morphologic examination, and the absence of degenerate areas. CRYOPRESERVATION OF CLEAVING EMBRYOS (5- TO 10 CELL) The methods used for freezing and thawing cleaving embryos were similar to those described by Trounson and Mohr (1983)2 and Mohr et al. (1985), 7 the only difference being the use of IVF culture medium with 15% maternal serum instead of phosphate-buffered saline supplemented with 15% fetal calf serum. Embryos with between five and ten cells were frozen on day 3 or 4 after insemination. Dimethylsulfoxide (DMSO) was added at room temperature in four steps of 0.25, 0.5, 1.0, and 1.5 M for 10 minutes each, and the embryos were then transferred to sterile glass ampules with about 0.5 ml of culture medium containing 1.5 M DMSO at room temperature. The ampules were immediately sealed and cooled in a biologic cell freezer (Planer Products Ltd., Sundbury-on-Thames, UK) from room temperature to - 6 C at 2 C minute - \ when ice nucleation was induced with cold forceps, held at - 6 C for 20 minutes, and then slowly cooled at 0.3 C minute -1 to - 80 C. They were then transferred to liquid nitrogen. Cleaving embryos were thawed by placement of the ampules in the cell freezer at - 80 C and warming at a rate of 8 C minute - 1 to 1 C, where the temperature was held for 5 minutes. They were then warmed slowly to room temperature. DMSO was removed by six serial incubations for 10 minutes at room temperature with 1.5, 1.25, 1.0, 0.75, 0.5, and 0.25 M DMSO; then the embryos were washed in culture medium several Vol. 44, No.5, November 1985 Fehilly et al. Cryopreservation of human embryos 639

3 times and slowly wanned to 37 C. Thawed embryos were cultured between 2 and. 43 hours and replaced into their mothers only if at least 50% of the original blastomeresappeared morphologically intact. It is well-known 8 9 that an intact zona pellucida is thought to be necessary for the survival in vivo of the precompaction embryo, and cracks and holes are sometimes observed in the zona pellucida after thawing. Some embryos received prolonged culture in vitro (more than 8 hours) after thawing to allow the embryo to undergo compaction before transfer to the uterus; whether or not defects were actually observed in the zona pellucida. Cleaving embryos were thawed between 67 and 84 hours after the start of the urinary LH surge (day 2 after ovulation) or 88 to 91 hours after an injection of hcg. 2 CRYOPRESERVATION OF BLASTOCYSTS Blastocysts were classified as follows3: stage A, the blastocoele just appearing as one or more vacuoles; stage B, a distinct cavity equal to one quarter or one half the volume of the embryo, the diameter of the embryo being unchanged; stage C, a single cavity occupying most of the volume of the embryo, and the zona pellucida becoming thinner as the blastocyst expands; and stage 0, the blastocyst fully expanded for 24 hours or more. It was demonstrated previously that stage C and 0 blastocysts gave the best results,3 and expansion was therefore used as one of the main criteria when storing blastocysts. Most blastocysts were frozen on day 5 or 6 after insemination, with only a few on day 4 or 7. The method of freezing and thawing expanded blastocysts has been fully described elsewhere.3 Glycerol was added in steps (five steps, 1%, 2%, 4%, 6%, and 8% glycerol for 10 minutes each) at room temperature, and the ampules cooled to - 7 C at 1 C minute -1, then at C minute- 1 to - 36 C, followed by immediate cooling to -196 C. We thawed the blastocysts 118 to 132 hours after the urinary LH surge (day 4 after ovulation) or 132 to 135 hours after an injection of hcg3 by stirring the ampule in a water bath at 30 C. The cryoprotectant was then slowly removed at room temperature: 8% glycerol for 10 minutes, 6% for 12 minutes, 5% for 14 minutes, 4% for 16 minutes, 3% for 18 minutes, 2% for 20 minutes, and 1% for 20 minutes. The embryos were washed several times in culture medium and then slowly warmed to 37 C. Embryo replacement did not take place until the blastocyst reexpanded to some extent, which indicated its