Vol. 11, Suppl. 3 51 The genetic screening of preimplantation embryos by comparative genomic hybridisation Maria V Traversa 1, James Marshall, Steven McArthur, Don Leigh Genea, Sydney, Australia Received: 12 August 2011; accepted: 30 November 2011 SUMMARY Comparative genomic hybridization (CGH) is an indirect DNA-based test which allows for the accurate analysis of aneuploidy involving any of the 24 types of chromosomes present (22 autosomes and the X and Y sex chromosomes). Traditionally, embryos have been screened using fluorescence in situ hybridization (FISH) a technique that was limited in the number of chromosomes able to be identified in any one sample. Early CGH reports on aneuploidy in preimplantation embryos showed that any of the 24 chromosomes could be involved and so FISH methods were going to be ineffective in screening out abnormal embryos. Our results from routine clinical application of array CGH in preimplantation genetic diagnosis (PGD) patients confirm previous reports on patterns of chromosomal contribution to aneuploidy. The pregnancy outcomes following embryo transfer also indicate that despite the requirement to freeze embryos, rates are encour- 1 Corresponding author: Genea, Level 2, 321 Kent Street, Sydney NSW 2000, Australia; e-mail: maria.traversa@genea.com.au Copyright 2011 by the Society for Biology of Reproduction
52 Improving IVF outcomes by CGH analysis aging, and successful ongoing pregnancies can be achieved. Reproductive Biology 2011 Suppl. 3: 51-60. Key words: aneuploidy, comparative genomic hybridization, PGD, blastocyst INTRODUCTION Chromosomal aneuploidy is a recognised significant contributing factor in spontaneous miscarriages with approximately two thirds of miscarriages being due to aneuploidy [4]. The types of aberrations encountered can range from simple trisomies or monosomies to more complex abnormalities involving multiple chromosomes. The outcomes for the pregnancy attempts can vary from implantation failure, early stage miscarriage to late stage miscarriage or abnormal live births depending on the chromosomes involved. Couples experiencing infertility need assisted reproductive technologies (ART) in order to achieve pregnancy. Many of these couples may undergo multiple IVF cycles and might not achieve a pregnancy and so often seek further options to improve their chances of a successful outcome. Preimplantation genetic diagnosis (PGD) offers IVF couples an additional selection tool over simple developmental quality for choosing the best embryo for transfer. The ability to screen embryos for chromosome errors can improve the chances of a successful implantation event and reduce the likelihood of aneuploid associated miscarriage [5]. Couples experiencing recurrent miscarriage may also consider IVF and PGD as a treatment option. As with repeated implantation failure couples, additional screening allows for the selection of chromosomally normal embryos resulting in improved implantation rates and ongoing pregnancy. Historically, aneuploid embryo screening has involved technologies where only a limited subset of chromosomes can be analysed. Fluorescence in situ hybridization (FISH) technology uses DNA probes to detect the presence of selected regions of chromosomes but is limited in any one assay to a maximum of five probe combinations. Evaluation of further chromosomes requires retesting of the original sample in multiple rounds of hybridization. The most commonly used assay tests for the majority of abnormalities that are detected during prenatal screening, i.e. chromosomes 13, 18, 21, X
Traversa et al 53 and Y. Extended assays (nine chromosome and twelve chromosome panels) are available and include chromosomes more commonly involved in early stage miscarriages, namely chromosomes 8, 14, 15, 16, 17, 20 and 22 [6]. Whilst it is possible to achieve results for up to 12 chromosomes, the multiple hybridization events necessary to do so can compromise the integrity of the test sample and hence the accuracy of the results. Comparative genomic hybridisation (CGH) is a technique that allows for comprehensive aneuploidy screening through its ability to indirectly detect imbalances involving any of the 24 chromosomes. In this technique, a reference genome is used to compare chromosomal complements of the embryo and detect gains and losses involving a significant portion any of the chromosomes. Array CGH (acgh) is a method which uses microarray technology to present selected regions of chromosomes for analysis. A typical microarray comprises 40,000 to 1,000,000 unique chromosome probe regions. As this technology requires a minimum input of DNA, typically 250 ng 2 µg, a method for DNA amplification of the primary biopsy sample is required. Whole genome amplification (WGA) provides sufficient product for acgh analysis. The complete technique (amplification and CGH analysis) requires up to several days to complete and so current IVF cycle management involves freezing of embryos and subsequent transfer in a freeze/thaw cycle. Initial reports on the use of this technology in embryos have documented improved identification of abnormalities as well as high pregnancy outcomes following transfer of screened embryos [2, 3, 7]. Our aim was to apply this technique for the selection of embryos in a group of patients requesting aneuploidy screening for miscarriage reduction or previous IVF implantation failure and measure pregnancy outcomes to identify if CGH is an appropriate application for such conditions. MATERIALS AND METHODS A total of 54 couples presenting with a history of miscarriage or implantation failure consented for their embryos to undergo preimplantation genetic screening in an IVF PGD cycle. Preclinical workup included karyotyping of both partners to exclude constitutional abnormalities that may be con-
