S.Kahraman 1,4, M.Bahçe 2,H.Şamlı 3, N.İmirzalıoğlu 2, K.Yakısn 1, G.Cengiz 1 and E.Dönmez 1

Human Reproduction vol.15 no.9 pp.2003 2007, 2000 Healthy births and ongoing pregnancies obtained by preimplantation genetic diagnosis in patients with advanced maternal age and recurrent implantation failure S.Kahraman 1,4, M.Bahçe 2,H.Şamlı 3, N.İmirzalıoğlu 2, K.Yakısn 1, G.Cengiz 1 and E.Dönmez 1 1 Sevgi Hospital, Assisted Reproductive Technologies and Reproductive Endocrinology Unit, 2 G.A.T.A., Genetic Division and 3 Sevgi Hospital, Genetic Division, Ankara, Turkey 4 To whom correspondence should be addressed at: İstanbul Memorial Hospital, Reproductive Endocrinology and ART Unit, Piyalepasa Bulvar1, 80270, Okmeydan1 Şişli İstanbul, Turkey. E-mail: semrakahraman@hotmail.com Preimplantation genetic diagnosis (PGD) and subsequent embryo development was evaluated in 72 couples presenting at our centre for intracytoplasmic sperm injection (ICSI) due to severe male factor. The embryo biopsies were performed in Ca 2 /Mg 2 -free medium. These patients were further divided into those with advanced maternal age (AMA, n 49) and those with recurrent implantation failure (RIF, n 23). Fluorescence in-situ hybridization (FISH) was carried out on 329 blastomeres (91.3%) with probes for the X, Y, 13, 18 and 21 chromosomes. The chromosomal abnormality rate was 41.3% with no significant difference between the AMA and RIF groups. Aneuploidy accounted for the majority (72.8%) of chromosomal abnormalities. Out of 329 embryos, 84.2% had cleaved after 24 h and 15.1% had arrested. Embryos were transferred in 70 patients and 22 pregnancies were achieved (31.4% with an ongoing pregnancy rate of 28.5%). There were no significant differences between the pregnancy rates of the AMA and RIF groups (32.5 and 30% respectively). Therefore PGD should be offered to patients with AMA and RIF. Furthermore, the use of Ca 2 /Mg 2 -free medium during the blastomere biopsy facilitates the procedure, while further embryo cleavage, ongoing pregnancies and healthy births are possible. Key words: aneuploidy/ca 2 /Mg 2 -free medium/chromosomes X, Y, 13, 18, 21/FISH/preimplantation genetic diagnosis Introduction Preimplantation genetic diagnosis (PGD) allows for the selection of chromosomally normal embryos for transfer after IVF or intracytoplasmic sperm injection (ICSI) in patients with advanced maternal age (AMA) carrying a risk for aneuploidy. Aneuploidy may be a major contributing factor of implantation failure in women of AMA undergoing assisted reproduction treatment. Therefore, the selection of normal embryos could increase the rate of pregnancy and healthy offspring. PGD offers an alternative to the prenatal diagnosis in couples with a risk of transmitting inherited diseases. Furthermore, recurrent implantation failure (RIF) occurs in some young patients even when good quality embryos are transferred. Chromosomal evaluation of the embryos in these patients could provide further insights into the reasons for RIF. PGD is still being evaluated by several centres for patients with RIF and recurrent abortion if aneuploidy is present. It has been reported that embryo biopsy is easier to perform in the absence of Ca 2 /Mg 2 ions (Grifo et al., 1990; Santalo et al., 1996). No detrimental effects on the viability of biopsied mouse embryos were observed during the limited time of exposure to Ca 2 /Mg 2 -free medium compared to the control group. However, it has been reported that a prolonged exposure to Ca 2 /Mg 2 -free medium resulted in impaired embryonic development (Ducibella and Anderson, 1976; Reeve, 1981). The possible toxic effects on embryo and embryonic development to the blastocyst stage were studied when Ca 2 /Mg 2 - free medium was used in humans (Dumoulin et al., 1998). In this study, human embryos that were left over after in-vitro fertilization (IVF) and that were considered unsuitable for freezing were used to assess the effect of Ca 2 /Mg 2 -free medium on human embryos, further embryo development and pregnancies after embryo biopsy for PGD. To our knowledge, this is the first study reporting