You are here

Preimplantation genetic diagnosis to improve pregnancy outcomes in subfertility

Best Practice & Research Clinical Obstetrics & Gynaecology, 6, 26, pages 805 - 815

Pre-implantation genetic diagnosis provides prenatal genetic diagnosis before implantation, thus allowing detection of chromosomal abnormalities and their exclusion from embryo transfer in assisted reproductive technologies. Polar body, blastomere or trophectoderm can each be used to obtain requisite genetic or embryonic DNA. Pre-implantation genetic diagnosis for excluding unbalanced translocations is well accepted, and pre-implantation genetic diagnosis aneuploidy testing to avoid repeated pregnancy losses in couples having recurrent aneuploidy is efficacious in reducing miscarriages. Controversy remains about whether pre-implantation genetic diagnosis aneuploidy testing improves take home pregnancy rates, for which reason adherence to specific indications is recommended while the issue is being adjudicated. Current recommendations are for obligatory 24 chromosome testing, most readily using array comparative genome hybridisation.

Keywords: array CGH, blastocyst, embryo, infertility, PGD safety, polar body, pregnancy rates, preimplantation aneuploidy testing, preimplantation genetic screening (PGS), repeated pregnancy losses, subfertility.

Preimplantation genetic diagnosis: indications and practicalities

Now over 20 years old, pre-implantation genetic diagnosis (PGD) is well established as an integral component of the prenatal genetic diagnosis armamentarium, a prime example of the translational fruits of the genome era. Complexities and safeguards, however, are necessary to assure accuracy. Although costs are higher than with traditional prenatal genetic diagnosis, PGD is sometimes the only practical way to achieve a goal. This applies, in particular, to improving pregnancy rates in assisted reproductive technologies (ART). Most pregnancy loses are caused by chromosomal abnormalities; thus, detecting and excluding these disorders by PGD becomes applicable to couples of known infertility or subfertility. The latter may be experiencing unrecognised embryonic losses, and be labelled clinically as infertile.

In this chapter, I first review approaches and pitfalls for obtaining gamete or embryonic DNA necessary for PGD. I then consider well-accepted and still arguable indications for applying PGD for subfertility, their diagnostic accuracy, and recent trends in diagnostic approaches for aneuploidy detection.

Obtaining cells for pre-implantation genetic diagnosis

Pre-implantation genetic diagnosis requires access to DNA from gametes or embryos before 6 days of conception, when implantation occurs. The three approaches are (1) polar-body biopsy assessing female gametes; (2) blastomere biopsy (aspiration), assessing the 3-day, six-to eight-cell cleaving embryo; and (3) trophectoderm biopsy, assessing the 5- to 6-day blastocyst.

Blastomere biopsy in the cleavage stage embryo (six to eight cells)

In blastomere biopsy, the glycoprotein layer surrounding the embryo, zona pellucida is traversed by mechanical, laser, or chemical means in order to extract a cell(s). The first two methods are most commonly used. Almost all centres remove only one cell because even one cell less is believed to reduce embryo survival as manifested by a 10% reduction in pregnancy rate; removal of two cells reduces the pregnancy rate considerably more. 1 These figures are derived from data correlating pregnancy rates with numbers of blastomeres remaining after the thaw of cryopreserved embryos. Extrapolation of these pregnancy rates to those of biopsied embryos not subjected to cryopreservation may or may not be completely applicable. The 40–50% reduction in pregnancy rates associated with loss of two blastomeres, however, 1 cast aspersions on protocols involving removal of two blastomeres. An experienced centre in Brussels, 2 which once routinely removed two cells, recently reached similar conclusions; their live birth rates were 37.4% and 22.4% after removal of one versus two cells, respectively. 2 The ‘take home’ baby rates are comparable with non-biopsied embryos. A total of 2005 cycles were monitored in the European Society of Human Reproduction and Embryology (ESHRE) PGD Consortium data collection; delivery rates were 16% per oocyte retrieval and 22% per embryo transfer; clinical pregnancy rates were 19 and 26%, respectively.3 and 4

Polar-body biopsy

Oocyte genotype can be deduced readily by analysis of the first and second polar biopsy. 5 The underlying principle is that the first polar body from a heterozygous individual showing a mutant maternal allele should be complemented by a primary oocyte having the normal allele. Oocytes deduced to be genetically normal can be allowed to fertilisein vitroand be transferred for potential implantation. Conversely, a normal polar body indicates an abnormal oocyte; thus, fertilisation would not proceed. If the assessment is for aneuploidy, a polar body with 24 or 22 chromosomes would indicate the oocyte scheduled for fertilisation would have 22 or 24 chromosomes, respectively; thus, the embryo would be monosomic or trisomic and not suitable for fertilisation. The same principle would apply with analysis of the second polar body.

Polar-body biopsy has advantages and disadvantages. One advantage is that first polar body is present before fertilisation; thus, its analysis offers the unique possibility of preconceptional diagnosis. The second polar body is not extruded until the mature oocyte is exposed to sperm. This becomes the only useful option if one must limit the number of oocytes that can be fertilised or embryos transferred; biopsy of the first polar body allows, in the absence of recombination, identification of a normal oocyte. Even without examining the second polar body, however, most euploid oocytes may be fertilised and reasonable pregnancy rates maintained, despite restrictive legislation. 6 Another advantage of polar-body biopsy is that no reduction occurs in cell number. The obvious disadvantage is inability to assess paternal genotype, obviously precluding application if the father has an autosomal dominant disorder and making analysis less efficient in managing couples at risk for autosomal traits. On the other hand, 95% of aneuploidies arise in maternal meiosis, so there is little loss in efficiency in PGD aneuploidy testing.