viability. The culture period varied from a few hours to overnight (2 to 20 hours). RESULTS The present results were obtained between February 1984 and February 1985, during which time a total of 1208 extra embryos were available from 432 patients who each had three fresh embryos replaced. All extra embryos were cultured in vitro and frozen if considered suitable. During the 12-month period of the study, there were three periods of approximately 4 months each when the preferential stage for freezing was as follows: period 1, expanded blastocyst (138 patients had extra embryos potentially for freezing); period 2, cleaving embryos, five to ten cells (162 patients had extra embryos potentially for freezing); period 3, expanded blastocysts (132 patients had extra embryos potentially for freezing). STORAGE OF CLEAVING EMBRYOS During period 2 of the study, 424 extra embryos were available for freezing, of which 269 (63%) developed to between five and ten cells (Table 1). A total of 179 ofthese 269 cleaving embryos were frozen on days 3 or 4, the remaining 90 embryos being considered unsuitable for freezing. However, these 90 embryos remained in culture, but none of them developed into satisfactory blastocysts. STORAGE OF BLASTOCYSTS During periods 1 and 3 of the study, when extra embryos, were cultured to the blastocyst stage, only one quarter of them expanded in vitro (Table 1). Expansion took place in 45% ofthe embryos on day 5 and in 49% by 1 day later. One embryo expanded on day 4 and seven others on day 7. Development ended most obviously in the remainder during compaction and expansion of the embryo (Table 1). Eleven of the 197 expanded blastocysts obtained during these two periods appeared partly necrotic after exposure to glycerol and were not frozen: on exposure to glycerol the blastocyst contracts, and it may be obvious that the space between the embryo and the zona pellu- 640 Fehilly et ai. Cryopreservation of human embryos

4 Table 1. Incidence of Embryo Storage in Relation to the Number of Extra Embryos Not Replaced Embryos developing into Test period and No. of fresh Proportion of emstage of em- pronucleate or 5- to 8-cell Compacted Cavitated Fully expanded bryos suitable bryos frozen cleaving embryos embryos embryos embryos blastocysts for cryostorage a not replaced Expanded blastocyst /441 b 23 Cleaving stages e 179/ Expanded blastocyst /343 b 24 aresults of the chi-square test: group 1 versus group 3, not significant; groups 1 and 3 versus group 2, P < 0.00l. beleven embryos appeared unsatisfactory after exposure to glycerol and were not stored; three completely hatched embryos were not stored: CNinety of these embryos had many fragments, were partially degenerated, or had arrested development; these embryos were not frozen. cida contains degenerate or retarded loose cells not incorporated in the blastocyst. Some embryos become pale or lyse after exposure to glycerol. Three embryos that hatched from their zona pellucida were also not frozen. Several other embryos were partially hatched, the main mass remaining inside the zona pellucida, and these contracted entirely within this investment upon exposure to glycerol and were frozen. THAWING AND REPLACEMENT OF CLEAVING EMBRYOS Many cleaving embryos were found to have damaged blastomeres after thawing. Others had sufficient intact blastomeres to be considered suitable for replacement (Table 2) and were replaced in their mother within 6 hours of thawing. Freezing can fracture the zona pellucida, and in other mammals this is believed to lead to embryonic death due to the invasion of the precompaction embryo by macrophages within the uterus. 