Performance comparison of two whole genome amplification techniques in frame of multifactor preimplantation genetic testing

Preimplantation genetic testing (PGT), formerly known as PGD for monogenic disease testing or PGS for chromosome screening, is an alternative to prenatal testing for couples being at risk of transmitting a genetic disorder to their offspring [38]. PGT allows exclusion of affected embryos before a clinical pregnancy has been established thus avoiding invasive prenatal testing and elective termination of pregnancy due to prenatally confirmed diagnosis. The material for PGT can be collected from day 3 or day 5 of developing embryo before its transfer to the uterus. The process initially requires controlled ovarian hyperstimulation, oocyte retrieval and subsequent oocyte in vitro fertilization (IVF), most commonly by intracytoplasmic sperm injection (ICSI) followed by embryo cultivation until the desired stage of development as well as a biopsy procedure [1]. Depending on a protocol, PGT can be done with or without embryo vitrification for the time of testing. Only embryos proved free of the disease-causing variant under consideration are subsequently transferred into the uterine cavity.

The success of the whole procedure depends mostly on competence and appropriate collaboration of the multidisciplinary team consisting of a clinical geneticist, reproductologist, gynaecologist, embryologist and molecular geneticist, and is achieved through safety and accuracy, improving genetic and reproductive medicine practices [6]. PGT is currently performed for single gene disorders (SGDs), late-onset disorders with genetic predisposition, chromosomal disorders, including aneuploidy and structural rearrangements, and HLA (human leukocyte antigen) typing to improve the access to HLA-matched stem cell transplantation [28].

The history of PGT goes back to 1989 when A. Handyside performed first preimplantation genetic diagnostic (PGD) cases detecting a Y chromosome-specific region with PCR in case of X-linked adrenoleukodystrophy and X-linked mental retardation [13]. Now, defining embryo gender is known as sexing and can complement to genetic testing of monogenic disorders linked to the sex chromosomes.

With time, PGT underwent significant methodological and approach changes, starting from polar body testing and blastomere analysis and now adapting trophectoderm biopsy with subsequent blastocyst freezing [26]. The analysis of more than a single cell leads to a more robust downstream molecular investigation, which sets among the reasons blastocyst stage biopsy strategy [5]. Molecular genetic testing developed from single loci directs PCR till sophisticated single cell whole genome amplification [9]. Embryo haplotyping offers a more generic approach to preimplantation diagnosis, and is especially useful for diseases with a wide spectrum of causative variants, such as cystic fibrosis and Duchenne muscular dystrophy [26].

Despite technological improvements, development of PGT protocols is challenging and prone to amplification failure, DNA contamination and ADO (allele dropout)—a phenomenon common to all single cell-based PCR tests, thus affecting the reliability of the test. ADO can be defined as amplification failure affecting only one of the parental alleles. ADO’s incidence varies, but in extreme cases has affected 20% of amplifications and in the past has led to several misdiagnoses [3]. The causes of misdiagnosis include swap of samples, transfer of the wrong embryo, maternal or paternal contamination, ADO, use of inappropriate probes or primers, probe or primer failure and chromosomal mosaicism [15].

ADO rates should be as low as possible, preferably less than 10%. Higher ADO rates can be tolerated when dealing with WGA-based protocols and autosomal recessive diseases compared to autosomal dominant or compound heterozygous cases. However, in such cases an increased number of linked markers have to be used [15].

Choosing the WGA type is also challenging due to difficulties in interpretation of downstream applications like short tandem repeat (STR) marker sizing with fluorescent polymerase chain reaction (fPCR) or array comparative genomic hybridization [25]. At the moment, several WGA technologies exist [40], for example, PCR-based approaches like degenerate oligonucleotide primer (DOP) [31] or primer extension (PEP) PCR technology [39]. Leading positions are taken by OmniPlex linear WGA [4, 33] technology developed by Rubicon Genomics and multiple displacement isothermal synthesis by a Phi29 polymerase approach (Alan H. [14]). Both of them have advantages and disadvantages. The use of Taq DNA polymerase in PCR-based approaches limits the fragment lengths to 3 kb. The Phi29 polymerase used for MDA generates DNA fragments up to 100 kb and has a 3′ → 5′ exonuclease proofreading activity. Often, it is not clear which technology could be prioritized in custom-designed protocols [40].