If one is carrying out PGD for detecting single nucleotide polymorphisms (SNPs) on mutant alleles causing single-gene disorders, one must take into account recombination, a phenomenon that obligatorily occurs between homologous chromosomes. If recombination of chromatids were not to occur, the two chromatids of a single chromosome in the first polar body would be identical in genotype and complementary to the oocyte; the second polar body (chromatid) would be identical to the oocyte. If crossing over were, however, to involve the region containing the gene in question, the single chromosome in the first polar body would show different alleles on its two chromatids (heterozygosity). Genotype of the oocyte could thus not be predicted without either biopsy of the second polar body or biopsy of the embryoper se(blastomere). In practice, both first and second polar bodies are biopsied in almost all centres.

Polar-body biopsy has recently gained popularity for an unexpected reason: greater validity in correlating results of a single-cell analysis. It is paradoxical but true that chromosomal status of the oocyte, only deduced from its complementary first and second polar bodies, is a more reliable indicator of embryo status than a blastomere from the embryo itself. Sequential testing of blastomere or blastocyst would still be necessary to exclude the rare 5% of paternally derived aneuploidies.7 and 8That analysis of a blastomere is less accurate than polar-body biopsy reflects the occurrence in blastomeres of mitotic non-disjunctional events that are unrepresentative of the status of the embryo's remaining cells. A randomised-controlled trial (RCT) planned by ESHRE to assess whether PGD aneuploidy testing increases ART success rates will use polar-body biopsy. 9

Centres technically experienced in polar-body biopsy seem to have pregnancy rates comparable to those achieved using blastomere biopsy. No attempts have been made to compare relative safety of blastomere versus polar-body biopsy, reflecting, until recently, the reality that few centres carried out polar-body biopsy. In fact, RCTs could be potentially misleading if technical expertise in a given centre is not comparable with both techniques.

Blastocyst biopsy

Biopsy of the trophectoderm in the 5- to 6 day, 120-cell blastocyst is the third approach. More cells can be removed at this stage, potentially facilitating diagnosis. Because trophectoderm forms the placenta, embryonic cellsper secould theoretically not be removed. In fact, long before the development of PGD, 10 human blastocysts were recovered by uterine lavage. Lavage to recover blastocysts was envisioned to be the approach by which cells removed by trophectoderm biopsy could allow genetic diagnosis. 11 After the development of the polymerase chain reaction, lavage for PGD was no longer pursued. Blastocyst biopsy received little attention, as ART centres routinely generated salutary pregnancy rates by transferring 3- to 4-day morulae.

Practitioners of ART have now returned to the blastocyst as a preferred stage of transfer. The additional 2–3 days in culture, beyond that required for an eight-cell embryo, allows some self-selection against non-thriving and aneuploid embryos. About one-third of embryos with chromosomal abnormalities are selected against between 3 and 5. Pre-implantation genetic diagnosis is still necessary to exclude remaining aneuploidies, whose frequency depends upon maternal age. Australian12 and 13and other 14 investigators have demonstrated the feasibility of blastocyst biopsy for PGD. One problem is increased rate of monozygotic twinning and concern that some morphologically good 4–5 days embryos may not thrive on in-vitro media constructed for use up to 3 days. Another potential problem is that culture media suitable for day 1-3 embryos may not always be optimal for day 4–5 embryos.

Embryo biopsy and cryopreservation

Embryo survival after polar body or blastomere biopsy was once considered almost impossible. It was thus believed that biopsied embryos must be transferred by day 6, with all required diagnostic results available by then. Cryopreservation of biopsied embryos is now feasible.15, 16, and 17As a result, one promising approach can be envisioned: blastocyst biopsy, vitrification, unhurried diagnostic analysis, thawing 1 month or more later and transfer in synchrony with a natural cycle. The intervening month allows diagnostic application, like array comparative genome hybridisation (CGH) or multiplex genome analysis using polymorphisms (SNPs) or copy number variants (CNVs).14, 18, 19, and 20Although all these can be carried out in ‘real time’, allowing fresh embryo transfer, this is a technical tour de force and now unnecessary.

Diagnostic methods for selecting chromosomal abnormalities

Because a karyotype cannot be obtained reliably from a single cell, chromosome-specific fluorescent in-situ hybridisation (FISH) probes have traditionally been used. Pivotal to obtaining accurate results is an intact nucleus. A high percentage of informative cases can be achieved only if removal of a blastomere or polar body can be accomplished in 3–5 mins, preferably less. Otherwise, the embryo (and blastomere or polar body) may be damaged as result of desiccation, temperature changes or alterations in osmolarity. The high error rates and high numbers of non-informative embryos in certain RCTs may reflect embryo damage.21, 22, and 23

Diagnostic sensitivity with chromosome-specific probes logically should increase as the number of chromosomes tested increases. About 70–80% of aneuploidy embryos can be detected using eight to12 chromosomes (e.g. X, Y, 13, 15, 16, 17, 18, 21, 22). These chromosomes must be interrogated over two to three hybridisation cycles, given that no more than five probes per hybridisation are recommended. Munné et al. 24 observed that actually 90% of aneuploidy can be detected by interrogating only 10 to 12 chromosomes. The reason is that rarer trisomies (double trisomy) often accompany more traditional aneuploidies (e.g. numbers 16 or 21). Thus, screening with only 10 chromosome-specific probes identifies 89% (382 out of 427) of embryos that are either normal or abnormal; screening with 12 probes identifies 91% (389 out of 427).