8, 9 Embryonic compaction prevents phagocytosis, and thawed cleaving embryos from ten patients were cultured in vitro for at least 1 day so that they compacted before replacement. This was a precautionary measure. THAWING AND REPLACEMENT OF BLASTOCYSTS Most blastocysts were contracted when thawed, no cavity being visible, and they contracted even further during removal of the glycerol. Contraction was reversed after the complete removal of glycerol in most of the blastocysts, and all intact blastocysts except one expanded within several hours (Table 2). COMPARISON BETWEEN CLEAVING EMBRYOS AND BLASTOCYSTS BEFORE AND AFTER CRYOSTORAGE The number of cleaving embryos available for freezing was 42% (179 of 424) of the total number of extra embryos, whereas of those left to grow to blastocysts in vitro, only 23% (183 of 784) were considered to be suitable for freezing as expanded blastocysts (Table 1). This difference was smaller when evaluated in terms of patients rather than embryos, because approximately half the number of patients had a cleaving embryo frozen, versus one third of other patients having expanded blastocysts frozen (Table 3). None of seven patients receiving cleaving embryos that were apparently intact upon thawing and replaced immediately became pregnant. In contrast, four of ten patients became pregnant when (partially) intact embryos were cultured to compaction in vitro before replacement. Curiously, in one of these cases, the embryo had not cleaved or compacted despite prolonged culture for 43 hours. Table 2. Thawing of Embryos from 49 Patients and Outcome of Replacements Stage of embryo freezing Cleaving stages Expanded blastocysts Fisher's exact test No.ofpatients whose embryos were thawed Intact /38 45 P< 0.02 Condition of thawed embryos Partially intact 19/ /38 24 Not significant Suitable.for replacement 26/ /38 66 P < Proportion of pregnancies Mter replacement of thawed embryos 4/ /15 53 Not significant Per patient whose embryos were thawed 4/ /23 35 Not significant Vol. 44, No.. 5, November 1985 Fehilly et ai. Cryopreservation of human embryos 641

5 Table 3. Incidence of Embryo Storage in Relation to the Number of Patients with Extra Embryos Patients with at least one embryo developing into Proportion of pa- No. of patients 5- to 8-cell Com~acted Cavitated Full}; expanded tients with em- Test period and stage with at least embryo em ryo embryo b astocyst bryos frozen" of embryos frozen one extra embryo Expanded blastocyst Cleaving stages / Expanded blastocyst "Results of chi-square test: group 1 versus group 3, not significant; group 1 versus group 2, P < 0.01; group 2 versus group 3, P < Blastocysts were thawed from 23 patients. In 12 of these cases the blastocysts had been frozen on day 5, and in 9 patients they were considered suitable for replacement after thawing; 6 of these patients became pregnant. In the remaining 11 patients, the blastocysts had expanded and been frozen on day 6; upon thawing, 6 of these 11 patients had embryos that were considered suitable for replacement and two pregnancies resulted (Table 4). The proportion of pregnant patients when one, two, and three thawed blastocysts were replaced was 4 of7, 4 of6, and 0 of2, respectively. Three patients received partially intact blastocysts in which more than half the tissue had survived; one patient became pregnant after the replacement of an embryo in which approximately 80% of the cells were intact. Significantly more blastocysts survived the freezing process than cleaving embryos. However, the incidence of implantation between the two groups evaluated in terms of patients was not significantly different because of the small size of the samples (Table 2). PREGNANCIES ESTABLISHED AFTER THE REPLACEMENT OF FROZEN-THAWED EMBRYOS Details of 12 patients becoming pregnant from the replacement of frozen-thawed embryos are given in Table 5. The average duration ofinfertility was 7.5 years, and four patients had secondary infertility. Two pregnant patients had prolonged (10 years) idiopathic infertility, and the couples had no intercourse during the replacement cycle. Altogether, 12 of the 17 embryps replaced into these patients implanted, which demonstrates the high survival rate of such embryos. Both patients who aborted were older than 40 years of age. One patient who aborted received a cleaving embryo that failed to compact despite prolonged culture in vitro, and her pregnancy was confirmed 642 Fehilly et ai. Cryopreservation