Preimplantation Genetic Diagnosis Comes of Age

On July 15, 2011, preimplantation genetic diagnosis (PGD) came of age with the 21st birthday of Natalie and Danielle, the first twin girls to be born in the world to a couple at risk of the X-linked condition adrenoleukodystrophy.[1] Since then, many thousands of couples have undergone in vitro fertilization (IVF) and genetic analysis of single or small numbers of cells biopsied from preimplantation embryos, and thousands of infants were born following transfer of embryos diagnosed as unaffected by genetic defects identified in one or both parents.[2] The goal of helping at-risk couples avoid the trauma of prenatal diagnosis and the possibility of terminating affected pregnancies has thus been achieved, and PGD is now clinically well established worldwide.

In this issue, we have assembled a series of expert comprehensive and detailed reviews on the achievements and controversies of the last 2 decades as well as more recent developments in PGD, focusing mainly on the current clinical state of the art. PGD is a complex clinical procedure. To help a couple have a healthy child free of severe genetic disorders requires the close collaboration of several disciplines including clinical genetics, reproductive medicine, clinical embryology, and molecular genetics to identify the genetic defect in the family, inform the couples of the risks as well as the potential benefits of PGD, manage the ovarian stimulation safely to maximize the chance of identifying unaffected embryos, biopsy the embryos minimizing any damage, and finally perform highly accurate single-cell genetic testing.

With hindsight, PGD was born prematurely. In the late 1980s, pregnancy rates following IVF were much lower than those routinely achieved today. The genes involved in common inherited diseases, such as cystic fibrosis and Duchenne's muscular dystrophy, had only recently been identified. The discovery of the polymerase chain reaction (PCR) for amplifying short fragments of DNA was promising, but the amplification of single-copy sequences from single cells was an enormous challenge. The oligonucleotide primers required to define the fragment to be amplified were difficult to design, with often limited access to sequence information, and they were unreliable. Heroic numbers of PCR cycles were required to produce enough DNA to visualize by basic methods, and consequently they were exquisitely sensitive to contamination with previously amplified PCR products. Single-cell PCR was also error prone because one parental copy (allele) sometimes failed to amplify (i.e., allele dropout [ADO]), a problem that remains unsolved today.

PGD therefore gained a reputation as a difficult and inefficient technology with a high risk of misdiagnosis, which it has never completely shaken off. Furthermore, fueled by the intense debate over the ethics of human embryo research, particularly in the United Kingdom, and the publicity surrounding the first cases that involved identifying the gender of the embryo to avoid sex-linked diseases typically only affecting males, but with obvious potential for social sex selection, the epithet “designer baby” was coined and to this day has mired PGD in unnecessary regulation not applied to other methods of prenatal diagnosis.[3]

Today, with improvements to embryo culture media that enable extended culture of human embryos to the blastocyst stage, implantation rates per embryo transferred have increased significantly. Thus prospects for at-risk couples to achieve a pregnancy following PGD have increased correspondingly. This is particularly so if they are young and fertile and have one or more good quality unaffected blastocysts for transfer. Until recently, PGD has been performed either on the two by-products of female meiosis, the first and second polar bodies, as pioneered by Yury Verlinsky and colleagues,[4] or on single cells biopsied from embryos at cleavage stages when the embryo has divided to the 6- to 10-cell stage early on the third day following fertilization.[5] [6] Many clinics, particularly in Europe, now actively promote single blastocyst transfer to reduce the risk of multiple pregnancies and the associated risks of prematurity and neonatal complications. Consequently, there has been renewed interest in the biopsy of small pieces of the outer trophectoderm layer from expanded blastocysts. Xu and Montag discuss the technical aspects of the different biopsy strategies and review their pros and cons for diagnosing different types of genetic abnormality.

With the sequencing of the human genome, we now take it for granted that with a few clicks, we have access to complete sequence information that has been made publicly available on the Internet. When PGD began in the 1990s, many inherited diseases caused by single-gene defects had not even been mapped accurately to a particular chromosome, and identifying mutations in the parents or affected children was a complex and often lengthy process. Furthermore, only a comparatively few polymorphic markers of known location were available, and these were mainly restriction fragment length polymorphisms that were not ideally suited to straightforward PCR analysis. Now we have thousands of polymorphic short tandem repeat and millions of single nucleotide polymorphism (SNP) markers distributed at known locations across the genome. This resource, together with the use of fluorescence PCR and highly sensitive detection and fragment length analysis by capillary electrophoresis using automated sequencers, has revolutionized our ability to analyze multiple fragments from single cells facilitating combined mutation detection and marker analysis. This strategy is applicable to almost all single-gene defects and has largely eliminated the main causes of technical diagnostic errors, undetected contamination, and ADO. Fiorentino reviews these developments and how they have been extended, for example, to combined HLA matching and single-gene defect diagnosis. Also, more recent developments, harnessing the power of microarrays for simultaneous analysis of thousands of DNA sequences following whole genome amplification from single cells, are critically evaluated.