Notwithstanding excellent results with chromosome-specific FISH probes over the past 10–15 years, the current standard is to obtain information on all 24 chromosomes. In fact, hybridisation can be carried out to cover all chromosomes. For example, using single blood cells 25 enumerated all 24 chromosomes on a single slide, carrying out six successive FISH cycles each with four chromosomes. Griffin et al. 26 developed a 24-colour FISH assay, using four rounds of hybridisation completed within 24 h. The preferred approach, however, is to apply genome-wide molecular approaches (cytogenomics). One approach is to use SNPs, using several informative per chromosome. Given DNA from male and female partners and other family member(s), one can deduce the number of chromosomes in a single cell on the basis of number of SNPs (alleles) on a given chromosome. For example, one normally expects two different SNPs, or a 1:1 allele ratio. If there are three different SNPs or an altered ratio (2:1), trisomy can be deduced. Single-cell aneuploidy testing using genome-wide SNPs can be accomplished reliably on a single blastomere.18, 19, and 20Results can be provided quickly enough for normal embryos to be transferred in the same cycle. Using SNP analysis, 27 accomplished this in 4 h.

Comparative genomic hybridisation (chromosomal microarrays)

Comparative genomic hybridisation is an exciting molecular cytogenetic technique that allows comprehensive analysis of the entire genome, based on single-stranded DNA annealing (hybridising) with a complementary single-stranded DNA. Hybridisation occurs whether denatured DNA is from the same individual or from different individuals. Typically, normal (control) DNA is labelled with a fluorochrome of one colour (e.g. green), whereas test (patient) DNA is labelled with a fluorochrome of a different colour (e.g. red). Given that both test and control DNA are denatured (single stranded), hybridisation occurs. If equal amounts of control and test DNA are present, the colour of the hybridised mixture should be yellow. If the test DNA is in excess (e.g. trisomy), the entire region for that chromosome would reveal more of the colour used to connote test (patient) DNA. This would be red in the previous example.

In practice, small amounts of single-stranded DNA of known sequence are placed by photolithography onto a platform (array) in ordered fashion. The amount of DNA in each ‘spot’ is small (i.e. micro). The number of sequences is set in advance, but is expected to encompass the entire genome, one sequence overlaying the adjacent one (‘tiling'). The ‘control’ DNA embedded by photolithography is labelled with a fluorochrome of one colour, and is exposed to single-stranded test DNA (e.g. patient), now labelled with a fluorochrome of different colour as reasoned above. Again, if control and test DNA are quantitatively equal for a given sequence, the colour is yellow. If test DNA is in excess (trisomy), one colour predominates; if test DNA is deficient, the other colour predominates.

Various commercial platforms are available, all interrogating sequences of DNA along every chromosome. ‘Coverage’ varies slightly based on sensitivity sought, but in all there is redundancy. That is, a given region is interrogated more than once to ensure replicability before making a diagnosis. Arrays used to interrogate blood from liveborn infants or chorionic villi or fluid cells are designed to be more sensitive than those used for PGD. The latter are designed to detect aneuploidies reliably only from all chromosomes.

Wells et al. 28 used array CGH (Illumina platform) and found the probability of an individual interrogated by array CGH embryo generating a pregnancy was 66.7% compared with 27.9% without PGD. Sher et al. 29 reported a remarkable increase in birth rates transferring blastocysts which, as 3-day embryos, underwent biopsy and CGH followed by vitrification and then thawing for transfer at a later time: 48% birth rate (45 out of 94) per transferred CGH-tested blastocyst compared with only 15% (57 out of 382) in non-CGH tested blastocysts. In 2012, array CGH is the preferred diagnostic approach to assess 24 chromosome aneuploidy. Of note, the platform detects both whole chromosome and chromatid errors, the latter of which now seems to be the more common cytologic explanation for aneuploidy.30 and 31

Pre-implantation genetic diagnosis detection of chromosomal rearrangements

Chromosomal rearrangements (translocations or inversions) may result in unbalanced gametes and, hence, an unbalanced zygote. Many couples with rearrangements are diagnosed only after repeated spontaneous miscarriages, reflecting lethality conferred by unbalanced gametes. Other couples probably never even recognise they were pregnant. They may be labelled as ‘infertile’ or ‘subfertile’. Given the unavoidable high proportion of unbalanced embryos in translocation heterozygotes, reproductive efficiency should logically be improved by transferring only cytogenetically normal embryos. 32 Such transfers would preclude another pregnancy loss and also exclude abnormal liveborns. Given that relatively few embryos are either normal or balanced, success requires many embryos from which one can choose the few that are normal.

Use of commercially available (less expensive) chromosome-specific probes does not allow an embryo with a balanced translocation to be distinguished from one that is genetically normal. Breakpoint-specific probes could accomplish this, and were indeed used in the early years of PGD translocation analysis 33 ; however, costs are prohibitive. An alternative that can distinguish genetically normal embryos from those having a balanced translocation is ‘conversion’. The second polar body or biopsied blastomere can be fused to a diploid mouse cell to generate metaphase chromosomes amenable to analysis using chromosome-specific FISH probes. 34 Other methods for conversion include subjecting blastomeres to caffeine or electrical stimulation.35, 36, 37, and 38These approaches were not initially efficient, but recent experience has been more promising. Single nucleotide polymorphisms analysis can also be applied using the principles described above. 39

Ideally, RCTs would have demonstrated efficacy for PGD by showing increased pregnancy rates in couples having a balanced translocation. No RCTs have been conducted, but data are nonetheless compelling. With the use of PGD, the frequency of miscarriages seems no greater than the 30–40% in women lacking translocations, and probably considerably less. Otani et al. 40 observed only 5.3% losses after using PGD in couples with a translocation, far fewer than expected. The lifetime cumulative pregnancy rate using PGD was 57.6%, requiring an average of only 1.24 cycles. Translocation couples undergoing PGD became pregnant much more rapidly than those translocation couples who do not use PGD (mean 4–6 years).41, 42, and 43It is for this reason that the Society for Assisted Reproductive Technology guidelines 44 support this indication for PGD.