of human embryos by rising heg, but an empty gestation sac was visualized by ultrasonography at 9 weeks. Approximately 50% of the thawed cleaving embryos had completed their cleavage cycle at the time of freezing and were regular 8-cell embryos during the cooling process. Their survival was not higher than that of other cleaving embryos (5-, 6-, 7-, 9-, or 10-cell). The four pregnancies established after replacement of cleaving embryos were obtained by cryostorage at the 6- or 7 -cell stage. DISCUSSION The present study demonstrates the advantages of storing expanded human blastocysts for the establishment of pregnancy. Two thirds of thawed blastocysts reexpanded and were considered suitable for replacement, compared with only one third of thawed cleaving embryos. The implantation rate was also higher with blastocysts, with 35% of all patients who had blastocysts thawed becoming pregnant, including those embryos that were not replaced because the blastocysts were considered to be unsuitable. The incidence of pregnancy was similar whether one or two blastocysts were replaced in the mothers, and Table 4. Rate of Development of in Vitro-Fertilized Oocytes into Expanded Blastocysts and Outcome of Treatment After Thawing and Replacement of Cryopreserved Blastocysts Outcome of treatment after thawing of blastocysts No embryo survival Embryo survival and replacement but not pregnant Replacement and pregnant Proportion of patients whose embryos were thawed/day after insemination when blastocyst expanded and frozen 3/ /9 4/6 6/9 2/6 "Fisher's exact test: no significant difference between day 5 and day 6.

6 Table 5. Data on Patients Becoming Pregnant After Replacement of Freeze-Thawed Embryos Proportion Cycle for freeze- No. of embryos Age offemaie Duration of Cause of of oocytes Stage of thawed embryo Outcome of partner infertility infertility fertilized a freezing replacement Thawed Replaced treatment yrs 27 b 7 Tubal 11/11 6- and 8- Natural 6 3 Ongoing cell 43 b 7 Tubal 5/6 7-cell Clomid 1 1 Aborted 33 b 3 Tubal 5/6 6- and 8- Natural 2 1 Ongoing cell 28 5 Tubal 7/11 6-,7-, and Natural 4 2 Ongoing 8-cell Idiopathic 616 Blastocyst Clomid 1 1 Aborted 34 8 Tubal Blastocyst Natural 2 1 Delivered one male infant 37 b 6 Tuballastheno- 9/10 Blastocyst Clomid 3 2 Delivered one spermia male infant 26 7 Tubal Blastocyst Natural 5 2 Delivered one female infant Idiopathic 9/12 Blastocyst Natural 1 1 Ongoing 37 b 11 Tubal 9/lD Blastocyst Natural 2 1 Ongoing 36 9 Tubal 9/12 Blastocyst Clomid 2 2 Ongoing 35 7 Tubal 3/5 Hatching blas- Clomid 1 1 Ongoing tocyst athe first 11 patients had three fresh embryos replaced. Fresh embryos were not replaced in the last patient because of severe gastrointestinal trauma after laparoscopy. bsecondary infertility. no pregnancies resulted when three thawed blastocysts were replaced. This sequence probably reflects our methods of thawing, rather than any underlying physiologic factor, because embryos were thawed singly until one satisfactory for replacement was obtained. Patients receiving two or three embryos might therefore have had only a single embryo capable of implanting. More than half of the patients who had thawed blastocysts replaced became pregnant, the incidence of implantation being 24%; this incidence is similar to that reported by others with comparable methods. 7 These preliminary results can be completely evaluated only when the proportion of embryos available for freezing is taken into consideration. One quarter of the fresh embryos not replaced developed into expanded blastocysts, many being arrested as morulae or during expansion; hence, the number of expanded blastocysts available for freezing was one half that of the number of cleaving embryos. This difference becomes smaller when evaluated in terms of patients, rather than embryos. The low proportion of embryos reaching the expanded blastocyst stage was compensated for by their high incidence of survival and implantation. The present results are based on small numbers of embryos but indicate that freezing blastocysts