Better PGT results are now achieved through combining direct and indirect testing, using platforms like Karyomapping [12] for genome-wide linkage analysis or turning to next generation sequencing (NGS) protocols (Francesco [10]). However, current studies still highlight clinically important limitations in the reliability of the technologies; for example, using Karyomapping, ∼ 14% of embryos are expected to remain without a conclusive result [19]. NGS has potential power to increase throughput and evaluate multiple genetic loci in parallel, but it is also well known for sequencing artefacts, which may complicate its application to PGD [32]. Also, costs are still quite high especially for limited sample amounts.

Regardless of the fact that PGT is recognized for its benefits, it is still relatively unregulated and lacks standardization compared with other forms of diagnostic testing [15]. This is partially because PGD lies at the intersection of two technologies with a confusing regulatory status: assisted reproduction and genetic testing [8]. It is admitted that a robust PGT test should be able not only to distinguish between a normal and affected embryo, but also to highlight all the unexpected events that may happen during meiosis, fertilization or PGD experimental procedure, and thus it should detect recombination, monosomy or trisomy and therefore diagnose abnormal embryos and detect ADO and contamination [18]. In case of adverse misdiagnosis, lessons can be very painful to patients and staff [16, 36].

Despite numerous advances, assisted reproductive technology (ART) live birth rates are still low ranging from 27 to 55%, depending on the patient age group and methodology used [6]. Another step in reaching considerably good results for SGD-PGT is embryo aneuploidy exclusion since it is well known that preimplantation human embryos are prone to chromosome instability [34] and high aneuploidy rates [18, 35]. Early results show that combined PGD and PGS increase patient chance for a healthy childbirth [20, 30].

Taking into account the aforementioned information, the aim of our study was to develop an individualized effective and robust multifactor embryo testing protocol and show a performance comparison of two WGA techniques in four different downstream applications—STR sizing, Sanger sequencing, aCGH and SNaPshot technology. We present our PGT experience for single gene diseases of autosomal dominant (genes: ACTA2, HTT, KRT14), autosomal recessive (genes: ALOX12B, TPP1, GLB1) and X-linked (genes: MTM1, DMD) types of inheritance.

It is widely recognized that development of preimplantation genetic testing protocols is time consuming, costly and laborious, because of a wide spectrum of technical complications and biology-driven obstacles [3, 36]. It is essential to remember that the interpretation of results may influence not only particular family wealth but also in the long term even the well-being of the whole society, since PGT is a potential tool to cease out at least some genetic conditions.

Performance and outcome requirements for our approach were subjected to the following measures: possibility to combine several technologies in order to distinguish a normal embryo from a carrier and an affected one, to distinguish possible contamination and loci/allelic dropout events and to perform embryo chromosome screening. It was relevant to get a conclusion on all embryos subjected to biopsy and avoid any additional embryo manipulations like repeated thawing and rebiopsy. Such a wide spectrum of requirements was set carefully taking into consideration all the previous historical obstacles of PGT. We aimed first to meet the highest safety standards and secondly to prioritize a purpose of achieving the desired pregnancy in a personalized and customized manner saving patient time and expenses, which was shown through a comprehensive comparison of two different WGA techniques.

In general, obtaining micrograms of DNA through WGA of day-5 embryo biopsies allowed us to perform embryo haplotype analysis, aneuploidy screening by aCGH and direct mutation testing through SNaPshot, Sanger sequencing or fragment size analysis. As shown before, direct variant locus testing boldly complements the indirect one [18, 21], since crossover events cannot be completely ruled out and ignored [11] especially in the case of ADO, which was also true in our cohort. We designed hemi-nested primers for all assessed loci following PGDIS guidelines for good practice in PGD [16, 24]. We conclude that having as much as possible a number of semi-informative and/or informative linked markers within reasonable distance upstream and downstream from a gene is the best way to minimize the risk of misdiagnosis or no conclusive diagnosis for a particular embryo.