Pre-implantation genetic diagnosis aneuploidy testing for repeated pregnancy loss (recurrent aneuploidy)

At least 50% of first-trimester spontaneous miscarriages have numerical chromosomal abnormalities (aneuploidy). Non-random distribution occurs among successive miscarriages; abortuses are usually either successively aneuploid or successfully not (euploid). 45 In pre-implantation embryos, non-random aneuploidy distribution also occurs in successive cycles. 46 Of further relevance is that at least 50% of morphologically normal embryos in women over the age of 35 years are chromosomally abnormal. 47 Thus, selecting an embryo for transfer cannot be based solely upon morphology. Deferral of transfer until 5 days (blastocyst) selects against only one-third of embryos that are chromosomally abnormal at 3 days.

Given all the above, the rationale for carrying out PGD aneuploidy testing and transferring only euploid embryos would seem obvious. Because the rationale is mostly applicable for couples experiencing recurrent aneuploidy, ideally at least one loss should have documented aneuploidy. If no abortus information is available, one can still carry out FISH or array CGH on archived specimens embedded in paraffin. If not possible, one should acknowledge pitfalls, specifically that one-half of pregnancy losses will not have been caused by aneuploidy; thus, PGD aneuploidy testing is less likely to be beneficial.

Although RCTs have not been conducted to determine efficacy of PGD aneuploidy testing for this indication, PGD is accepted as beneficial.48, 49, 50, 51, and 52Preferable to strictly descriptive studies is the use of a surrogate to which comparison can be made. The Brigham formula 53 takes into account maternal age and the number of previous abortions, deriving the likelihood of a pregnancy loss. With this assessment tool, 50 observed pregnancy losses took place in only 13% of couples who used PGD, compared with the much higher expected rate of 33% in couples not undergoing PGD. Benefit was predictably greatest among women over age 35 years (P < 0.001).

Pre-implantation genetic diagnosis aneuploidy testing to improve pregnancy rates in assisted reproductive technologies

Controversy exists whether PGD aneuploidy testing in Europe (often termed preimplantation genetic screening or PGS) improves pregnancy rates in women who require ART for reasons other than for genetic indications. Analogous to the indication of recurrent pregnancy losses, rationale seems unassailable because the primary reason pregnancy rates decline in ART in the fourth decade is the high embryonic loss caused by aneuploid embryos. Aneuploidy increases with increasing age; thus, clinical miscarriage rates do as well. In both infertile and subfertile groups, the obvious strategy is to carry out PGD aneuploidy testing, transfer euploid embryos, and increase the proportion of viable pregnancies. Of course, the same strategy could be applicable to infertile couples who have not experienced a clinical pregnancy loss. Those with previous biochemical pregnancies (subfertile) would be especially good candidates. On the basis of descriptive studies comparing success rates before and after PGD, favourable results were in the late 1990s reported from experienced centres worldwide48, 49, 54, 55, 56, 57, and 58; The same results were evident when comparisons were made with historical expectations for age-matched women not undergoing PGD.

Two RCTs conducted in the USA59 and 60showed improved pregnancy rates, but neither was sufficiently powered to show statistical benefit. Larger PGD and ART centres in the USA and Europe offered PGD to improve pregnancy rates in older women; however, these large centres were not able to complete an RCT. Smaller centres have conducted RCTs, mostly in Europe. All failed to show benefit. Thus, guidelines have understandably not recommended PGD aneuploidy testing for this reason. Many studies have methodological flaws 23 and cogent objections can be raised. As an illustration, let us consider the well-publicised RCT of Mastenbroek et al. 61 Pre-implantation genetic diagnosis aneuploidy was diagnosed on the basis of seven chromosomes tested. Transfer of at least one euploid embryo resulted in a pregnancy rate of 16.8% per embryo. When no biopsy was attempted (true control), the pregnancy rate was 14.7% per embryo, or 13% lower than with PGD. In an additional 20% of cases, blastomere biopsy was carried out but no results were obtained; this rate was twice that expected in PGD aneuploidy testing.62 and 63This problem was exacerbated in this study by the low mean number of embryos present before PGD (4.8 per cycle). This mean was lower than the minimum number of six recommended at the time for proceeding with PGD.23 and 62Given lack of informative results, a third (unintended) group inadvertently evolved in which biopsy was carried out but no diagnosis resulted. The pregnancy rate was only 6% in this group. The investigators strictly adhered to intent-to-treat statistical analysis, which dictates that all cases remain assigned to their original group even if they do not complete their assigned technical protocol (i.e. PGD). With this method (in my opinion illogical when evaluating a surgical procedure like embryo biopsy), the true PGD group (16.8% pregnancy per embryo) was pooled with the de-facto sham group (6%) to yield a blended ‘PGD’ live birth rate of only 24% per cycle, compared with the statistically higher 35% in the non-biopsied, non-PGD group.

Despite methodological flaws in this and certain other PGD aneuploidy RCTs, it remains that no RCT has shown beneficial results in improving pregnancy rates. Thus, in 2010, the ESHRE PGD Consortium generated a position statement recommending that PGD aneuploidy testing (termed PGS) interrogate embryonic cells other than blastomeres and use methods other than FISH. 64 Polar-body biopsy using array CGH was recommended, and an RCT initiated. 9

When considering whether to carry out PGD aneuploidy testing, these guidelines might apply: (1) limit indications to women of relatively advanced maternal age, perhaps 37 years or older; (2) initiate PGD only if highly skilled embryologists are available; (3) proceed until completion of the cycle only if there are six to eight morphologically normal embryos from which two to three chromosomally normal embryos can thus be reasonably expected. If fewer embryos exist, PGD probably should not be pursued in that cycle; (4) interrogate all 24 chromosomes by array CGH.