further human embryo development, ongoing pregnancies and healthy births with Ca 2 / Mg 2 -free medium for embryo biopsy. Material and methods This study was undertaken at Sevgi Hospital Embryology and Genetic Laboratory between December 1998 and June 1999. Third day embryos were biopsied in 72 patients for chromosomal evaluation. AMA (49 patients) and RIF groups with more than two cycles were included. From these two groups eight patients had centrally located granular oocytes (Figure 1) in their present and previous cycles and six male partners had megalo-pin head spermatozoa globozoospermic forms and round spermatids in their ejaculates (Figure 2). Patients signed a consent form before PGD consenting to the use of their embryos for blastomere biopsy and in case of pregnancy, amniocentesis. This study has been approved by the local ethical committee of Sevgi Hospital. An embryo biopsy was performed at 64 68 h after ICSI and only embryos with more than seven or eight blastomeres were selected for biopsy. Pedigree analysis, family history and appropriateness of cases for PGD were evaluated. Patients were counselled regarding the advantageous of aneuploidy screening. Furthermore, couples were counselled regarding possible increased risk of sex chromosomal abnormalities in offspring and the need for follow-up prenatal testing. Males were karyotyped and counselled for the possibility of gonadal European Society of Human Reproduction and Embryology 2003

S.Kahraman et al. mosaicism or Y chromosome deletion/insertion. From February 1999, Y chromosome deletions are performed in our Molecular Genetic Laboratory. Figure 1. Centrally located granular oocytes. Excessive central granular structures (big arrows). Figure 2. A megalohead spermatozoon (MHS) (big arrow) together with pin-head spermatozoon (PHS) (small arrow) and round spermatid (RS). Figure 3. Trisomy 13. A blastomere FISH analysis with an extra signal of chromosome 13. 2004 Embryo biopsy A stimulation protocol previously described (Kahraman et al., 1999) was employed. In 72 cases with 72 cycles an ICSI procedure was performed due to severe male infertility with AMA and RIF. Embryos were placed in HEPES-buffered medium without Ca 2 /Mg 2 ions (EB TM -10 Medium, Scandinavian IVF Science, Gothenburg, Sweden) before the biopsy procedure. Embryos were incubated for 5 25 min with EB TM -10 medium according to the degree of compaction between blastomeres. Embryos with compacted blastomeres on the third day were exposed to between 15 and 25 min to an EB TM -10 medium. Granulation of blastomeres was observed in these embryos exposing a relatively longer duration. Each Petri dish contained 10 ml of EB TM -10 medium covered with sterile equilibrated mineral oil. The first droplet was prepared for biopsy procedure and the second one was to replace the biopsied blastomere(s) in the same Petri dish. Three-dimensional partial zona dissection in the shape of crosshatching or V-shape was performed to facilitate the biopsy procedure as was originally introduced by Verlinsky s group (Cieslak et al., 1999). V-shaped assisted hatching was performed in most of the embryos to create a triangular flap opening with a diameter of 25 30 mm to allow the replacement of the blastomere biopsy pipette. In 67.2% of the embryos, blastomeres were biopsied by pressing the zona pellucida with a biopsy pipette to make the blastomere protrude from the zona opening. In the remaining embryos (32.8%), blastomeres were biopsied by entering the opening of the zona pellucida with the biopsy pipette. The biopsy procedure was executed 64 68 h after insemination for embryos obtained by ICSI. Only embryos with seven or more blastomeres were selected for biopsy. Embryos with a poor morphology with uneven blastomeres and a fragmentation rate of 30% were excluded. After the biopsy was done on third day embryos without compaction, the embryos were replaced in the EB TM -10 medium for approximately between 5 and 15 min. When compaction with tight junction between blastomeres was observed, 15 25 min of exposure to EB TM 10 medium was preferred. Each biopsy