54 Improving IVF outcomes by CGH analysis tributing to their infertility, (e.g. chromosome translocations, or mosaic aneuploidies such as 45X or 47 XXY) as well as other relevant pre-cycle hormonal assays conducted as per standard IVF practices. Embryos were cultured to blastocyst stage in sequential media as reported previously [1]. On day 3 of development, viable embryos underwent an assisted hatching procedure by making a small opening in the zona pellucida with the aid of a Zilos TK laser (Hamilton Thorne Biosciences, Beverly, MA, USA). Embryos were assessed early in the morning of Day 5 for the presence of expanding trophectoderm which was herniating suitably from the zona breaching. Those with sufficient trophectoderm formation underwent a biopsy procedure as reported previously [1]. Any embryos requiring further development before biopsy were cultured for another 8 24 h before reassessment was carried out. Commercial whole genome amplification kits (Rubicon Genomics or Sigma WGA4) were used to amplify DNA from the biopsy material. Typical yields are from 4 7 µg. Amplified DNA was purified and subsequently labelled with CY-3 or CY-5 fluorescent tags. A reference DNA sample was treated in a similar manner and labelled with the alternate fluorophor. The two samples were mixed in equal amounts and hybridized on the micro array slide for up to 24 hours according to manufacturer s instructions (Agilent Technologies). Following a washing step, arrays were scanned (Agilent Technologies DNA microarray scanner) and results analysed using software which calculates log ratios of the two fluorophors. Graphic representation (Genomic Workbench Standard Edition 5.0.14, Agilent) enabled interpretation of dosage for each chromosome. Where available, reports on products of conception from IVF couples attending the clinic from July 2009 through to December 2009 were used to identify any chromosomal aneuploid involvement in the miscarriage. The technique cited on the report used to detect abnormalities in this group was multiplex ligation probe assay (MLPA). RESULTS A total of 167 embryos from the 54 couples electing to have CGH were analysed for all 24 chromosome types with 71 (43%) showing abnormal
Traversa et al 55 chromosome numbers. Of the 71 abnormal embryos, 39 (55%) showed aneuploidy involving a single chromosome, 29 (41%) showed aneuploidy involving two chromosomes and 5 (7%) showed the involvement of more than two chromosomes (chaotic abnormalities). Figure 1 summarises the distribution of chromosomes involved in the aneuploidies detected in this subset of embryos and compares this distribution to that from samples from miscarriages (products of conception) of patients who had previously undergone standard IVF treatment. The embryo data shows that any of the chromosomes can contribute to an abnormal result and a subsequent failure of treatment outcome. Analysis of these same embryos using the standard five chromosome aneuploidy FISH screening panel (chromosomes 13, 18, 21, X and Y) would have revealed only 35% of the chromosomally abnormal embryos. While this figure would have Figure 1. Comparison of aneuploidy rates for individual chromosomes for preimplantation genetic diagnosis embryos (PGD) and in vitro fertilization (IVF) miscarriage samples.
56 Improving IVF outcomes by CGH analysis increased to 72% and 78% with the use of nine and twelve chromosome panels it still means that nearly 25% of chromosomally abnormal embryos, screened through the biggest FISH panel, would have been considered suitable for transfer. Transfer of any of these embryos, while having a low possibility of an abnormal live birth, would have resulted in implantation failure or first trimester miscarriage both of which having emotional and financial costs to the patients. Figures 2 and 3 show examples of CGH profiles for four different embryos. Embryo A shows a partial gain of the long arm of chromosome Figure 2. Comparative genome hybridization (CGH) array profiles showing log ratios for individual chromosome array feature. Individual probes are shown as red dots (gain), green dots (loss) or black dots (balanced), including plot of the moving average over 10 Mb regions (shown as slid vertical line). Embryo A shows a partial gain of 6q22.31 qter, embryo B shows complete loss of chromosome 18 (each indicated by a blue horizontal rectangle).
Traversa et al 57 6 (6q22.31 qter,) embryo B shows complete loss of chromosome 18, embryo C shows complete gain of chromosome 7 and embryo D shows a partial loss of the long arm of chromosome 2 (2q36.1 qter). Of the 54 couples entering the study, 29 (average maternal age at transfer being 38.6 years, age range 29.8-43.7) have had a tested embryo transferred in a freeze/thaw cycle. The overall implantation rate for these patients, measured as the detection of a positive fetal heart beat at week 7 ultrasound testing, was 46% (18 of the 39 embryo transfers). Breakdown of the pregnancy data by age, shows that women <38 yrs achieved an implantation Figure 3. Comparative genome hybridization (CGH) array profiles showing log ratios for individual chromosome array features (individual probes shown as red dots (gain), green dots (loss) or black dots (balanced), including plot of the moving average over 10 Mb regions (shown as slid vertical line). Embryo C shows complete gain of chromosome 7 and embryo D shows a partial loss of 2q36.1 qter (each indicated by a blue horizontal rectangle).