In the mid-1990s, the use of fluorescence in situ hybridization (FISH) with chromosome-specific probes was developed for the detection of abnormal chromosome copy number (aneuploidy) in the interphase nuclei of single cells. The first application was to identify the sex of the embryo with X and Y specific probes in couples at risk of X-linked disease.[7] This led rapidly to the use of multicolor FISH in two or more sequential hydridizations to screen embryos for chromosome aneuploidy more generally, mainly in women of advanced maternal age but also other indications such as repeated miscarriage or in vitro fertilization (IVF) failure. Because only a limited number of fluorochromes are available, the number of chromosomes that can be analyzed in a single hybridization is also limited. The main focus initially, therefore, was on those chromosomes, which could result in viable trisomies (i.e., chromosomes 13, 16, 18, 21, X and Y).[8] With further experience, however, this was later extended to include those chromosomes most commonly associated with aneuploidy at these early stages, including chromosomes 15 and 22.

Because of the well-known technical limitations of multicolor FISH with interphase nuclei, i.e., overlapping or split signals, falsely indicating monosomy or trisomy in a euploid nucleus, respectively, it was necessary to validate the single-cell results with the rest of the embryo. This confirmed that there was a high incidence of aneuploidy detected in most, if not all, of the other nuclei, indicating that one of the gametes had been aneuploid at fertilization. Furthermore, the incidence of these “uniform” aneuploidies increased with maternal age in the decade preceding the menopause as would be expected from the well-established predominance and age-related incidence of maternal aneuploidies in pregnancy and miscarriage following natural conception.[9] However, this work also led to one of the major discoveries of preimplantation genetics: In about a third of embryos examined, aneuploidies detected in the single cell were only present in a minority of the other cells, and conversely, in embryos identified as euploid by single cell analysis, a minority of the other cells of the embryo often had one or more aneuploidies for the chromosomes studied.[10] [11] This postzygotic chromosomal mosaicism, which is thought to arise through malsegregation of chromosomes in the mitotic divisions of the cleavage-stage embryo following fertilization, was unexpected.

Despite these technical limitations and the complication of chromosomal mosaicism, and encouraged by promising improvements in pregnancy rates, preimplantation genetic screening (PGS) for aneuploidy became widely practiced, especially in the United States, and it remains by far the most frequent use of PGD. In 2007, however, a large randomized multicenter clinical trial of its use in all women between the ages of 35 and 41 demonstrated that live-birth rates were significantly decreased in those women having PGS.[12] This led to an intense and sometimes acrimonious debate about the technical aspects of the trial both in the United States and Europe, polarized by the divergent prejudices of those in the private sector versus those in countries where IVF treatment is provided by the health system.

Nevertheless, a consensus emerged that screening for a limited number of chromosomes by FISH needed to be replaced by a technology that allows the detection of aneuploidy for all 24 different chromosomes. Furthermore, the problem of chromosomal mosaicism at cleavage stages has reopened the debate about the optimal stage to biopsy the embryo. Should the by-products of female meiosis, the first and second polar bodies, be analyzed to focus on the major source of aneuploidy, or is it more effective to biopsy a small number of trophectoderm cells at the blastocyst stage to avoid chromosomal mosaicism and include detection of any paternal aneuploidies? Treff compares the methods available for 24-chromosome screening and critically evaluates how accurate they are at the single-cell level. Fragouli and Wells then review the data emerging from the use of the new technologies, mainly comparative genomic hybridization either to metaphase chromosomes or more recently using microarrays (array CGH), and what the results tell us about how aneuploidies arise in early human development. Finally, Vanneste et al review their discovery that not only is the cleavage-stage embryo prone to chromosome instability resulting in mosaicism, but there is also a high incidence of de novo subchromosomal structural abnormalities.

In view of all of these exciting technical developments, which now provide a highly accurate genetic diagnosis of both chromosomal aneuploidy and single-gene defects, clinical pregnancy rates remain relatively low, and the incidence of multiple pregnancies and prematurity are still too high.[2] One of the reasons for this has always been that couples often only consider PGD late in their reproductive history when IVF is much less effective as a result of an increased incidence of maternal aneuploidy and the reduced number and quality of oocytes. Tur-Kaspa reviews the clinical management of PGD patients with different genetic problems to optimize a successful clinical outcome.