may be the superior method. The longer period of culture required for embryos to reach the blastocyst stage may act as a quality control to select those unable to display continued growth. During this study period of 12 months, 1313 patients had laparoscopy for oocyte recovery and IVF and 432 of them had at least one extra embryo after the replacement of three fresh embryos. This ratio was used in Table 6, together with parameters of embryo survival and implantation found in this study. We estimate that the overall incidence of pregnancy will increase by 3% with the cryostorage of blastocysts, provided three fresh embryos are still replaced into each patient (Table 6). It would be interesting to calculate the chances in pregnancy rate if none, one, or two fresh embryos were replaced during the treatment cycle, the others being stored as cleaving embryos or blastocysts. This approach would obviously avert the risk of multi pregnancies after the replacement of two or more fresh embryos. The chance of embryos surviving cryostorage might also improve, because the most healthy fresh embryos (i.e., those with the best morphologic features and good cleavage rates) are presumably being selected for replacement, and many of them would be capable of developing into blastocysts. The hazards of cryostorage might thus be less for such embryos, as compared with those now being used for cryopreservation, thus raising the incidence of survival and implanta- Vol. 44, No.5, November 1985 Fehilly et ai. Cryopreservation of human embryos 643

7 Table 6. Estimated Effect of Blastocyst Freezing on the Total Incidence of Pregnancy (Normal Incidence of Pregnancy After Replacement of Fresh Embryos is Set at 25%) Step Laparoscopies 1000 Expected patients with extra 329 embryos Expected patients becoming II pregnant after replacement of fresh embryos Expected patients with ex 115 III panded blastocysts (35% of I) Patients not becoming pregnant 86 IV after replacement offresh embryos (75% of III) Patients becoming pregnant af 30 3 V ter replacement of freeze thawed embryos (35% of IV) Total patients expected to be pregnant (II + V) tion. Evidence has suggested that prefreeze development in vitro, assessed by morphologic examination and rate of development, is an important parameter in embryo survival and the establishment of pregnancy. 3 Acknowledgments. We thank Jonathan Hewitt, M.R.C.O.G., George Rowland, M.R.C.O.G., Patrick Steptoe, F.R.C.O.G., and John Webster, M.R.C.O.G. for their continuo ous support of this study. Mrs. Jean Crow and Ms. Helen Izzard are gratefully acknowledged for technical assistance. Eurof Walters, B.Sc., is acknowledged for his statistical ad vice. REFERENCES 1. Zeilmaker GH, Alberda AT, van Gent I, Rijkmans CMPM, Drogendijk AC: Two pregnancies following transfer of intact frozen-thawed embryos. Fertil Steril 42:293, Trounson AO, Mohr L: Human pregnancy following cryopreservation, thawing and transfer of an 8-cell human embryo. Nature 305:305, Cohen J, Simons RF, Edwards RG, Fehilly CB, Fishel SB: Pregnancies following the frozen storage of expanding human blastocysts. J In Vitro Fertil Embryo Trans 2:59, Willadsen S, Polge C, Rowson LEA, Moor PM: Deep freezing of sheep embryos. J Reprod Fertil 46:151, Department of Health and Social Security (U.K.): Report of the Committee ofinquiry into Human Fertilization and Embryology. London, Her Majesty's Stationery Office, Edwards RG, Fishel SB, Cohen J, Fehilly CB, Purdy JM, Slater JM, Steptoe PC, Webster JM: Factors influencing the success of in vitro fertilisation for alleviating human infertility. J In Vitro Fertil Embryo Trans 1:3, Mohr LR, Trounson AO, Freemann L: Deep-freezing and transfer of human embryos. J In Vitro Fertil Embryo Trans 2:1, Bronson RA, McLaren A: Transfer to the mouse oviduct of eggs with and without the zona pellucida. J Reprod Fertil 22:129, Trounson AO, Moore NW: The survival and development of sheep eggs following complete or partial removal of the zona pellucida. J Reprod Fertil 41:97, Febilly et al. Cryopreservation of human embryos