procedure was performed by the same operator and was observed by a second person by means of a monitor, connected to a video camera mounted on a microscope. The duration of the biopsy procedure was noted from the beginning (from fixing embryos with a holding pipette to replacing single biopsied blastomeres in a second EB TM -10 droplet) to the end. The embryo to be biopsied was placed in a Petri dish in Ca 2 /Mg 2 -free medium containing 3 mg/ml HSA and 100 IU/ml Pc-6 EB TM -10 after equilibration at 37 C and 5% CO 2. The number of blastomeres in the beginning of the procedure, biopsied blastomeres, nucleus, multinucleated blastomeres, damaged blastomeres in the embryo or damaged biopsied blastomeres were recorded. The type of biopsy protrusion and subsequent development of embryos were observed for 24 h more to follow their developmental capacity. Any signs of cell or embryo damage were recorded. Biopsied embryos were cultured overnight (24 h more) both to observe their further development and to allow for a sufficient period for FISH procedure. Statistical analysis was performed using a Student s t-test or χ 2 test, where appropriate. Fluorescence in-situ hybridization (FISH) The isolated blastomeres were analysed with a previously described protocol (Munné et al., 1993) except for the hybridization time,

Survey of PGD in patients with AMA and RIF Table I. Results of chromosomal analysis (n 329) Results Embryos analysed n (%) Normal 193 57.8 Abnormal 136 41.3 Aneuploidy 99 72.8 Trisomy 47 47.5 Monosomy 41 40.8 Double monosomy 4 3.9 Double trisomy 4 3.9 Trisomy monosomy 4 3.9 Complex aneuploidy 17 12.5 Polyploidy 13 9.6 Haploidy 7 5.1 which was 3 h. A Nikon E-600 microscope with filters recommended by Vysis for the probe mixture was used. The scoring criteria described previously (Munné and Weier, 1996) were followed and minor hybridization spots with low fluorescence intensity were not scored. For the multicolour FISH analysis, five DNA probes were used for the simultaneous detection of chromosomes X, Y, 13, 18 and 21 (Figure 3). Embryos were classified as complex abnormal when two or more chromosomes had an abnormal count but were not completely polyploid or haploid. The blastomeres of abnormal embryos not being transferred were fixed and evaluated by FISH procedure to confirm the diagnosis. Results In a total of 72 patients, PGD was performed for AMA in 49 (68%) patients and RIF in 23 (32%) patients. The mean ages of AMA and RIF patients were 37.9 2.1 and 30.3 3.1 respectively (P 0.05). In all, 360 embryos were selected for biopsy. A successful FISH analysis was carried out in 329 out of 360 blastomeres (91.3%); the rates of normal and abnormal chromosomal evaluation of embryos were 58.7 and 41.3% respectively. In 99 (72.8%) of abnormal embryos, aneuploidy was diagnosed. In seven (5.1%) of the embryos haploidy, in 13 (9.6%) polyploidy and in 17 (12.5%) complex aneuploidy were diagnosed. Out of 329 embryos analysed, 58.7% were found to be normal while 41.3% had chromosomal abnormality. The majority of the abnormal embryos were found to have aneuploidy (72.8%). Polyploidy, complex aneuploidy and haploidy were found to be 9.6, 12.5 and 5.1% respectively. Monosomy was diagnosed in 40.8% of the aneuploid embryos; 47.5% trisomy, 3.9% double monosomy, 3.9% double trisomy, 3.9% trisomy monosomy were observed respectively. The most significant chromosomal abnormality was observed in chromosome 18 (P 0.05) (35.3%) in the aneuploid embryo group, with chromosome 21 second at 27.1%, and chromosome 13 third at 22.4%. Table I shows results of chromosomal analysis. In the AMA group, the rates of chromosomally abnormal and normal embryos were 40.4 and 59.6% respectively. The majority of the abnormal embryos had aneuploidy (76.1%). Polyploidy, complex aneuploidy and haploidy were found in 10.3, 9.1 and 4.5% of embryos respectively. The rates of chromosomally abnormal and normal embryos in the RIF group were 43.2 and 56.8%. The majority of abnormal Table II. Distribution of the further development of embryos and duration of biopsy n 3 min 3 min Mean SD Mean SD Normal cleaved