Pre-implantation genetic diagnosis for single-gene abnormalities and potential causes of subfertility

Pre-implantation genetic diagnosis can be carried out for any single-gene disorder whose chromosomal location is known; even if the causative mutation is not known linkage analysis can still be carried out. 65 In the 2005 ESHRE PGD Consortium cycles,3 and 4the most common single-gene indications among 500 cycles were myotonic dystrophy (n = 76), Huntington disease (n = 56), cystic fibrosis (n = 55) fragile X syndrome (n = 51), spinal muscular atrophy (n = 27), tuberous sclerosis (n = 15) and Marfan syndrome (n = 13); beta-thalassemia and sickle-cell anaemia combined for 61. A total of 110 other conditions were interrogated, including 18 cases undergoing human leukocyte antigen typing and not at risk for a genetic condition. It can be predicted that pregnancy losses, and by extension sub-fertility, can be avoided or ameliorated by applications of the same technology. Thus, principles underpinning this approach will be discussed briefly.

Minimising diagnostic accuracy by linkage analysis

Pre-implantation genetic diagnosis for disorders caused by a single mutant gene is highly accurate in experienced hands, but several pitfalls must be avoided. One potential problem, contamination, is now largely obviated by obligatory use of ICSI. The major problem remains allele drop out (ADO). A genuine biologic explanation is double-stranded breakage involving DNA strands of one allele, thereby resulting in only the complementary allele able of being amplified. More commonly, both alleles actually are amplified, but one so much greater than the other that it appears ADO has occurred. This phenomenon might especially be prone to occur in poor quality embryos or nuclei. Irrespective, amplification rates rarely exceed 90–95% per allele even in experienced hands. 62 Less than 100% amplification could reflect stochastic phenomena (e.g. failure of probes to locate patient DNA), perhaps aggravated by embryo damage that has resulted in loss of embryonic DNA. Allele drop out can be managed.

Liebaers et al. 66 reported 0.6% mis-diagnosis in 581 PGD pregnancies, excluding one case in which linkage had been deduced incorrectly. 67 Linkage data are required to avoid the pitfall of ADO, obligatory in single-gene PGD. This approach confirms results of mutation analysisper seand allows transfer of embryos even if ADO has occurred at the locus being interrogated. It is necessary to establish the parental haplotype phase, and has traditionally involved determining polymorphic short tandem repeats or SNPs on either side of the mutant and normal allele (i.e. phase [cis]). Phase is typically determined by analysing affected and unaffected family members. The same approach could be taken in sub-fertile couples, even if all relatives are ostensibly ‘normal’.

Safety of pre-implantation genetic diagnosis

Removal of one or more blastomeres or polar bodies might logically be injurious, and we have noted that pregnancy rates are indeed decreased. The totipotential (pluripotent) nature of embryonic cells, however, confers theoretical safety against organ-specific differentiation resulting in liveborns with anomalies. Loss of one or more cells before organ system differentiation is readily obviated by another cell having capacity to accomplish that same purpose. Thus, the malformation rate should be similar to that in the general population. In fact, the birth defect rate in non-PGD ART is 30% higher than the background rate. 68 This seems not to be to caused by ARTper sebut by the underlying infertility that necessitated ART. 69

Liebaers et al. 66 have conducted the most thorough study of PGD offspring, carrying out a physical examination 2 months after births in their centre in Brussels. Anomalies in 563 PGD liveborns, 18 stillborns, and nine neonatal deaths were compared with these in a previously reported cohort study of ICSI offspring not undergoing PGD. 70 About one-half of the PGD cases followed by Liebaers et al. 66 were at risk for a single-gene disorder, whereas the others underwent aneuploidy testing. Structural malformations were found in 2.13% of those undergoing PGD alone and 3.38 % in those requiring ICSI. No differences were observed between offspring resulting from single-gene PGD and PGD aneuploidy testing. A smaller matched pair study (n = 102 in each arm) using additional assessments instruments also showed no statistical difference. 71

The anomaly rate observed in PGD cases observed by Liebaers et al. 66 was similar to that observed in the Chicago Reproductive Genetics Institute cohort: 1.9% of 1230 babies, 72 pooling all indications. Thus, cohorts from two experienced centres have failed to show increased anomalies or anomalies disproportionally clustered in any given organ system. Information into the effects of cryopreservation of biopsied embryos is still needed, but PGD is safe for liveborns. 73


Subfertility is not infrequently the result of unrecognized genetic abnormalities in one or both partners. This currently applies to chromosomal abnormalities but increasingly can be expected to extend to single gene causes. As readily available tests are able to detect more genetic explanations for subfertility, prediction will improve and ART efficiency likewise increase. Preimplantation genetic diagnosis is likely to become more widely utilized, not only array CGH for aneuploidy but whole genome analysis. Cryopreservation of biopsied blastocysts will permit analysis for multiple genes pivotal for development, with transfer in a later cycle only of ostensibly normal embryo(s).

Practice points


  • PGD aneuploidy testing reduces the number of miscarriages in couples presenting with repeated spontaneous abortions.
  • PGD aneuploidy testing for improving ART pregnancy rates is highly plausible, but RCTs have not yet proved beneficial statistically. Methodological problems, however, exist in reported trials.
  • Pre-implantation genetic diagnosis aneuploidy testing is best carried out on polar bodies and should use 24 chromosome array CGH, a much more complete test than the selected chromosome-specific FISH probes used in reported RCTs.
  • Embryo biopsy seems safe, with no increased rate of birth defects in resulting liveborns.
Research agenda


  • Conduct RCTs or best alternative to determine if application of 24 chromosome array CGH to PGD aneuploidy testing will increase the ‘take home baby’ rate in couples presenting with history of repetitive spontaneous miscarriages and no evidence of chromosomal translocation in either partner.
  • Conduct RCTs or best alternative to determine if application of 24 chromosome array CGH to PGD aneuploidy testing will increase the ‘take home baby’ rate in couples having no history of spontaneous abortion but presenting with infertility and requiring ART. Stratify by male (obligatory ICSI) and female infertility.
  • Conduct molecular studies (e.g. methylation of genes) to determine if cryopreservation after embryo biopsy (especially blastocyst) results in genomic alteration compared with embryos that are frozen but not previously biopsied, or not frozen.