To our knowledge, our work is the first attempt in evaluating Picoplex and MDA amplifier performance across different downstream applications in frame of embryo preimplantation genetic testing. Provided figures give insight in understanding the applicability of both WGA methodologies to different molecular techniques and assist in choosing one when customizing PGT depending on the mutation type and technical equipment of the laboratory.

Currently, single-/few-cell WGA might be done with a wide array of amplification strategies [2]. We conclude that methodology choice depends on multiple factors like desired downstream application techniques as well as embryo amount. STR analysis efficacy including possible ADO event detection depends mostly on particular genomic region nucleotide composition and can be improved through PCR reaction condition optimization. The MDA WGA product compared to OmniPlex produces more heavy DNA strings, thus exhibiting properties closer to genomic DNA, and therefore, electropherograms are much clearer. Our results are consistent with other group findings that the per base error rate for MDA is at least two times lower compared to PCR-based approaches as shown for Sanger sequencing and SNaPshot applications. In general, MDA shows better genome recovery sensitivity as also concluded before [17] while allowing for a more convenient genotyping. However, MDA results in significant amplification bias [7], which contributed to the observed high aCGH noise levels. For full-fledged analysis, we recommend usage of both WGA techniques dividing the embryo cohort if the embryo amount is big enough. If the number of (semi-)informative markers is low, it is favourable to use the MDA technique since this will result in more robust SGD locus analysis. If STR marker informativenes is high enough, ADO will not drastically affect the result when detecting possible crossover events, and one might consider using OmniPlex since it gives more reliable aCGH profiles.

It is known that embryo aneuploidy and implantation potential are highly correlated with biopsy stage. Cleavage-stage embryo blastomere biopsy still represents the most commonly used method in Europe, although this approach has been shown to have a negative impact on embryo viability and implantation rates [5, 29]. Therefore, day 5 biopsy is highly favourable. In our study, trophectoderm biopsy performance was additionally complemented by usage of a time-lapse embryo imaging system, which not only aims at biopsy timing, but also can give a clue for the best choice of developing embryo for transfer through assessment of embryo rating by a time-lapse system algorithm when multiple embryos are SGD free and euploid.

Our experience with preimplantation testing began with a lot of goals and aims that were expected from a clinical and molecular point of view. We tried to set up a diagnostic algorithm that would suit every case and be foolproof. It became apparent already with our first cases that the approach should be more patient tailored than universal and more based on close communication between patients, clinical geneticists, reproductologists, embryologists and molecular geneticists than on pure data analysis. Proper genetic counselling before planning a PGT case is crucial as the patient has to be acquainted to any potential pitfalls to give a fully informed consent for testing. The final strategy of molecular testing should better be made after taking into consideration available embryo amount and morphology, type of disorder and family specifics and preferences. Although the main goal during monogenic disease preimplantation testing would always be disease-causative variant-free embryo selection, we found it expedient to use aneuploidy testing besides morphological embryo evaluation to determine the embryo most suitable for eSET thus increasing the chance for successful embryo implantation and development saving extra efforts and costs. The final result will always depend on a lot of different factors—even after all embryo testing is done, there is a possibility of failed implantation due to maternal age factor, endometrial receptivity problems and many more—this is why a multidisciplinary approach is a key to success for each family and thus the community altogether.

Single blastocyst biopsy whole genome amplification ensures possibility of multifactor preimplantation genetic testing without compromising embryo viability and in general chance of achieving a healthy pregnancy. A semi-nested direct and indirect testing system minimizes embryo misdiagnosis risk due to allelic dropout, nonspecific amplification or contamination. 24-Chromosome aneuploidy screening when performed concurrently with single gene disorder preimplantation embryo testing provides valuable information for embryo selection excluding leading failed embryo implantation cause and notably improving single embryo transfer rates thus saving time and money leading to higher pregnancy rates. A developed protocol can be further applied to customize PGT protocols for families seeking alternatives for prenatal testing.