embryos 222 2.2 1.3 2.7 1.5 Slow developed embryos 57 1.3 0.8 1.4 0.6 Arrested embryos 50 1.5 0.7 1.3 0.6 There were no significant differences between the two durations of biopsy groups. Table III. A comparison of characteristics of assisted reproduction cycles in patients with advanced maternal age (AMA) and repeated implantation failure (RIF) AMA RIF P-value* n 49 n 23 mean SD mean SD Age 37.9 2.1 30.3 3.1 0.05 No. of oocytes obtained 11.7 3.8 14.1 2.9 Fertilization with 2 PN 8.1 3.9 (69) 10.6 4.1 (75) 0.05 Cleaved embryos 7.4 3.3 (91) 8.6 3.4 (81) NS Biopsied embryos 4.9 2.0 (67) 5.4 2.3 (63) NS Abnormal embryos 2.0 1.4 (39) 2.2 1.1 (46) NS Transferred embryos 2.8 1.2 (57) 2.8 0.8 (53) NS *t-test. NS not significant. Values in parentheses are percentages. embryos had aneuploidy (66.7%). Polyploidy, haploidy and complex aneuploidy were found in 8.3, 6.2 and 18.8% of embryos respectively. The mean duration of the biopsy procedure from the beginning of assisted hatching to the replacing blastomere(s) in another droplet containing Ca 2 /Mg 2 -free medium was 4.2 3.2 minutes per biopsy procedure. Table II shows the further development of embryos exposed for 3 min (1 3 min) and 3 min (3 8 minutes) to EB TM -10 medium during the biopsy procedure. In 32.8% of embryos a blastomere biopsy was performed by pressing zona pellucida with a biopsy pipette to make the blastomere protrude from the zona opening whilst in 67.2% of embryos, biopsy pipette was inserted into the embryo to remove one or two blastomeres. Table III represents a comparison of characterictics of assisted reproduction cycles in patients with AMA and RIF. No statistical significant difference was observed between the rate of cleaved, biopsied, abnormal and transferred embryos (P 0.05). The fertilization rate was found to be higher in the RIF group probably due to the higher number of oocytes obtained (P 0.05). Although the number of embryos (n 21) evaluated chromosomally was low in cases with megalo spermic samples (n 6), high rates of aneuploidy (62.5%) and polyploidy (31.3%) were observed, probably due to a high incidence of diploidy in megalo head spermatozoa. A high aneuploidy rate was found (76%), while 24% of embryos were chromosomally normal in this group. 2005

S.Kahraman et al. Eight patients had centrally located excessive granular oocytes (CLGO) in their previous and present cycles. As very low implantation and ongoing pregnancy rates (4.7%, unpublished data) were achieved. PGD was performed to evaluate the chromosomal complement of this group. Thirtynine percent of embryos were found to be chromosomally abnormal in the CLGO group, while 61% of embryos were found to be normal. The majority of abnormal embryos had aneuploidy (56.2%) in the same group. The rate of normally cleaving embryos on the fourth day was 84.2 and 15.8% of the embryos arrested. Out of 84.2% normally cleaved embryos, 18.1% of them exhibited a slow cleavage (an increase of less than two blastomeres in 24 h). Of the 72 cases, two patients had no transfer due to the lack of chromosomally normal embryos. Out of 110 chromosomally normal embryos transferred, a total of 22 pregnancies were achieved in 70 patients (31.4%). One pregnancy ended with a spontaneous abortion at 8 weeks and one ended with a blighted ovum at 6 weeks. The cytogenetic evaluation of both patients pregnancy material revealed 46, XY; 46, XX chromosomal complement, confirming the results of PGD. The rate of ongoing pregnancy was found to be 28.5% in total, a 25.5% ongoing pregnancy rate was achieved in AMA and 30.4% in RIF group respectively (P 0.05). Ten healthy births have been achieved; other pregnancies are continuing into the third trimester. Discussion Many factors have been attributed to the low success rate of assisted reproduction in women of advanced maternal age. These factors can either be endometrial or oocyte related. Defective hormonal milieu, poor uterine flow and inadequate endometrial proliferation cause reduced endometrial receptivity and thus implantation failure. Alternatively, reduced embryo quality, ageing of oocytes and chromosomally