  • 1 J. Cohen, D. Wells, S. Munne. Removal of 2 cells from cleavage stage embryos is likely to reduce the efficacy of chromosomal tests that are used to enhance implantation rates. Fertil Steril. 2007;87:496-503 Crossref
  • 2 A. De Vos, C. Staessen, M. De Rycke, et al. Impact of cleavage-stage embryo biopsy in view of PGD on human blastocyst implantation: a prospective cohort of single embryo transfers. Hum Reprod. 2009;24:2988-2996 Crossref
  • 3 V. Goossens, M. De Rycker, A. De Vos, et al. Diagnostic efficiency, embryonic development and clinical outcome after the biopsy of one or two blastomeres for preimplantation genetic diagnosis. Hum Reprod. 2008;23:481-492 Crossref
  • 4 V. Goossens, G. Harton, C. Moutou, et al. ESHRE PGD Consortium data collection VIII: cycles from January to December 2005 with pregnancy follow-up to October 2006. Hum Reprod. 2008;23:2629-2645 Crossref
  • 5 Y. Verlinsky, A. Kuliev. Practical preimplantation genetic diagnosis. (Panthenon Publishing Group, London, 2005)
  • 6 Gianaroli L, Magli M, Lappi M, et al. Preconception diagnosis. 9th Annual International Conference on Preimplantation Genetics 2009, Repro. Biomed. Online: Miami, Florida. p. S5.
  • 7 A. Kuliev, J. Cieslak, Y. Verlinsky. Frequency and distribution of chromosome abnormalities in human oocytes. Cytogenet Genome Res. 2005;111:193-198 Crossref
  • 8 A. Kuliev, Y. Verlinsky. Preimplantation genetic diagnosis: technological advances to improve accuracy and range of applications. Reprod Biomed Online. 2008;16:532-538 Crossref
  • *9 J. Geraedts, J. Collins, L. Gianroli, et al. What next for preimplantation genetic screening? A polar body approach!. Hum Reprod. 2010;25:575-577 Crossref
  • 10 J.E. Buster, M. Bustillo, I.A. Rodi, et al. Biologic and morphologic development of donated human ova recovered by nonsurgical uterine lavage. Am J Obstet Gynecol. 1985;153:211-217 Crossref
  • 11 S. Carson. Biopsy of blastocysts. Y. Verlinsky, A. Kuliev (Eds.) Preimplantation genetics (Plenum Press, New York, 1991) 85-88 Crossref
  • 12 S.J. McArthur, D. Leigh, J.T. Marshall, et al. Pregnancies and live births after trophectoderm biopsy and preimplantation genetic testing of human blastocysts. Fertil Steril. 2005;84:1628-1636 Crossref
  • 13 S.J. McArthur, D. Leigh, J.T. Marshall, et al. Blastocyst trophectoderm biopsy and preimplantation genetic diagnosis for familial monogenic disorders and chromosomal translocations. Prenat Diagn. 2008;28:434-442 Crossref
  • 14 W.B. Schoolcraft, E. Fragouli, J. Stevens, et al. Clinical application of comprehensive chromosomal screening at the blastocyst stage. Fertil Steril. 2010;94:1700-1706 Crossref
  • 15 W.T. Zheng, G.L. Zhuang, C.Q. Zhou, et al. Comparison of the survival of human biopsied embryos after cryopreservation with four different methods using non-transferable embryos. Hum Reprod. 2005;20:1615-1618 Crossref
  • 16 M. Kuwayama. Highly efficient vitrification for cryopreservation of human oocytes and embryos: the Cryotop method. Theriogenology. 2007;67:73-80 Crossref
  • 17 Y. Schlenker, J. Stevens, M. Rawlins, et al. Clinical success with vitrification following trophectoderm biopsy for comprehensive chromosomal screening. 65th Annual Meeting of the American Society for Reproductive Medicine, Atlanta, Georgia, October 17-21. Fertil Steril. 2009;92(Suppl.) O-243, S71
  • 18 A.H. Handyside, G.L. Harton, B. Mariani, et al. Karyomapping: a universal method for genome wide analysis of genetic disease based on mapping crossovers between parental haplotypes. J Med Genet. 2009;47:651-658
  • 19 Vanneste E, Voet T, Le Caignec C, et al. Chromosome instability is common in human cleavage stage embryos. Nat Med 15: p. 577–583.
  • 20 D.S. Johnson, G. Genelos, J. Baner, et al. Preclinical validation of a microarray method for full molecular karyotyping of blastomeres in a 24-h protocol. Hum Reprod. 2010;25:1066-1075 Crossref
  • 21 S. Munné, J. Cohen, J.L. Simpson. In vitro fertilization with preimplantation genetic screening. N Engl J Med. 2007;357:1769-1770 author reply 1770–1771
  • 22 S. Munné, L. Gianaroli, I. Tur-Kaspa, et al. Substandard application of preimplantation genetic screening may interfere with its clinical success. Fertil Steril. 2007;88:781-784
  • *23 J.L. Simpson. What next for preimplantation genetic screening? Randomized clinical trial in assessing PGS: necessary but not sufficient. Hum Reprod. 2008;23:2179-2181 Crossref
  • 24 S. Munne, E. Fragouli, P. Colls, et al. Improved detection of aneuploid blastocysts using a new 12-chromosome FISH test. Reprod Biomed Online. 2010;20:92-97 Crossref
  • 25 J. Aurich-Costa, C. Ng, S. Selvaggio, et al. Oligonucleotide (ODN) fluorescence in situ hybridization (Oligo-Fish) and conventional FISH allow enumeration of 24 chromosomes in 6 successive hybridizations performed in a single day., 65th Annual Meeting of the American Society for Reproductive Medicine: Atlanta, Georgia. October 17–21. Fertil Steril. 2009;92(Suppl.) O-170, S50