One approach to improving pregnancy rates per embryo transfer would to combine aneuploidy screening with the genetic diagnosis to avoid the transfer of nonviable aneuploid embryos. Genotyping hundreds of thousands of SNP markers by microarray and, for example, karyomapping each chromosome to identify the parental origin, provides a universal linkage-based test for single-gene defects that also detects meiotic trisomy and monosomy.[13] Also, with the costs of next-generation screening rapidly coming down, in the future, it may be increasingly common for couples to elect to have preconception screening for hundreds of severe recessive childhood diseases before starting a family.[14] For couples in their 20s or early 30s, prospects of a healthy live birth from even a single cycle of PGD would be much enhanced.

One of the disappointments of the last 20 years has been that PGD, possibly because of its early reputation as an expensive and error-prone technology, has failed to become widely practiced and still only accounts for a fraction of prenatal testing.[15] As and when it becomes possible to screen the population for common inherited diseases such as cystic fibrosis, providing IVF with PGD would be a highly cost-effective way to prevent the birth of affected children, particularly when compared with the cost of treating and caring for them throughout their lifetimes.[16] In the next 20 years, it may become possible to diagnose some genetic conditions noninvasively early in pregnancy using next-generation sequencing to analyze cell-free fetal DNA in the maternal circulation, if costs can be reduced.[17] Together these technologies should provide real alternatives to invasive prenatal diagnostic procedures at later stages of pregnancy to benefit couples at risk of having children affected by an inherited disease.

• References

• 1 Handyside AH, Kontogianni EH, Hardy K, Winston RM. Pregnancies from biopsied human preimplantation embryos sexed by Y-specific DNA amplification. Nature 1990; 344 (6268) 768-770
• 2 Harper JC, Coonen E, De Rycke M , et al. ESHRE PGD Consortium data collection X: cycles from January to December 2007 with pregnancy follow-up to October 2008. Hum Reprod 2010; 25 (11) 2685-2707
• 3 Handyside A. Let parents decide. Nature 2010; 464 (7291) 978-979
• 4 Verlinsky Y, Rechitsky S, Evsikov S , et al. Preconception and preimplantation diagnosis for cystic fibrosis. Prenat Diagn 1992; 12 (2) 103-110
• 5 Hardy K, Martin KL, Leese HJ, Winston RM, Handyside AH. Human preimplantation development in vitro is not adversely affected by biopsy at the 8-cell stage. Hum Reprod 1990; 5 (6) 708-714
• 6 Handyside AH, Lesko JG, Tarín JJ, Winston RM, Hughes MR. Birth of a normal girl after in vitro fertilization and preimplantation diagnostic testing for cystic fibrosis. [see comments]. N Engl J Med 1992; 327 (13) 905-909
• 7 Griffin DK, Handyside AH, Penketh RJ, Winston RM, Delhanty JD. Fluorescent in-situ hybridization to interphase nuclei of human preimplantation embryos with X and Y chromosome specific probes. Hum Reprod 1991; 6 (1) 101-105
• 8 Munné S, Lee A, Rosenwaks Z, Grifo J, Cohen J. Diagnosis of major chromosome aneuploidies in human preimplantation embryos. Hum Reprod 1993; 8 (12) 2185-2191
• 9 Munné S, Alikani M, Tomkin G, Grifo J, Cohen J. Embryo morphology, developmental rates, and maternal age are correlated with chromosome abnormalities. Fertil Steril 1995; 64 (2) 382-391
• 10 Munné S, Weier HU, Grifo J, Cohen J. Chromosome mosaicism in human embryos. Biol Reprod 1994; 51 (3) 373-379
• 11 Delhanty JD, Griffin DK, Handyside AH , et al. Detection of aneuploidy and chromosomal mosaicism in human embryos during preimplantation sex determination by fluorescent in situ hybridisation (FISH). Hum Mol Genet 1993; 2 (8) 1183-1185
• 12 Mastenbroek S, Twisk M, van Echten-Arends J , et al. In vitro fertilization with preimplantation genetic screening. N Engl J Med 2007; 357 (1) 9-17
• 13 Handyside AH, Harton GL, Mariani B , et al. Karyomapping: a universal method for genome wide analysis of genetic disease based on mapping crossovers between parental haplotypes. J Med Genet 2010; 47 (10) 651-658
• 14 Bell CJ, Dinwiddie DL, Miller NA , et al. Carrier testing for severe childhood recessive diseases by next-generation sequencing. Sci Transl Med 2011; 3 (65) 66ra4
• 15 Handyside AH. Preimplantation genetic diagnosis after 20 years. Reprod Biomed Online 2010; 21 (3) 280-282
• 16 Tur-Kaspa I, Aljadeff G, Rechitsky S, Grotjan HE, Verlinsky Y. PGD for all cystic fibrosis carrier couples: novel strategy for preventive medicine and cost analysis. Reprod Biomed Online 2010; 21 (2) 186-195
• 17 Lo YM, Chan KC, Sun H , et al. Maternal plasma DNA sequencing reveals the genome-wide genetic and mutational profile of the fetus. Sci Transl Med 2010; 2 (61) 61ra91