imbalance may explain a low success rate in AMA. This explanation is supported by the fact that, if oocytes from younger women are donated to the older women, successful pregnancies can be achieved (Navot et al., 1994). PGD for women aged between 36 and 45 years undergoing assisted reproduction were evaluated by using multiprobe FISH for chromosomes X, Y, 18,13 and 21. It has been demonstrated that additional chromosomes, in particular chromosome 16, were most frequently involved in aneuploid karyotypes (Bahce et al., 1999). Thus it might be beneficial to analyse a larger set of chromosomes to improve pregnancy rates in these patients. The minimum and maximum exposure of embryos to the EB medium during biopsy was between 5 and 25 min depending on the severity of compaction between blastomeres. The subsequent embryo development was not affected when the cell number was seven or more. More compacted embryos were exposed for 10 min. The minimum and maximum duration of the biopsy procedure from the beginning of the assisted hatching to the end of the biopsy was 1 min 20 s and 8 min respectively. The use of Ca 2 /Mg 2 -free medium for biopsy on human embryos to seven or more cells allows an easy biopsy with a low rate of cell lysis as well as a shorter 2006 time needed to perform the biopsy. A shorter biopsy procedure was carried out with the use of Ca 2 /Mg 2 -free medium on human embryos to remove blastomeres with a low rate cell lysis (Dumoulin, 1998). However, the possible toxic effects on embryo and human blastocyst were not studied on embryos that would be transferred. To our knowledge, this is the first study to evaluate the effect of Ca 2 /Mg 2 -free medium on the development of human embryos and pregnancies. If the majority of embryos to be biopsied were found to have five to six cells on the morning of the third day, they were checked 4 5 h later. Unfortunately, no pregnancy was achieved by transferring these slow growing embryos. This may indicate a poor implantation due to chromosomal, physiological or unknown reasons. The parameter used in our study was to assess further embryo viability. Slow growing embryos, arrested embryos or normally cleaving embryos were evaluated. Only 15% of embryos were found to be arrested 24 h later. After a further 2 3 day observation of arrested embryos, very few reached the blastocyst stage. If an increase of fewer than two blastomeres was observed, these embryos were recorded as slow growing embryos (17%). There was no statistical difference between the cleavage rates and biopsy duration was less or more than 3 min (P 0.05). The time needed for removal was noted together with the rate of intact cells. Biopsy time was calculated from the embryo fixation with a holding pipette to placing the biopsied blastomere(s) to another droplet. In fact, the length of fixation of embryos with opened zona to the placing EB TM -10 medium was almost always 1 min. Based on embryo morphological evaluation, 360 embryos were selected for blastomere biopsy. The links between the numbers of cumulus oocyte complexes (COC) and biopsied embryos and between the numbers of COC and unaffected embryos were found to be highly suggestive (P 0.05) (Vandervorst, 1998). It has been claimed that it is justifiable to cancel PGD cycles in which it is expected that fewer than six COC will be retrieved. The mean fertilization rates with 2PN were 8.1 3.9 and 10.6 4.1 in AMA and RIF groups respectively in our study (P 0.05). The fertilization rate was found to be higher in RIF group. This may be due to the number of oocytes obtained. The mean numbers of chromosomally abnormal embryos were found to be similar in the AMA and RIF groups (2.0 1.4 versus 2.2 1.1) (P 0.05). Embryos with only grade I and II were used for biopsy. Poor embryo morphology and a low rate of cleavage are associated with a high incidence of chromosome disorder when compared with normally developing embryos. In our experience with slow developing embryos where biopsies were performed on the third day, no pregnancy was achieved. Six patients had megalohead and pin-head sperm samples in their ejaculates. A high number of sperm heads without tails and abundant round spermatid forms were commonly observed. Round-headed spermatozoa were seldom accompanied by these severely abnormal spermatozoa. Low fertilization and pregnancy rates were achieved (43.2 and 9.1%) with these severely defective forms in our recent study (Kahraman et al., 1999). Low fertilization and pregnancy rates might be due to a high incidence of chromosomal abnormalities. PGD was