  • 26 D.K. Griffin, D. Ioannou, A.S. Ganriel, et al. Novel perspectives on 24 chromosome diagnosis in human preimplantation embryos. Chromosome Res. 2009;17:554
  • 27 N. Treff, X. Tao, J. Su, et al. Four hour 24 chromosome anueploidy screening using high throughput PCR SNP allele ratio analysis. 65th Annual Meeting of the American Society for Reproductive Medicine, Atlanta, Georgia, October 17-21. Fertil Steril. 2009;92(Suppl.) O-269, S49
  • 28 D. Wells, E. Fragouli, S. Alfarawaty, et al. Increased embryo implantation and high birth rates following comprehensive chromosomal screening of in vitro fertilized embryos. 9th Annual International Conference on Preimplantation Genetics, Miami, Florida, April 23–25. Reprod Biomed Online (Suppl). 2009;18:S10
  • 29 G. Sher, L. Keskintepe, M. Keskintepe, et al. Genetic analysis of human embryos by metaphase comparative genomic hybridization (mCGH) improves efficiency of IVF by increasing embryo implantation rate and reducing multiple pregnancies and spontaneous miscarriages. Fertil Steril. 2009;92:1886-1894 Crossref
  • *30 A. Kuliev, S. Rechitsky. Polar body based preimplantation genetic diagnosis for Mendelian disorders. Mol Hum Reprod. 2011;17:275-285 Crossref
  • *31 A. Kuliev, Z. Zlatopolsky, I. Kirillova, et al. Meiosis errors in over 20,000 oocytes studied in the practice of preimplantation aneuploidy testing. Reprod Biomed Online. 2011;22:2-8 Crossref
  • 32 S. Munne, M. Sandalinas, T. Escudero, et al. Outcome of preimplantation genetic diagnosis of translocation. Fertil Steril. 2000;73:1209-1218 Crossref
  • 33 S. Munne, J. Fung, C. Cassel, et al. Preimplantation genetic analysis of translocations: case-specific probes for interphase cell analysis. Hum Genet. 1998;102:663-674 Crossref
  • 34 Y. Verlinsky, D. Dozortsev, S. Evsikov. Visualisation and cytogenetic analysis of second polar body chromosomes following its fusion with a one-cell mouse embryo. J Assist Reprod Genet. 1994;11:123-131 Crossref
  • 35 Y. Verlinsky, S. Evsikov. A simplified and efficient method for obtaining metaphase chromosomes from individual human blastomeres. Fertil Steril. 1999;72:1127-1133 Crossref
  • 36 Y. Verlinsky, J. Cieslak, S. Evsikov, et al. Nuclear transfer for full karyotyping and preimplantation diagnosis for translocations. Reprod Biomed Online. 2002;5:300-305 Crossref
  • 37 A. Shkumatov, V. Kuznyetsov, J. Cieslak, et al. Obtaining metaphase spreads from single blastomeres for PGD of chromosomal rearrangements. Reprod Biomed Online. 2007;14:498-503 Crossref
  • 38 A. Kuliev, J. Ciesalk, J. Jansen, et al. Conversion and non-conversion approach to preimplantation diagnosis for chromosomal rearrangements in 475 cycles. Reprod Biomed Online. 2010;21:93-99 Crossref
  • 39 M. Traversa, D. Leigh. The application of a molecular strategy using STR for routine PGD in both reciprocal and Robertsonian translocation carriers. 9th International Conference on Preimplantation Genetics, Miami, Florida, April 23–25. Reprod Biomed Online (Suppl). 2009;18:S10
  • 40 T. Otani, M. Roche, M. Mizuike, et al. Preimplantation genetic diagnosis significantly improves the pregnancy outcome of translocation carriers with a history of recurrent miscarriage and unsuccessful pregnancies. Reprod Biomed Online. 2006;13:869-874 Crossref
  • 41 M. Goddijn, J.H.K. Joosten, A.C. Knegt, et al. Clinical relevance of diagnosing structural chromosome abnormalities in couples with repeated miscarriage. Hum Reprod. 2004;19:1013-1017 Crossref
  • 42 M. Sugiura-Ogasawara, Y. Ozaki, T. Sata, et al. Poor prognosis of recurrent aborters with either maternal or paternal reciprocal translocations. Fertil Steril. 2004;81:367-373 Crossref
  • 43 M.D. Stephenson, S. Sierra. Reproductive outcomes in recurrent pregnancy loss associated with a parental carrier of a structural chromosome rearrangement. Hum Reprod. 2006;21:1076-1082
  • 44 M.A. Fritz, G. Schattman. Reply of the Committee: Parental translocations and need for preimplatation genetic diagnosis? Disorting effects of ascertainment bias and the need for information-rich families. Fertil Steril. 2008;90:892-893 Crossref
  • 45 D. Warburton, L. Dallaire, M. Thangavelu, et al. Trisomy recurrence: a reconsideration based on North American data. Am J Hum Genet. 2004;75:376-385 Crossref
  • 46 C. Rubio, C. Simon, F. Vidal, et al. Chromosomal abnormalities and embryo development in recurrent miscarriage couples. Hum Reprod. 2003;18:182-188 Crossref
  • 47 S. Munne, M. Alikani, F. Tomkin, et al. Embryo morphology, developmental rates, and maternal age are correlated with chromosome abnormalities. Fertil Steril. 1995;64:382-391
  • 48 L. Gianaroli, M.C. Magli, A.P. Ferraretti, et al. Preimplantation diagnosis for aneuploidies in patients undergoing in vitro fertilization with a poor prognosis: identification of the categories for which it should be proposed. Fertil Steril. 1999;72:837-844 Crossref