58 Improving IVF outcomes by CGH analysis rate of 56% (9/16) and women >38 yrs achieved 39% (9/23). The clinical pregnancy rate per oocyte retrieval for these patients equates to 75% (9/12) for women <38 yrs and 53% (9/17) for women >38 yrs. DISCUSSION The ability to screen for abnormalities involving any of the 24 chromosomes represents a major advantage over FISH based technologies. The results presented here confi rm that any of the chromosomes are capable of contributing to abnormalities found at the blastocyst stage of embryo development. Aneuploidy involving the larger chromosomes, which typically are not part of commercially available FISH based kits, were detected in several blastocyst stage embryos, contradicting the belief that embryos with errors involving these chromosomes are less likely to develop to later stages this data shows that they can and do form quality blastocysts. In contrast to FISH which identifies only a single point on a chromosome, the many probes used in CGH span the entire length of a chromosome, and so valuable information involving structural errors where only parts of chromosome are involved, (e.g. partial duplications and deletions) can also be detected. The ability to determine loss/gain of parts of a chromosome opens the possibility of using CGH to analyse embryos of translocation couples instead of FISH, which has been reported to suffer from a high false positive abnormality rate [8]. While improved PCR analysis methods have been reported for translocation analysis [8], it is technically not easy to incorporate further aneuploid analysis in these methods. In situations where translocated fragments are large enough for detection, CGH offers the opportunity to determine balanced chromosome sets as well as ploidy status of the other 22 chromosomes present. CGH provides an averaged chromosome result across the entire tissue sample biopsied from the embryo unlike FISH where each cell is seen and analysed individually. This averaging partially overcomes the well documented interpretation problems associated with low level mosaicism
Traversa et al 59 where a single chromosomally abnormal cell would often result in an abnormal assessment of the embryo in spite of the awareness that a small subset of aberrant cells often does not result in a clinically significant abnormality. It has been observed that low level mosaicism can commonly be found in rapidly dividing cells but unfortunately in FISH PGD, this may result in a false positive abnormal result and an otherwise clinically suitable embryo is deemed unsuitable for transfer and discarded. Finally, our results here demonstrate the reliability and feasibility of this CGH approach of detection of aneuploidy in blastocyst stage embryos and subsequent improved outcomes when tested embryos are selected for transfer. The outcomes achieved with CGH for patients experiencing repeated implantation failure and recurrent miscarriage show that extended aneuploidy screening can have a positive effect on the likelihood of a successful transfer and subsequent pregnancy in otherwise poor prognosis patients. In addition, the patients with tested embryos for transfer are saved the cost and time of futile transfers and have a reduced risk of miscarriage. REFERENCES 1. De Boer KA, Catt JW, Jansen RPS, Leigh D, McArthur S 2004 Moving to blastocyst biopsy for PGD and single embryo transfer at Sydney IVF. Fertility and Sterility 82 295-298. 2. Fragouli E, Katz-Jaffe M, Alfarawati S, Stevens J, Colls P, Goodall N, Tormasi S, Gutierrez-Mateo C, Prates R, Schoolcraft WB, Munne S, Wells D 2010 Comprehensive chromosome screening of polar bodies and blastocysts from couples experiencing repeated implantation failure. Fertility and Sterility 94 875-887. 3. Fragouli E, Lenzi M, Ross R, Katz-Jaffe M, Schoolcraft WB, Wells D 2008 Comprehensive molecular cytogenetic analysis of the human blastocyst stage. Human Reproduction 23 2596-2608. 4. Lathi RB, Westphal MD, Milki AA 2008 Aneuploiy in the miscarriages of infertile women and the potential benefit of Preimplantation genetic diagnosis. Fertility and Sterility 89 353-357. 5. Munne S, Wells D, Cohen J 2010 Technology requirements for Preimplantation genetic diagnosis to improve assisted reproduction outcomes. Fertility and Sterility 94 408-430.
60 Improving IVF outcomes by CGH analysis 6. Munne S, Fragouli E, Colls P, Katz-Jaffe MG, Schoolcraft WB, Wells D 2010 Improved detection of aneuploid blastocysts using a new 12 chromosome FISH test. Reproductive BioMedicine Online 20 92-97. 7. Schoolcraft WB, Fragouli E, Stevens J, Munne S, Katz-Jaffe MG, Wells D 2010 Clinical application of comprehensive chromosomal screening at the blastocyst stage. Fertility and Sterility 94 1700-1706. 8. Traversa MV, Carey L, Leigh D 2010 A molecular strategy for routine preimplantation genetic diagnosis in both reciprocal and Robertsonian translocation carriers. Molecular Human Reproduction 16 329-337.