Survey of PGD in patients with AMA and RIF performed to evaluate a chromosomal complement in this group. A high aneuploidy rate was found (76%), while 24% of embryos had normal chromosomes. Therefore, couples should be counselled about low rates of fertilization and pregnancy with ICSI when only pin-head megalohead forms are present in the ejaculate. Pregnancy rates were found to be similar in both AMA and RIF groups. PGD also resulted in similar pregnancy rates for both groups. PGD also resulted in similar findings for both groups. In patients with RIF, a 30.4% ongoing pregnancy rate was encouraging. In AMA group patients aged between 36 45, a 25.5% pregnancy rate was achieved. Only one pregnancy resulted in a blighted ovum in AMA group presenting a low (7.6%) biochemical abortion rate. Ten healthy babies were born and the other pregnancies are continuing into the third trimester. Thus PGD may be offered to the patients with repeated implantation failure and to women with advanced maternal age. Ca 2 /Mg 2 -free medium allows for an easier biopsy procedure during PGD with less chance of damaging the embryo. At the same time, it does not result in any loss developmental potential to the morula or even blastocyst stage. A lower rate of cell lysis and a shorter time are needed for this procedure. The use of calcium and magnesium free media is facilitates biopsy of cleavage stage embryos with seven to eight cells or morula stages. References Bahce, M., Cohen, J. and Munné, S. (1999) Preimplantation genetic diagnosis of aneuploidy: were we looking et the wrong chromosomes? J. Assist. Reprod. Genet., 16, 176 181. Cieslak, J., Ivakhnenko, V., Wolf, G. et al. (1999) Three-dimensional partial zona dissection for preimplantation genetic diagnosis and assisted hatching. Fertil. Steril., 71, 308 313. Ducibella, T. and Anderson, E. (1979) The effect of calcium deficiency on the formation of the zona occludens and blastocoel in the mouse embryo. Dev. Biol., 73, 46 58. Dumoulin, J.C.M., Bras, M., Coonen, E. et al. (1998) Effect of Ca 2 /Mg 2 - free medium on the biopsy procedure for preimplantation genetic diagnosis on further development of human embryos. Hum. Reprod., 13, 2880 2883. Grifo, J.A., Boyle, A., Ficher, E. et al. (1990) Preembryo biopsy and analysis of blastomeres, by in situ hybridization. Am. J. Obstet. Gynecol., 163, 2013 2019. Kahraman, S., Akarsu, C., Cengiz, G. et al. (1999) Fertility of ejaculated and testicular megalohead spermatozoa with intracytoplasmic sperm injection. Hum. Reprod., 14, 726 730. Munné, S. and Weier, H.U.G. (1996) Simultaneous enumeration of chromosomes 13,18,21, X and Y in interphase cells for preimplantation genetic diagnosis. Cytogenet. Cell. Genet., 75, 263 270. Munné, S., Grifo, J., Cohen, J. et al. (1993) Diagnosis of major chromosome aneuploidies in human preimplantation embryos. Hum. Reprod., 8, 2185 2191. Navot, D., Drews, M.R., Bergh, P.A. et al. (1994) Age related decline in female fertility is not due to diminished capacity of the uterus to sustain embryo implantation. Fertil. Steril., 61, 97 107. Reeve, W.J.D. (1981) The distribution of ingested horsea dish peroxidanse in the 16-cell mouse embryo. J. Embryol. Exp. Morphol., 66, 191 207. Santalo, J., Grossman, M. and Egozcue, J. (1996) Does Ca 2 /Mg 2 -free medium have an effect on the survival of the preimplantation mouse embryo after biopsy? Hum. Reprod. Update, 2, 257 261. Vandervorst, M., Liebaers, I., Sermon, K. et al. (1998) Succesful preimplantation genetic diagnosis is related to the number of available cumulus oocyte complexes. Hum. Reprod., 13, 3169 3176. Received on October 15, 1999; accepted on June 5, 2000 2007