  • 49 S. Munne, C. Magli, J. Cohen, et al. Positive outcome after preimplantation diagnosis of aneuploidy in human embryos. Hum Reprod. 1999;14:2191-2199 Crossref
  • 50 S. Munne, S. Chen, J. Fischer, et al. Preimplantation genetic diagnosis reduces pregnancy loss in women aged 35 years and older with a history of recurrent miscarriages. Fertil Steril. 2005;84:331-335 Crossref
  • 51 Y. Verlinsky, I. Tur-Kaspa, J. Cieslak, et al. Preimplantation testing for chromosomal disorders improves reproductive outcome of poor-prognosis patients. Reprod Biomed Online. 2005;11:219-225
  • *52 Y. Verlinksy, A. Kuliev. Atlas of preimplatation genetic diagnosis. 2nd ed. (Taylor and Francis, Boca Raton, Florida, 2005)
  • 53 S.A. Brigham, C. Conlon, R.G. Farquharson. A longitudinal study of pregnancy outcome following idiopathic recurrent miscarriage. Hum Reprod. 1999;14:2868-2871 Crossref
  • *54 L. Gianaroli, C. Magli, A.P. Ferraretti, et al. The beneficial effects of preimplantation genetic diagnosis for aneuploidy support extensive clinical application. Reprod Biomed Online. 2005;10:633-640 Crossref
  • 55 S. Munne, M. Sandalinas, T. Escudero, et al. Improved implantation after preimplantation genetic diagnosis of aneuploidy. Reprod Biomed Online. 2003;7:91-97 Crossref
  • 56 Y. Verlinksy, A. Kuliev. PGD and its role in ART. P. Brisdon (Ed.) Textbook of IVF and assisted reproduction (Parthenon Publishing Group, London, 2005) London
  • *57 Y. Verlinksy, A. Kuliev. Practical preimplantation genetic diagnosis. (Springer-Verlag, London, 2005)
  • 58 A. Kuliev, S. Rechitsky, O. Verlinsky, et al. Preimplantation testing for chromosomal disorders improves reproductive outcome of poor-prognosis patients. Reprod Biomed Online. 2005;11:219-225
  • 59 L. Werlin, I. Rodi, A. DeCherney, et al. Preimplantation genetic diagnosis as both a therapeutic and diagnostic tool in assisted reproductive technology. Fertil Steril. 2003;80:467-468 Crossref
  • 60 J.E. Mersereau, E. Pergament, X. Zhang, et al. Preimplantation genetic screening to improve in vitro fertilization pregnancy rates: a prospective randomized controlled trial. Fertil Steril. 2008;90:1287-1289 Crossref
  • *61 S. Mastenbroek, M. Twisk, J. van echten-Arends, et al. In vitro fertilization with preimplantation genetic screening. N Engl J Med. 2007;357:9-17 Crossref
  • *62 Preimplantation Genetic Diagnosis International Society (PGDIS). Guidelines for good practice in PGD: programme requirements and laboratory quality assurance. Reprod Biomed Online. 2008;16:134-147
  • 63 C. Gutierrez-Mateo, J. Sanchez Garcia, J. Fisher, et al. Preimplantation genetic diagnosis of single-gene disorders: experience with more than 200 cycles conducted by a reference laboratory in the United States. Fertil Steril. 2009;92:1544-1556 Crossref
  • 64 J. Harper, E. Coonen, F.C.S. Ramaekers, et al. What next for preimplantation genetic screening (PGS)? A position statement from the ESHRE PGD Consortium Steering Committee. Hum Reprod. 2010;25:821-823 Crossref
  • 65 Y. Verlinksy, A. Kuliev. Preimplantation genetic diagnosis. A. Milunsky, J.M. Milunsky (Eds.) Genetic disorders and the fetus (Blackwell Publishing, Sussex, 2010) 950-977
  • 66 I. Liebaers, S. Desntttere, W. Verpoest, et al. Report on a consecutive series of 581 children born after blastomere biopsy for preimplantation genetic diagnosis. Hum Reprod. 2010;25:275-282 Crossref
  • 67 L. Wilton, A. Thornhill, J. Traeger-Synodinos, et al. The causes of misdiagnosis and adverse outcomes in PGD. Hum Reprod. 2009;24:1221-1228 Crossref
  • 68 M. Hansen, C. Bower, E. Milne, et al. Assisted reproductive technologies and the risk of birth defects: a systematic review. Hum Reprod. 2005;20:328-338
  • 69 J.L. Zhu, O. Basso, C. Obel, et al. Infertility, infertility treatment, and congenital malformations: Danish national birth cohort. BMJ. 2006;333:679-683
  • 70 M. Bonduelle, I. Ponjaert, A.V. Steirteghem, et al. Developmental outcome at 2 years of age for children born after ICSI compared with children born after IVF. Hum Reprod. 2003;18:342-350 Crossref
  • 71 S. Desmyttere, M. Bonduelle, J. Nekkebroeck, et al. Growth and health outcome of 102 2-year-old children conceived after preimplantation genetic diagnosis or screening. Early Hum Dev. 2009;85:755-759 Crossref
  • 72 Ginsberg N, Rechitsky S, Kuliev A, et al. Clinical outcomes of over thousand deliveries after preimplantation genetic diagnosis (PGD) for genetic and chromosomal disorders. 9th Annual International Conference of Preimplantatiion Genetics 2009: Miami, Florida. April 23–25. Reprod Biomed Online 18: p. S35.
  • *73 J.L. Simpson. Children born after preimplantation genetic diagnosis show no increase in congenital anomalies. Hum Reprod. 2010;25:6-8


March of Dimes Foundation, White Plains, NY, USA

Tel.: +1 914 997 4555; Fax: +1 914 997 4560.